Director of the National Bureau of Statistics answered a reporter’s question on the national economic operation in 2023.

(January 17, 2024)

 

 

  On January 17th, the State Council Press Office held a press conference. Kang Yi, director of the National Bureau of Statistics, and Wang Guanhua, spokesman of the National Bureau of Statistics, introduced the national economic operation in 2023 and answered questions from reporters. The transcript of the press conference is as follows:

 

  Shou Xiaoli, deputy director and spokesperson of the State Council Information Office.

 

  Good morning, ladies and gentlemen! Welcome to the press conference of the State Council Information Office. Today, we released economic data routinely. We invited Mr. Kang Yi, director of the National Bureau of Statistics, and Ms. Wang Guanhua, spokesperson of the National Bureau of Statistics, to introduce the national economy in 2023 and answer your concerns.

 

  Next, let’s first invite Mr. Kang Yi to make an introduction.

 

  Kang Yi, Director of National Bureau of Statistics.

 

  Thank you, host. Good morning, media friends. First of all, let me introduce the national economy in 2023.

 

  In 2023, the national economy rebounded and high-quality development was solidly promoted. In 2023, in the face of the complicated and severe international environment and arduous and arduous tasks of domestic reform, development and stability, under the strong leadership of the CPC Central Committee with the Supreme Leader as the core, all localities and departments conscientiously implemented the decision-making arrangements of the CPC Central Committee and the State Council, adhered to the general tone of striving for progress while maintaining stability, comprehensively implemented the new development concept, accelerated the construction of a new development pattern, comprehensively deepened reform and opening up, and intensified macro-control. Efforts were made to expand domestic demand, optimize the structure, boost confidence and prevent risks. China’s economy rebounded, supply demand improved steadily, transformation and upgrading were actively promoted, employment prices were generally stable, people’s livelihood security was strong and effective, and high-quality development was solidly promoted. The main expected goals were successfully achieved. (For details of the national economy in 2023, please refer to the link https.://www.stats.gov.cn/sj/zxfb/202401/t20240117_1946624.html)

 

  Next, I will give you a briefing on the improvement of the unemployment rate survey by age groups in cities and towns.

 

  The National Bureau of Statistics attaches great importance to the improvement of the labor force survey system, organizes relevant departments and experts to conduct in-depth discussions, study international standards and experiences and practices of various countries, and conducts on-the-spot investigations to carefully sort out the statistical methods and caliber of studying the unemployment rate. In order to reflect the employment and unemployment situation of youth more accurately and completely, from the point of view of fully considering the national conditions, the unemployment rate statistics of age groups have been adjusted in two aspects. One is to publish the unemployment rate of the labor force aged 16-24, which does not include students at school; The second is to increase the unemployment rate of the 25-29-year-old labor force excluding students at school. There are two main considerations:

 

  On the one hand, it is to monitor youth employment and unemployment more accurately. In 2023, on average, among the urban population aged 16-24 in China, students accounted for more than 60%, nearly 62 million; Non-school students account for more than 30%, about 34 million people. According to China’s national conditions, the main task of school students is to study, not to work part-time. If school students are included in the age group, young people who are looking for part-time jobs at school and those who are looking for jobs after graduation will be mixed together, which cannot accurately reflect the employment and unemployment situation of young people who really need to work in society. Calculating the unemployment rate of different age groups excluding students in school is conducive to more accurately reflecting the employment and unemployment situation of young people entering the society, giving them more accurate employment services and formulating more effective and targeted employment policies.

 

  On the other hand, it is to reflect the whole picture of employment and unemployment of young people in the process of graduation from school to stable work. The number of years of education of young people in China has been increasing. At present, the gross enrollment rate of higher education in China is nearly 60%. Most young people just graduated at the age of 24 and are still in the period of choosing jobs. Some of them are unemployed or unstable. By the age of 29, most of them have passed the period of choosing jobs, and the employment situation tends to be stable. All walks of life are very concerned about the employment situation of young people when they just leave school, and they are also very concerned about their employment situation for a period of time after graduation. Therefore, our bureau increased the calculation and released the unemployment rate of the 25-29-year-old labor force.

 

  Regarding the release method, in the future, our bureau will release the unemployment rate of the labor force aged 16-24, 25-29 and 30-59 excluding students in the data release database of the National Bureau of Statistics on a monthly basis. You can query the data in the data release database.

 

  I’ll stop here and answer your questions.

 

  Beijing Youth Daily reporter

 

  I want to ask about the general situation. In 2023, faced with multiple challenges such as weak world economic recovery, frequent domestic natural disasters and arduous tasks of reform, development and stability, what is the overall performance of China’s economic operation? From last year’s point of view, have our main objectives and tasks been better achieved throughout the year? Thank you.

 

 

  Kang Yi

 

  Thank you for your question. In the previous introduction, the annual performance of China’s economy was introduced in detail. On the whole, our main expected goals have been successfully achieved. It can be said that a transcript with good color and sufficient weight was handed over. For the economic performance of the past year, it can be summarized as a rebound, full of color, bright performance, and it is not easy.

 

  First, the recovery is good. From the perspective of economic growth, China’s GDP exceeded 126 trillion yuan in 2023, and the growth rate was 2.2 percentage points faster than that in 2022. Quarterly, it shows a trend of low, high and stable, and the positive trend is further consolidated. According to the comparable price calculation, the economic growth in 2023 will exceed 6 trillion yuan, which is equivalent to the total economic output of a medium-sized country in one year. The per capita GDP increased steadily, reaching 89,358 yuan in 2023, an increase of 5.4% over the previous year. From the perspective of employment, the employment situation has generally improved. The average unemployment rate in urban surveys has dropped by 0.4 percentage points over the previous year, especially for migrant workers. From the perspective of prices, prices generally maintained a moderate increase, with the annual CPI rising by 0.2% and the core CPI rising by 0.7%. From the perspective of international payments, the export of goods increased by 0.6% in the whole year, and the foreign exchange reserves at the end of the year exceeded 3.2 trillion US dollars.

 

  Second, the color is full. Last year, we coordinated the effective improvement of quality and the reasonable growth of quantity. The whole country closely focused on the primary task of high-quality development, and promoted the economy to continuously win advantages in structural adjustment and transformation and upgrading, so that high-quality development was more complete. The innovation-driven development strategy was implemented in depth, and the investment in innovation increased steadily. According to preliminary calculation, in 2023, the investment in research and experimental development of the whole society reached 3,327.82 billion yuan, and the intensity of R&D investment reached 2.64%, up by 0.08 percentage point over the previous year. In this year, some major scientific and technological innovations were reported frequently, especially the high-end, intelligent and green development of manufacturing industry was solidly promoted. With the optimization and upgrading of economic structure, the role of service industry and consumption as the main engines of economic growth has become more prominent. In 2023, the added value of service industry accounted for 54.6% of GDP, an increase of 1.2 percentage points over the previous year; The contribution rate of final consumption expenditure to economic growth reached 82.5%, an increase of 43.1 percentage points over the previous year. The foundation for safe development was further consolidated, and the annual grain output increased by 1.3% over the previous year; The output of raw coal, crude oil and natural gas in industrial enterprises above designated size increased by 2.9%, 2% and 5.8% respectively. We have kept the bottom line of no systemic risks, ensured economic and financial security, and ensured more effective people’s livelihood. The per capita disposable income of the national residents actually increased by 6.1% over the previous year.

 

  Third, the performance is bright. To observe the economic performance of China, we should not only compare it vertically with ourselves, but also compare it horizontally with other countries. In 2023, China’s economic growth rate of 5.2% is not only higher than the estimated global growth rate of about 3%, but also among the best in the world’s major economies. The contribution rate of China’s economy to world economic growth in 2023 is expected to exceed 30%, which is the biggest engine of world economic growth. At the same time, China’s exports have also achieved a small increase, and its share in the global market has remained stable. Another point is that the overall price increase is moderate, which is in sharp contrast with the high global inflation and the difficult balance between anti-inflation and steady growth in some countries.

 

  Fourth, it is not easy. It’s hard to know what’s going on until it’s gone through. In 2023, the world economy is in a downturn, the international pattern is complicated and evolving, geopolitical conflicts are frequent, and the complexity, severity and uncertainty of the external environment are rising. There are many periodic and structural contradictions in China, and natural disasters occur frequently. In this complicated situation, it is even more commendable to achieve such development achievements. These achievements are the result of the strong leadership of the CPC Central Committee with the Supreme Leader as the core, the scientific guidance of Socialism with Chinese characteristics Thought in the new era of the Supreme Leader, and the unremitting efforts of the people of the whole country. The achievements deserve full recognition and need to be cherished.

 

  Looking forward to 2024, we should also see that the current external environment is still complicated and severe, with insufficient domestic effective demand, overcapacity in some industries, weak social expectations and many hidden risks. To promote China’s economic recovery, we must overcome some difficulties and challenges. We should effectively deal with these difficulties and solve these problems in accordance with the decision-making arrangements of the Central Economic Work Conference, and constantly promote China’s economic stability and prosperity. Thank you.

 

  Global Times reporter

 

  I would like to ask the publisher to introduce the contribution rate and pulling point of the three major demands to economic growth in the fourth quarter of 2023 and the whole year. Are there any new changes? In addition, I would like to know what the development trend of domestic and international double circulation pattern was in the past year. Thank you.

 

 

  Kang Yi

 

  Thank you for your question. You are concerned about two aspects, one is the three major demand situations, and the other is the two-cycle situation. First of all, answer your first question, the contribution and pull of the three major needs. In 2023, final consumption expenditure, total capital formation and net exports of goods and services will drive economic growth by 4.3, 1.5 and -0.6 percentage points respectively, and their contribution rates to economic growth will be 82.5%, 28.9% and -11.4% respectively. In the fourth quarter, final consumption expenditure, total capital formation and net exports of goods and services boosted economic growth by 4.2, 1.2 and -0.2 percentage points respectively, and their contribution rates to economic growth were 80%, 23.1% and -3.1% respectively.

 

  The second question is the progress of domestic and international double circulation. It should be said that new progress has been made in building a new development pattern in 2023, which has three characteristics:

 

  First, the main role of the domestic macro-cycle is increasing. Based on the strategic basis of expanding domestic demand and releasing the market potential of strong domestic demand, the driving role of domestic circulation in economic development is obviously enhanced. In 2023, China’s total retail sales of consumer goods reached 47.1 trillion yuan, the scale of fixed assets investment reached 50.3 trillion yuan, and the contribution rate of domestic demand to economic growth reached 111.4%, an increase of 25.3 percentage points over the previous year. We are making great efforts to build a unified national market, get through the blocking points of the domestic big cycle, and smooth all links of production, distribution, circulation and consumption. The flow of factors tends to be active, and the connection between production and sales is gradually improving. In the fourth quarter, the utilization rate of industrial capacity was 75.9%, up 0.2 percentage points year-on-year; The product sales rate of industrial enterprises above designated size has remained stable at the level of over 97%. The annual commercial freight volume increased by 8.2% year-on-year, and the commercial passenger traffic increased by 66.5%.

 

  Second, the quality and level of international circulation have been further improved. Faced with the adverse effects of shrinking external demand, China has made efforts to promote the stable scale and excellent structure of foreign trade, and its exports have achieved positive growth. The horizontal ratio is better than that of major export-oriented economies, and the share of the international market remains generally stable. In 2023, the export volume of goods increased by 0.6% over the previous year, and the proportion of general trade and private enterprises’ import and export increased. At the same time, China is also actively expanding international economic and trade cooperation and building the "Belt and Road" with high quality. It has become a major trading partner of more than 140 countries and regions, and the scope of international circulation is broader and deeper. In 2023, China’s total import and export volume to countries building the "Belt and Road" increased by 2.8% over the previous year.

 

  Third, domestic and international circulation promote each other. The advantages of our super-large-scale market have emerged, and the expansion of domestic demand has boosted imports. In 2023, the import of agricultural products increased by 5%, consumer goods increased by 1.2% and energy products increased by 27.2%. The advantages of a sound industrial system and stable production capacity have also been brought into play. The promotion of stability and quality by exports has brought about the development of related domestic industries and helped the overall improvement of the economy. In 2023, China’s exports of goods reached a record high. Among them, the export of mechanical and electrical products increased by 2.9% over the previous year. The export value of the "new three products" represented by electric manned vehicles, solar cells and lithium-ion batteries also exceeded the trillion-dollar mark for the first time, with an increase of 29.9%.

 

  While we see the positive progress of domestic and international double circulation, we also see that there are still some blocking points in the domestic big circulation, the world economy continues to be in a downturn, the global industrial chain supply chain is deeply adjusted, and the international circulation is also facing some difficulties and challenges. In the next step, we should make overall plans to expand domestic demand and deepen supply-side structural reform, make overall plans for high-quality development and high-level security, make overall plans to promote deep-level reform and high-level opening-up, open up key blocking points that restrict the economic cycle, and promote the mutual promotion of domestic and international double cycles. Thank you.

 

  American international market news agency reporter

 

  According to recent data, the CPI in China showed negative growth for the third consecutive month in December. What is the prospect of CPI? Will the government take measures to deal with the low prices?

 

 

  Kang Yi

 

  Thank you for your question. The price issue is highly concerned by all sectors of society. From the overall situation in 2023, China’s prices generally maintained a moderate upward trend, and the annual CPI rose by 0.2%. The reporter mentioned that CPI has been negative for three consecutive months, and we should look at both the overall data and the structural data. The year-on-year decline of CPI in recent months is mainly structural and phased.

 

  First, the decline in CPI is structural. The recent decline in prices is mainly due to the fall in food and energy prices. If the influence of food and energy prices is deducted, the core CPI remains stable, which shows that the price decline is not universal and comprehensive, but local and structural. The decline in food and energy prices is not entirely a change in the relationship between supply and demand in the market. From the perspective of 2022 and 2023, it is mainly affected by some non-economic and unconventional factors. From the perspective of energy prices, the price of fuel now has a large weight in the basket of China’s CPI. In December 2023, the energy price decreased by 0.5% year-on-year, while it increased by 5.2% year-on-year. As we all know, energy prices were mainly affected by the Russian-Ukrainian war. In 2022, energy prices rose sharply, and in 2023, they fell back, with one positive and one negative, and the year-on-year downward pull was relatively large. From the perspective of food prices, it decreased by 3.7% year-on-year in December 2023 and increased by 4.8% in the same period of last year, that is, in December 2022. In December 2022, mainly affected by the epidemic situation, the logistics was not smooth, and the prices of various foods were rising. After the normal operation was resumed, the supply was sufficient, and the food prices naturally fell on the basis of the high base of the previous year.

 

  Second, the decline of CPI is phased. In particular, China’s economic recovery is improving, residents’ income is growing steadily, domestic total demand is expected to expand, and the price recovery of goods and services is based and conditional. The recent Spring Festival holiday is approaching, and the demand for food consumption is also increasing. People are more active in eating out, visiting relatives and friends, traveling and other services, which will boost the seasonal rebound of CPI. From the ring comparison, the CPI rose by 0.1% in December 2023; In the first ten days of January, 2024, according to the monitoring situation, some food prices maintained a steady and slightly rising trend. In addition, in addition to supply and demand, prices are also affected by expectations. Judging from the recent situation, the expected confidence of both enterprises and residents has recovered marginally. In December, the expected index of manufacturing production and business activities reached 55.9%, and the expected index of non-manufacturing business activities reached 60.3%, both of which were in a high boom zone. In the fourth quarter, the prosperity index of enterprises above designated size was 109 points, an increase of 0.4 points over the previous quarter. In December, the consumer confidence index rose by 0.6 points from the previous month.

 

  Third, the low price operation also reflects the problem of insufficient effective demand to some extent. Insufficient effective demand is a phased phenomenon in which our country’s economy is gradually moving towards normal state after three years of epidemic impact. In the short term, insufficient demand will lead to a downward trend in prices. The central government attaches great importance to the problem of insufficient demand. When planning economic work in 2024, the Central Economic Work Conference clearly emphasized that efforts should be made to expand domestic demand and promote consumption to shift from post-epidemic recovery to sustained expansion, and clearly pointed out a series of specific work directions. With the promulgation and implementation of these policies at the Central Economic Work Conference, the problem of insufficient effective demand will be gradually alleviated, and consumer prices are expected to stabilize and rebound. We expect that prices will rise moderately in 2024. Thank you.

 

  CCTV reporter from Central Radio and Television General Station

 

  High-quality development is the primary task of building a socialist modern country in an all-round way. What is the progress of high-quality economic development in China at present? What measures will be taken in the future to continuously promote high-quality economic development? Thank you.

 

 

  Kang Yi

 

  Thank you for your question. High-quality development is the last word in the new era and the primary task of building a socialist modern country in an all-round way. In 2023, all regions and departments have made great efforts to stabilize economic operation, and they have not slackened their efforts in promoting high-quality development. China’s economy has achieved effective improvement in quality and reasonable growth in quantity, and the road to high-quality development has become more firm and powerful. To sum up, it can be summarized by five "further".

 

  First, important progress has been made in the construction of a modern industrial system, and the transformation of development kinetic energy has been further accelerated. Modern industrial system is an important foundation for high-quality development. All localities and departments insist on scientific and technological innovation to lead the construction of modern industrial system, and solidly promote the high-end, intelligent and green transformation of manufacturing industry. The new kinetic energy and advantages of China’s economic development have been growing, and new progress has been made in the construction of modern industrial system.

 

  Second, the reform and opening up have been deepened and the vitality of development has been further released. In 2023, China made every effort to promote the construction of a unified national market, optimize the development environment of the private economy, and continue to create a market-oriented, rule-of-law and international first-class business environment, effectively enhancing the dynamic vitality of economic development. By the end of September 2023, there were 181 million registered business entities in China, including 122 million individual industrial and commercial households. Accelerate high-level opening-up, actively carry out international economic and trade exchanges and cooperation, build the Belt and Road with high quality, and successfully host China International Import Expo(CIIE), Service Trade Fair and Canton Fair. China’s import and export volume to countries that jointly built the Belt and Road increased by 2.8%. China International Import Expo(CIIE)’s annual intentional turnover increased by 6.7% compared with the previous session.

 

  Third, the green and low-carbon transformation has continued to deepen and the development mode has further changed. China insists on promoting economic development in the process of green and low-carbon transformation, and the green production mode and lifestyle have been accelerated. Actively build a clean, low-carbon, safe and efficient energy system and continuously optimize the energy consumption structure. According to preliminary accounting, the proportion of non-fossil energy consumption in total energy consumption in 2023 will increase by 0.2 percentage points over the previous year. By the end of 2023, the installed capacity of renewable energy in China will account for more than half of the total installed capacity, historically exceeding that of thermal power. The output of green and low-carbon products is also growing rapidly. In 2023, new energy vehicles increased by 30.3% over the previous year, and the production and sales volume were the first in the world; The export volume of electric manned vehicles increased by 67.1%.

 

  Fourth, people’s livelihood security is strong and effective, and people’s lives are further improved. Improving people’s livelihood and well-being is the fundamental purpose of development. In 2023, the income of urban and rural residents grew steadily, the level of public services and social security continued to improve, and the short-board field of people’s livelihood was gradually strengthened, further polishing the background of people’s livelihood with high-quality development. The per capita disposable income of residents increased by 6.1%, among which, the transfer income from government social relief and subsidies, policy life subsidies and cash policy subsidies for benefiting farmers increased by 10.3%. The investment in people’s livelihood has been increasing. The investment in the production and supply of electricity, heat, gas and water has increased by 23%, and the investment in agriculture has increased by 9.3%, all of which are obviously faster than the total investment.

 

  Fifth, the ability to ensure food and energy security has been improved, and the foundation for safe development has been further consolidated. In 2023, China ensured security in its development and achieved good development on the basis of security.

 

  Of course, at the same time, we must also see that China is still in the key period of transforming the development mode, optimizing the economic structure and transforming the growth momentum. To further promote high-quality development, we still need to overcome many difficulties and challenges. It is necessary to thoroughly implement the spirit of the Central Economic Work Conference, take high-quality development as the last word in the new era, make overall plans for high-quality development and high-level security, continuously improve the economic quality and reasonably increase the quantity, and turn the grand blueprint of Chinese modernization into a beautiful reality step by step. Thank you.

 

  Reuters reporter

 

  I am also concerned about the structural problems. I would like to ask, how can we realize the transformation of old and new kinetic energy in China’s economic development? It is also seen in the report that new energy vehicles and solar cells are growing rapidly, but in the process of vigorously developing advanced manufacturing industries, how can we prevent possible overcapacity problems? In addition, in the case of relatively insufficient domestic demand, will the rapid expansion of these industries bring downward pressure on prices and possible international trade frictions? Thank you.

 

 

  Wang Guanhua

 

  Thank you for your question. I want to answer your question mainly from two aspects:

 

  First of all, the first point is about the conversion of old and new kinetic energy. High-quality development is the development of implementing the new development concept, and "innovation" ranks first among the five development concepts. Adhering to the drive of innovation and promoting the transformation of development kinetic energy has always been an important task for high-quality development. Just now, Director Kang Yi gave a detailed introduction to the achievements of China’s high-quality development last year, especially the innovation-driven development. Here I would like to add a few more data. In 2023, the added value of China’s equipment manufacturing industry above designated size increased by 6.8% over the previous year, which played a key role in promoting the stable recovery of industry; In 2023, the report released by the World Intellectual Property Organization showed that China ranked 12th in the global innovation index, and the number of the top 100 scientific and technological innovation clusters in the world jumped to the first place in the world for the first time. It can be said that new kinetic energy has become an important engine to lead high-quality development.

 

  Secondly, last year, China’s economy rebounded, especially the market demand and domestic demand recovered well. Everyone should have personal feelings about this, especially with the increasing demand-driven role, China’s supply and demand convergence and economic cycle are also improving. Give you a few data for reference. In the fourth quarter, the product sales rate of industrial enterprises above designated size in China remained above 97%, which rose to 98.4% in December; Judging from the capacity utilization rate that reflects the capacity utilization situation, in the four quarters of 2023, the industrial capacity utilization rate was 74.3%, 74.5%, 75.6% and 75.9% respectively, showing a trend of quarterly recovery. This reflects that with the improvement of market demand, China’s capacity utilization is gradually recovering. At the beginning of this year, we also noticed that ice and snow tourism and ice and snow sports can be said to be "out of the circle", which not only conforms to the general trend of upgrading the consumption structure of residents, but also shows the potential of domestic demand in China. We are full of expectations and confidence in the recovery of the consumer market this year.

 

  Generally speaking, China is still in a critical period of economic recovery and transformation and upgrading. In the next step, in accordance with the decision-making arrangements of the Central Economic Work Conference, we should persist in striving for progress through stability, promoting stability through progress, establishing first and then breaking, continuously consolidate and strengthen the economic recovery to a good trend, and promote the effective improvement of quality and reasonable growth of quantity of the economy. Thank you.

 

  Singapore Straits Times reporter

 

  What I want to ask is, last year we saw the impact of real estate on the whole macro-economy. Can it remain stable this year? Thank you.

 

 

  Kang Yi

 

  Thank you for your question. Real estate has always attracted much attention. At the press conference last year, some reporters also mentioned this issue. What is the trend and what is the next step? Everyone is very concerned. According to our monitoring, after more than 20 years of development, the real estate market is in the process of adjustment and transformation. Under such a big background, all regions and departments are adapting to the new situation that the relationship between supply and demand in the real estate market has undergone major changes and adjusting and optimizing real estate policies in a timely manner. Judging from the current situation, there have been some positive changes in the real estate market, mainly in two aspects:

 

  First, the decline in real estate investment, sales and other indicators narrowed. In 2023, the investment in real estate development decreased by 9.6% compared with the previous year, and the decline rate narrowed by 0.4 percentage points compared with the previous year. The funds in place of real estate development enterprises decreased by 13.6%, which was 12.3 percentage points narrower than the previous year. The decline in commercial housing sales has narrowed significantly. In 2023, the sales area of commercial housing in China decreased by 8.5% compared with the previous year, and the sales volume decreased by 6.5%, both of which were declining, but the declines were significantly narrowed by 15.8 and 20.2 percentage points respectively. Since August, the online signing and filing volume of newly-built commercial housing has rebounded as a whole, rising by 20.2 percentage points in December compared with August. From the monitoring of 70 large and medium-sized cities, the transaction volume of new and second-hand houses is rising moderately. Second, the completed area of real estate increased rapidly. The work of "guaranteeing the delivery of the building" is progressing steadily, and the effect continues to appear. In 2023, the completed housing area of real estate development enterprises increased by 17% over the previous year.

 

  What do you think of the future trend of the real estate market? The long-term healthy development of China’s real estate market has a good foundation for several reasons:

 

  First, there is still a lot of room to improve the quantity and quality of urbanization. Just now, it was reported that the urbanization rate in 2023 is 66.16%, which is still room for improvement compared with the level of more than 80% in developed economies. The urbanization of China is still in the process of sustainable development. In the past five years, the urbanization rate has increased by 0.93 percentage points annually. Every year, more than 10 million rural residents enter cities and towns, and the scale of new citizens is relatively large, which will also bring a large number of new housing needs. Although the per capita housing area in our country is not small, the functions and structures of many houses are not reasonable, and many people need improved housing urgently, which will also form an important driving force of the real estate market, including the 70 large and medium-sized cities currently monitored. The demand for improved housing is very obvious, which is manifested in the fact that the transaction volume of second-hand houses in 70 large and medium-sized cities has exceeded the transaction volume of new houses.

 

  Second, there is great potential for building a new model of real estate development. The new model of real estate development is being actively constructed, which is the fundamental solution to solve the real estate development problems and promote the healthy development of real estate. Among them, the construction of affordable housing, the construction of public infrastructure for both ordinary and emergency use, and the reconstruction of villages in cities are all advancing rapidly. With the vigorous and orderly progress of these projects, it will help solve the urgent problems of people in housing and housing, and at the same time, it will also drive real estate-related investment and consumption and promote the healthy development of the real estate market. Thank you.

 

  China Daily reporter

 

  Recently, many international organizations and commercial organizations have raised their expectations for China’s economic growth, believing that China is still the biggest engine of global economic growth. What do you think of this? What is the economic trend of China in 2024? Thank you.

 

 

  Kang Yi

 

  Thank you for your question. How to look at the economic trend of China in 2024 is also a matter of great concern to everyone. 2024 is the 75th anniversary of the founding of People’s Republic of China (PRC), and it is a crucial year for us to implement the 14th Five-Year Plan. To predict the economic trend of this year, we should first see that we will face some challenges and difficulties, but more are favorable conditions and advantages. Taken together, the opportunities we face are greater than the challenges, and the favorable conditions are stronger than the unfavorable factors. The basic trend of China’s long-term economic improvement has not changed, and the factors supporting the high-quality development of China’s economy are accumulating and increasing. Therefore, we predict that China’s economy will continue to pick up in 2024. Specifically, there are five advantages.

 

  First, the economic growth is "good". In the four quarters of 2023, the gross domestic product (GDP) is growing positively year-on-year and quarter-on-quarter, and the economic scale is also expanding quarter by quarter, which is also a good momentum of recovery. In addition to the statistical accounting of economic aggregates, the National Bureau of Statistics has also done a job of monitoring and comparing some physical quantity indicators. Judging from the physical quantity indicators, the absolute quantities of most physical quantity indicators, such as electricity consumption, output of major industrial products, investment, import and export, have greatly exceeded the level in 2019. Some physical indicators are lower than the pre-epidemic level in 2019, mainly related to the problems mentioned by reporters just now, and are the output indicators of products related to real estate. This also reflects that our economic operation is improving as a whole. International organizations have raised their forecasts for China’s economic growth by 0.4 percentage points, the International Monetary Fund by 0.1 percentage points and the OECD by 0.1 percentage points, which shows that the international community is optimistic about China’s economic development prospects in 2024.

 

  Second, economic development is "resilient". Our country has a solid industrial base, and is the only country with all industrial categories in the United Nations industrial classification. Its supporting capacity and integration advantages are outstanding. The added value of our manufacturing industry accounts for nearly one-third of the world’s total, and our goods exports account for one-seventh of the world’s total. At the same time, China’s transportation, communication and other infrastructure networks are improving day by day, short-board areas such as education and medical care are constantly strengthening, and the supply quality of talents, funds and other factors is improving significantly. Food and energy security, industrial chain supply chain and other key areas of support capacity building have also achieved practical results, which have enhanced the resilience and room for manoeuvre of our country’s economic development, and are also the basis for the economy to be stable and far-reaching

 

  Third, high-quality development is "full of vitality". New industries are growing rapidly, new formats are improving continuously, new models are being cultivated rapidly, the economic structure is being continuously optimized, and the potential of economic development is expected to be further stimulated. In 2023, the added value of service industry accounted for 54.6% of GDP, and contributed more than 60% to economic growth. The investment in technological transformation of manufacturing industry increased by 3.8%, and the investment in high-tech industries increased by 10.3%, which was faster than the growth rate of all fixed assets investment. More importantly, China continues to promote the formation of a new situation of innovation-driven development, and China’s economy is constantly developing and growing while accelerating the cultivation of new quality productivity.

 

  Fourth, the reform and opening up has "many dividends". China adheres to and improves the basic socialist economic system, adheres to the "two unswerving", speeds up the construction of a unified national market, promotes the construction of a high-standard market system, optimizes the business environment, and creates a fairer competitive environment for all kinds of enterprises, which will continuously stimulate the enthusiasm and creativity of business entities. Including foreign-invested enterprises, continue to be optimistic about China. From January to November 2023, the number of newly established foreign-funded enterprises increased by 36.2% year-on-year, and China’s open dividend continued to be released.

 

  Fifth, the macro policy is "wide in space". The policy effects of issuing additional government bonds, reducing taxes and fees, and lowering the RRR and interest rates introduced in 2023 will continue to be released this year. This year, some new measures will be taken to optimize the reserves, and these new incremental measures and stock policies will be superimposed to protect the stable operation of the economy. At present, China’s government debt level and inflation rate are low, the policy toolbox is constantly enriched, and there is a lot of room for manoeuvre in fiscal, monetary and other policies. There are conditions and space for strengthening the implementation of macro policies.

 

  Although there will be some difficulties and challenges in promoting the sustained economic recovery in 2024, the Central Economic Work Conference held some time ago made a careful analysis of the difficulties and challenges and put forward specific countermeasures. It is necessary to fully implement the arrangements of the Central Economic Work Conference and constantly turn development advantages into development potential. This year, China’s economy will certainly be able to face difficulties and forge ahead, and achieve effective improvement in quality and reasonable growth in quantity. Thank you.

 

  The Paper reporter

 

  In 2023, the total retail sales of social consumer goods showed a trend of gradual recovery, but at present, China still faces the problem of insufficient demand. Excuse me, how do you view the performance of the consumer market in 2023? What is the prediction for 2024? Can the recovery of consumption continue? Thank you.

 

 

  Kang Yi

 

  Thank you for your question. The first question is the performance of the consumer market in 2023, and the second question is the prediction of the consumer market in 2024.

 

  Let me introduce the performance of the consumer market in 2023. Generally speaking, the consumer market will recover in 2023. In the three years since the epidemic, the consumer market has been hit hard, and many contact and aggregate consumption have been restrained. With the smooth transition of epidemic prevention and control, the economy and society have fully resumed normal operation, and consumption has shown a good recovery trend with frequent hot spots. Consumption has become an important force driving economic recovery in 2023. There are several characteristics: First, the consumption scale has reached a new high. In 2023, the total retail sales of social consumer goods exceeded 47 trillion yuan, reaching a record high. Second, consumption has become the main driving force of economic growth again. In 2023, the final consumption expenditure will drive economic growth by 4.3 percentage points, up by 3.1 percentage points over the previous year; The contribution rate to economic growth was 82.5%, an increase of 43.1 percentage points, and the basic role of consumption was more significant. Third, service consumption recovered quickly. The rapid recovery of service consumption is also a highlight of consumption recovery in 2023. The retail sales of services increased by 20% over the previous year, which was 14.2 percentage points faster than the retail sales of goods. The per capita service consumption expenditure of residents increased by 14.4%, accounting for 45.2% of the per capita consumption expenditure of residents, up by 2 percentage points over the previous year. Fourth, the structural upgrading of household consumption continues. Especially the improvement of people’s living standards and the steady growth of income, our country is currently in a period of rapid upgrading of residents’ consumption structure.

 

  We judge that there are many favorable conditions to support the sustained recovery of consumption in the next stage, and consumption will still maintain a good growth. Supporting factors: First, the consumption potential is still huge. With a population of more than 1.4 billion, the advantages of super-large-scale market are still obvious, coupled with the integrated development of urban and rural areas, the advancement of urbanization and the continuous upgrading of consumption structure, which provide a broad space for consumption growth. In particular, the consumption potential of medical care and health is expected to be further released. Second, the consumption base has been continuously consolidated. Income is the premise and foundation of consumption. With the sustained economic recovery and the overall improvement of employment situation, residents’ income is expected to maintain steady growth, which will strongly support the improvement of residents’ consumption power. Third, consumption highlights are constantly emerging. Digital consumption, green consumption, healthy consumption, cultural tourism consumption, etc. are all developing rapidly, and some consumption hotspots such as smart home, entertainment tourism, sports events, and domestic products are also heating up, which constantly adds momentum to the upgrading and expansion of the consumer market. The fourth is to promote the continuous development of consumption policies. All localities and departments insist on giving priority to restoring and expanding consumption, and have successively issued a series of policies to promote consumption, focusing on stabilizing and expanding traditional consumption, cultivating and expanding new consumption, and continuously optimizing the consumption environment, which will continue to play a positive role in stabilizing the consumer market and promoting consumption recovery. Therefore, we are optimistic about the consumption trend in 2024. Thank you.

 

  21st century business herald reporter

 

  I want to pay attention to the fifth national economic census. The fifth national economic census was officially launched on January 1 this year. What is the current progress? What are the innovations of this census compared with the previous four? How to ensure the quality of census data? Thank you.

 

 

  Kang Yi

 

  Thank you for your question. The fifth national economic census is an important survey of national conditions and national strength in the new era and new journey, which carries the important task of finding out the economic background of our country and reflecting the progress of high-quality development. At present, the unit inventory work has been successfully completed, and it has entered the stage of census registration since January 1, 2024. Now the census registration work is being carried out in an orderly manner. Especially since the beginning of the new year, on January 3rd, Vice Premier Ding Xuexiang made a survey to guide the on-site registration of the census, visited the grassroots enumerators and staff, and put forward clear requirements for doing a good job in census registration. At present, the census work is progressing in an orderly manner. January-April this year is the official registration stage of the census, which is also the most important stage of the census work and the most important stage for obtaining high-quality census data.

 

  The innovation of the fifth national economic census this year is mainly to focus on the development of high-quality services, enrich the survey content and innovate the survey methods. The fifth national economic census will comprehensively investigate the development scale, layout and benefits of China’s secondary and tertiary industries. On this basis, we should expand the field of statistical investigation, enrich the content of statistical investigation, improve the statistical investigation system and promote the construction of a high-quality statistical monitoring system. There are three main aspects of innovation:

 

  First, the input-output survey was carried out as a whole for the first time. Originally, the input-output survey and the economic census were conducted separately, and the fifth national economic census integrated these two separate surveys, which helped to promote the better connection between economic aggregate data and structural data.

 

  The second is to further improve the "three new" economic statistics. Add the content of digital economy survey, better reflect the digitalization process of China’s economic development, systematically promote the platform economy survey, and lay the foundation for finding out the development of brand-new industries, new formats and new development models.

 

  Third, there is innovation in the means and methods of investigation. Continue to deepen the application of departmental administrative records, add self-reporting methods to collect census data, develop mobile applet to collect data for the first time, and establish an input-output statistical electronic ledger for the first time to improve the efficiency of census work.

 

  The quality of census data is the most fundamental criterion to measure the success or failure of census. In the census work, we attach great importance to the quality of census data and take a series of measures to ensure the quality of census data.

 

  The first is to standardize data collection. In this economic census, census takers collect data from households, census subjects fill in their own reports and departments submit data, so as to strictly control human interference in data collection and ensure the quality of census source data.

 

  The second is to carry out data audit and inspection in various ways. We carried out the census data with the report and review, comprehensively used big data means and various analysis methods to carry out audit and verification, and organized on-site verification and inspection in time. After the census registration, the general office will also organize a spot check on the quality afterwards to comprehensively test the quality of the census registration data.

 

  The third is to resolutely investigate and deal with census fraud. Conduct a census in accordance with the law, resolutely resist all kinds of interference with census data, strengthen statistical law enforcement inspection, incorporate the fifth national economic census into statistical inspectors, and "strike out" violations of laws and regulations in the census, and seriously pursue accountability according to the law and regulations.

 

  At present, there are about 2.1 million census "two members" (census investigators and census counselors) visiting the streets, going deep into enterprises and merchants to collect data, and the census data are being reported in an orderly manner.

 

  Here, I would also like to thank all the respondents for their support and the hard work of the census workers. I also hope that the media friends will continue to support the census, publicize the census and supervise the census. Let us work together to hand over a satisfactory answer sheet for the fifth national economic census with high quality. Thank you.

 

  Cover journalist

 

  Recently, some media reported that there was a difference of 294.6 billion kWh between electricity consumption and power generation in the first 11 months of 2023. Why is there such a data difference? Thank you.

 

 

  Kang Yi

 

  Thank you for your question. We are concerned that some media discuss the difference between electricity consumption and electricity generation. Electricity generation and electricity consumption are two important indicators reflecting power operation. Just now, this reporter friend mentioned the difference between the data of these two statistical indicators from January to November, mainly because the caliber of the two indicators is different. One of these two indicators is the electricity consumption of the whole society, and the other is the power generation of industrial enterprises above designated size. In addition to industrial enterprises above designated size, with the rapid development of solar power generation and wind power generation, some industrial enterprises below designated size, various houses and merchants also generate electricity, but this part of the power generation is not within the industrial power generation above designated size from January to November.

 

  For your better understanding, I would like to explain these two indicators. Their survey objects, statistical caliber and coverage are different. First, the statistical caliber is different. The electricity consumption of the whole society is counted from the users of electricity, and the industrial power generation above designated size is counted from the suppliers. The scope of these two statistics is different. Specifically, the electricity consumption of the whole society is the full-scale electricity consumption, that is, the total electricity consumption, including the electricity consumption of enterprises and residents, as well as the self-produced and self-used electricity and line loss of power plants. The statistical caliber of industrial power generation above designated size refers to the power generation of industrial enterprises above designated size. The current standard of industrial enterprises above designated size is industrial enterprises with annual main business income of 20 million yuan or more, so it does not include the power generation of small industrial power generation enterprises below designated size, and it is not a full caliber power generation.

 

  Second, it is affected by some distributed power generation. In recent years, distributed solar energy and wind power generation, which are widely dispersed and small in size, have developed rapidly, and they are mostly distributed in various types of residential buildings, merchants and some industrial enterprises below the scale. Because the power generation enterprises are small in scale, this part of the power generation does not meet the industrial statistical standards above the scale, and some of them are merchants and residential buildings, so it is not counted in the monthly industrial power generation above the scale, but will be counted in the whole society’s power generation. As we all know, our data are usually released in the middle of the next month. So many small-scale power generation scattered around the country will be counted annually, and the power generation of the whole society will also be released in the annual statistical bulletin. Generally speaking, the industrial power generation above designated size accounts for about 95% of the total social power generation, which has fluctuated a little recently. The power generation below designated size is increasing, while the power generation above designated size is decreasing, but the decrease is not big, which is probably the ratio. Look, everybodymonthlyThe expression "industries above designated size" cannot be ignored in power generation. Statistical indicators are relatively rigorous, and everyone should also pay attention to the meaning and scope of statistical indicators when using them, so as to facilitate the more accurate use of data. Thank you.

 

  Southern Metropolis Daily reporter

 

  What is the employment situation in 2023? Did you complete the target task? The number of college graduates in 2024 is expected to reach a new high. What will be the employment trend? Thank you.

 

 

  Kang Yi

 

  Thank you for your question. Employment is also a matter of great concern to everyone, because it is the biggest livelihood. The CPC Central Committee and the State Council have always attached great importance to the employment issue, emphasizing the promotion of stable employment to a strategic level. All localities and departments insist on giving priority to employment and optimizing and adjusting policies and measures to stabilize employment. In the past year, we made every effort to stabilize the overall employment situation and the overall employment situation improved. There are several characteristics.

 

  First, investigate the decline of unemployment rate and the continuous increase of new jobs. On a quarterly basis, the national urban survey unemployment rate is 5.4%, 5.2%, 5.2% and 5.0% respectively, and the trend of gradual improvement in employment is obvious. From January to November, the number of new jobs in cities and towns was 11.8 million, an increase of 350,000 over the same period of last year.

 

  Second, key groups and difficult groups have strong employment security. A series of job stability support and job expansion incentives came into effect, and the classified assistance for groups with employment difficulties was effective, and the employment of migrant workers, young people, people with employment difficulties and other groups was effectively guaranteed. In 2023, the total number of migrant workers reached 297.53 million, an increase of 1.91 million over the previous year. The average unemployment rate of migrant agricultural registered population in cities and towns was 4.9%, a decrease of 0.7 percentage points over the previous year. From January to November, the number of unemployed people in cities and towns was 4.75 million, and the number of people with employment difficulties was 1.56 million.

 

  Third, the employment scale of people out of poverty has increased steadily. All localities and departments give full play to the mechanisms of labor cooperation, counterpart support and fixed-point assistance between the east and the west, and carry out in-depth special assistance for poverty alleviation and relocation in key counties and ex situ, so that the employment of poverty-stricken people will grow steadily. By the end of November 2023, the number of migrant workers will be 32.94 million, exceeding the target of 30 million.

 

  In this year’s employment situation, the overall judgment is that the pressure still exists, and the structural contradictions in employment of some groups and some industries will still be more prominent. However, with the recovery of economy, the acceleration of industrial transformation and upgrading, and the accumulation of positive factors to stabilize employment, China’s employment situation is expected to remain stable. There are several reasons:

 

  First, the expansion of economic scale has brought about an increase in employment. Economic growth is the basis for stabilizing and expanding employment. In recent years, the continuous expansion of China’s economic growth is the key to stimulating employment growth. In 2024, the increment of China’s economic creation is expected to be greater than last year, which will provide strong support for expanding employment. Second, the number of people leaving the labor market in 2024 will be larger than the number of people newly entering the labor market, which also provides more employment space for people looking for jobs. The third is to upgrade the industrial structure and expand the employment capacity. Compared with other industries, the service industry with higher labor intensity has obvious advantages in absorbing employment. Since last year, the service industry has recovered well, and its proportion in GDP is also increasing. The proportion of service industry in GDP has exceeded the level before the epidemic, and the employment-driven role of catering, transportation, wholesale and retail industries is more obvious. Looking forward to this year, service consumption will be more active, and the growth of service industry will continue to be one of the main forces driving employment absorption. At the same time, new industries, new formats and new business models are developing vigorously, generating many brand-new job demands, which is also conducive to expanding employment space and improving employment quality. Fourth, the effect of stabilizing employment policy continues to exert significant effects. The Central Economic Work Conference has also made arrangements for this, especially pointing out that more policies are needed to stabilize expectations, growth and employment. All localities and departments will give more prominence to employment priority orientation, increase employment assistance for key groups, and the release of policy dividends is expected to continue to provide a strong guarantee for employment stability. Thank you.

 

Don’t get me wrong about these English (8)

Eminem, an American rapper, brought the film to perfection in the last part of his semi-autobiographical film "Eight Miles". Jimmy, who he played, won the championship in the impromptu rap contest in Detroit’s black area. When all the rappers waved their arms up and down with the music, you might feel excited. Many rappers saw this part and strengthened their belief in making Chinese rap music, but this English that made the audience high was easily misunderstood as other meanings.

1. bring down the house.

Get a house full of applause, not knock down the house.

Don't get me wrong about these English (8)

2. have a fit.

Lose one’s temper, fly into a rage, not trying on clothes

Don't get me wrong about these English (8)

3. make one’s hair stand on end.

It’s creepy, not outrageous.

Don't get me wrong about these English (8)

4. be taken in.

Deceived, not accepted.

Don't get me wrong about these English (8)

5. pull up one’s socks.

Courage, not socks.

Don't get me wrong about these English (8)

In a blink of an eye, 2013 is over half. How is your dream coming true? Courage to pursue your dream can be not only someone else’s business, but also your business. I also wish you pull up your socks and chase your dream. See you next week!

Related reading

Don’t get me wrong about these English (7)

Don’t get me wrong about these English (6)

Don’t get me wrong about these English (5)

Don’t get me wrong about these English (4)

Don’t get me wrong about these English (3)

Don’t get me wrong about these English (2)

Don’t get me wrong about these English meanings (1)

About the author:

Don't get me wrong about these English (8)

Leon, graduated from Harbin Institute of Technology, is a bilingual psychotherapist. Former editor-in-chief, good at American English, friends ashes powder, hobby HipHop and raising dogs. Now I live in Beijing and have my own psychological studio.

(Editor of China Daily English Dianjin Julie)

Starting from 70,900 after the discount, I went to the store to shoot the third-generation Yidong of Chang ‘an, and the new Blue Whale was equipped with 1.5T power as standard.

In March of this year, Changan Automobile officially launched the third generation. Because of its renewed power and upgraded size, the third generation Yidong has attracted the attention of many friends. Now that the new car has been on the market for almost half a year, what kind of market is it in the terminal market at present? With such questions, we came to Chang ‘an 4S store to take a real shot, and also asked about the price to see what the performance of the third generation of Yidong was like.

Compared with the old models, the third-generation Yidong has actually changed a lot. First of all, from the appearance, the third-generation Yidong has obviously become more avant-garde and sporty. Among them, its front area uses a large-scale borderless grille, and the internal filling is still arranged in a lattice, but the spokes are relatively thin. The left and right headlights are designed in one piece, which is relatively slender as a whole. The front enclosure and the low-lying front create a good wide-body effect visually, and it seems that the gas field is still very strong.

On the side of the car body, the third generation Yidong uses a relatively simple design, and is also equipped with a popular hidden door handle, which is very modern. The combination of the multi-spoke and the bright decorative strips on the upper part of the window enhances the exquisite feeling of the side and looks very young.

The broken line design used in the tail is rich, and the rising tail, slender taillights and exhaust port decoration on both sides are all more interesting elements. In addition, the edge of the taillight with blackening effect and the LOGO of the car logo also make the sports characteristics of the tail more obvious and more suitable for the tastes of young people.

In the interior, the third-generation Yidong uses a 13.2-inch touch control panel and a 10.25-inch LCD instrument panel, which basically undertake the functions of information display and vehicle control in the car, so the physical buttons in the third-generation Yidong car are also much less.

Instead, it is touch control or voice control. The third generation of Yidong comes standard with AI Super Brain Xiaoan, which has stronger understanding ability and knowledge question-and-answer ability. In the actual experience, daily functions such as navigation, music playing and air conditioning control can be directly controlled by voice, and the system has a high recognition degree of voice commands. Like window control, driving mode switching, etc., personally, it is more convenient to use physical control operation, and it can also achieve blind operation, and personal experience will be better when driving.

In terms of interior configuration, the functions such as 540 panoramic image and Bluetooth key of mobile phone, which are more practical, are all standard in the third generation. In addition, the functions such as rear parking radar, cruise control, inductive tailgate and remote start are also available. Therefore, in my opinion, if the car purchase budget is limited and relatively cost-effective, the third-generation Yidong entry-level model is a more cost-effective option.

The third-generation easy-to-move seat has a certain sports style and is wrapped in leather, which feels good. The backrest and both sides of the seat are well supported, while the seat cushion is relatively soft, so the overall comfort is good. In terms of size, the third-generation Yidong has a certain upgrade compared with the old model. The length, width and height are 4770*1840*1440mm, and the wheelbase is 2765mm, while the Nissan classic of the same level has a wheelbase of 2700 mm. In contrast, the third-generation Yidong does have some advantages.

In terms of internal space, taking the experiencer with a height of 175cm as a reference, after coming to the main driving position and adjusting the sitting position, the experiencer can still have a punch left in the overhead space. Keep the angle of the main driver’s seat fixed and come to the second row to ride. At this time, the experiencer can still have one punch left in the overhead space and more than two punches in the leg space. Therefore, for ordinary people, the spatial performance of the third generation of escape is still relatively spacious.

In the trunk part, under normal circumstances, the volume of the third generation escape is 520L, and the internal loading capacity is still considerable, and the depth is relatively large. If the items are placed slightly in front, basically the whole upper body needs to reach the trunk to get things.

In terms of power and body, the third-generation Yidong comes standard with the new Blue Whale power, using the NE1.5T high-pressure direct injection engine of the new generation of Blue Whale and the 7-speed wet powershift of the new generation of Blue Whale. The maximum power of the engine is 125kW, the maximum torque is 260N·m, and the No.92 gasoline is filled. The comprehensive fuel consumption of WLTC is 6.14L/100km, the acceleration time of 0-100km/h is 9.1 seconds, and the maximum speed can reach. In terms of body, the third generation of Yidong used Changan Ark cage safety body, and used more than 70% ultra-high strength steel, which showed good passive safety performance.

The overall dynamic performance of the third-generation Yidong is still dominated by household comfort. The initial power output is relatively more sensitive in the standard mode, and the gear shift of the transmission is slightly delayed in the economic mode, in exchange for better ride comfort performance. Therefore, when driving in the middle of the city, the whole car performs well even in stop-and-go road conditions. In terms of acceleration in the middle and late stages, the power reserve of the third generation Yidong is also good, and overtaking is more powerful, which can bring better driving pleasure.

As a domestic family car with a history of 12 years and accumulated users exceeding 1.7 million, the starting price of 70,900 for the third generation Yidong seems to me to be quite favorable at present, and it has a high cost performance in combination with the vehicle itself. Therefore, if you want to buy a car this summer vacation, I think the third generation of Changan Yidong is worth considering, don’t you think?

Lantu Automotive CEO Lu Fang: Walking on thin ice every day, only by maintaining competition can we survive

Author | Chen Yixun

Edit | Sun Chunfang

2022 is destined to be an extraordinary year for all new energy vehicle companies. In the first half of 2022, the entire automotive industry encountered problems such as epidemic, rising raw material prices, and tight chip supply. During this period, it was inevitable that production and sales would be affected to a certain extent. How to overcome resistance and maintain the sustainability of enterprises has become a problem that new energy vehicle companies, including Lantu Automobile, need to think about together.

In November 2022, Lantu Auto announced the completion of the A round of financing of nearly 5 billion yuan, and the valuation of Lantu Auto after financing is nearly 30 billion yuan. After this round of financing, Lantu Auto does not rule out continued financing. At the same time, Lantu Auto will also consider the follow-up IPO plan based on the market and regulatory policies.

As a new energy vehicle brand under Dongfeng Group, Lantu Automobile was established in 2018 and later spun off from the parent company for independent operation in 2021. In December 2021, Lantu Automobile’s Series A financing was officially launched, and after nearly a year, it finally landed.

In the second half of 2022, new forces began to reshuffle, and capital markets became more cautious and conservative in investing in new energy vehicles. Lu Fang, CEO of Lantu Automobile, said that the current international economic situation is severe, and in the complex environment of multiple unfavorable factors, the investment and financing of the automobile and parts industry has become more difficult. In this context, the successful financing of Lantu Automobile reflects the full recognition and confidence of investors in the primary market in the value of Lantu Automobile’s investment.

However, Lu Fang also said: "Now I feel like walking on thin ice every day, I have to participate in the competition as a competitor, and I have to survive the competition."

Q: Lantu Auto recently received 5 billion yuan A round of financing. What is the main use of this money? Will there be financing plans in the future? Is there an approximate time point for listing, and is there some information to disclose and share?

A: This financing is mainly used to support Lantu technology research and development, digital system construction, capacity building and marketing investment and other related matters related to Lantu’s main business operation and expansion. It is expected that more than 40% of the expenses will be used for technology research and development. We hope to use the money to really support the sustainable development of Lantu.

In the future, on the basis of this round of successful financing, Lantu will continue to carry out follow-up rounds of financing work according to actual development needs and market conditions. Follow-up financing is not only financial investment, but also strategic coordination in business. Please refer to Lantu’s follow-up announcement for specific capital operation plans.

Q: Lantu Auto has recently gone to Norway. What is Lantu’s follow-up plan for going to sea? Did you encounter any difficulties in the process of going to sea?

A: Entering Norway is just the first step for Lantu to go overseas. At present, Lantu Dreamer is carrying out adaptive development in Europe and is expected to go overseas in 2023. In the future, Lantu will also enter the market of other European countries. The ultimate goal is to enter the German or French market through efforts in the next two or three years. Germany and France are traditional automobile powerhouses. If they can be recognized by local consumers in their markets, it will prove that Lantu’s cars can be recognized by most people in the world.

Lantu has entered Norway as a new brand that has only been established for more than two years. It is head-to-head with mature European brands. Its reputation and product strength need to be verified by the local market, and it will definitely face certain challenges. Lantu has been developing and testing corresponding products in accordance with the consumer needs, regulatory requirements, and market environment of Norway and other European countries.

At present, Chinese car companies go overseas to Norway in the form of vehicle exports. Under the influence of the pandemic, there will be a certain degree of delays and challenges in the transportation link. How to adapt to local laws and regulations, policies, user usage habits and local usage environment are some of the problems that car companies will face when going overseas. To enter the local market, the most important thing is to understand the needs of users, integrate with local consumers and culture, and then reach a consensus.

Q, the product pricing of Lantu Automobile is above 300,000, this kind of high-end play is very different from BYD and Aian starting from the low end and then challenging the high end. What kind of market and product reasons does Lantu make this decision?

A: At present, China’s manufacturing and consumption are upgrading, moving towards a higher direction, and the same is true for the automotive industry. Chinese cars cannot always compete at a low level, and high-end brand power can support sustainable development. At the same time, with consumption upgrading, consumers’ demand for high-end cars is strong, and China has better independent brands to meet consumers’ needs in this regard, rather than being occupied by BBA. At present, consumers also prefer to buy smart electric vehicles rather than fuel vehicles. At this time, domestic car brands are naturally needed to meet consumers’ needs.

It must be difficult to be a high-end brand at the beginning, but although things are difficult, someone has to do it. Lantu is a high-end electric brand under Dongfeng, carrying the important tasks of the group’s brand upward and the country’s independent mastery of core technologies. High-end is not high-end in price, but should be high-end in all aspects of products, services and experiences. At the same time, high-end brands can also adapt to the needs of the overall economy and user consumption upgrades after they are made.

Q. Intelligence is a major selling point of new energy vehicles. What stage has Lantu Automobile reached in the development of intelligent technology? Are there any plans to acquire mobile phone companies or cooperate with other Internet companies?

A: LANTU has been doing electrification and intelligence since the beginning, and has done a lot of work in intelligent driving, intelligent cockpit, etc. Last month, LANTU released its ESSA + SOA intelligent electric bionic, which solves the problems faced by new energy vehicles in both electrification and intelligence to the greatest extent. By decoupling the software from the hardware, the car becomes a self-defining product like a mobile phone, and LANTU can open up a lot of content to developers and users. This is a revolutionary change in the electronic and electrical architecture of the whole vehicle.

At present, Landmap has no plans to cooperate with mobile phone companies.

Q. How many people are there in the current Lantu automotive team? What is the proportion of R & D personnel in it? How much is the annual investment in R & D?

A: Landmap Automotive currently has more than 4,000 personnel. In the field of R & D, Landmap has assembled a R & D team with electrification and intelligent engineers as the core. Landmap R & D personnel account for 38% of the total number of technology companies, and the proportion of R & D teams is the largest in new vehicle companies. The number of personnel in the core R & D field increased by 73% compared to October 2021. The average age of the team is 32.6 years old, and the proportion of graduate students and above is 38%.

In addition to full-stack self-research, Dongfeng Group has also provided a lot of technical support to Landmap in terms of Sandian, intelligent networking, etc. Landmap will turn these advantages into the source of its own development.

Q. Previously, Lantu Automobile also talked about that it should not become Wei Xiaoli, nor will it become Wei Xiaoli. Lantu’s positioning is a national high-end new energy brand, and its mission is to compete head-on when BBA is fully electrified. BBA has also begun its electric transformation, and BBA has the genes of traditional luxury brands. As a brand new brand, what is the biggest advantage of Lantu Automobile in competing with BBA?

A: In recent years, in terms of the development of China and the international automobile market, whether it is supply chain, technology or product strength, Chinese automobiles have been completely inferior to international traditional automobile powers, especially China’s leading technological advantages in the "New Four Modernizations" and "Three Electrics", so that China’s automobile product level, R & D capabilities and production quality have the strength to compete with foreign high-end brands on the same stage.

In terms of technology, the electronic and electrical architecture of Lantu vehicles is completely based on pure electric drive, and it is a real electric vehicle rather than an oil-to-electricity product. In fact, many products, including BBA, are still based on the development concept of oil-to-electricity; secondly, from joint ventures to autonomous Chinese cars, whether in the development of vehicle performance or in the control of vehicle quality, including in the supply chain, we have the ability to create products that are not inferior to BBA or even superior to them. More importantly, in high-end smart electric vehicles, China is far ahead, and many ecological partners in Europe also admit that the intelligence of Chinese electric vehicles is at least 5-10 years ahead of them.

Frankly speaking, the current Chinese car brand power may not be as good as that of foreign countries, but with the passage of time, China’s brand power may soon surpass them at some point. In a few years, the advantage of Chinese brands may become more obvious, and even part of the BBA market will be competed for and transformed into China’s own high-end new energy vehicle brand market.

Q. Many companies are talking about user-based enterprises and user co-creation. What is Landmap’s understanding of user-based enterprises?

A: The so-called user-based enterprises, it is crucial to understand the user’s pain points. Many times, the user pain points that enterprises understand are not the same as the user’s own pain points. The key is to understand from the user’s perspective. Returning to the source depends on how to continuously provide users with reliable products and services through technological innovation.

Landmap’s understanding of user-based enterprises is not a NIO-style service type, nor is it only focused on the single dimension of cockpit or intelligent driving. Landmap has vehicle technology, Internet technology, and an ecological development platform based on the future. It integrates self-developed technical capabilities such as platforms and architectures to build connections with users, focusing on users’ needs in vehicle driving control, intelligence, and safety, and realizes comprehensive technology.

Q. What do you think of the new energy vehicle market in the past two years, the rise of Tesla and new Chinese automakers?

A: New forces have a lot of innovative thinking worth learning. Together, they can expand the market, which is good for all high-end electric brands. Electric vehicles have the ability to compete with traditional fuel vehicles. Only with competition can they grow. Only with excellent competitors can they stimulate their maximum potential and continuously move towards a better direction.

But at the same time, the competition in the future will also be very cruel. Not only are traditional car companies making new energy vehicles, but many new players are also entering the industry. If you look at the mobile phone industry in the past 10 or 20 years, the new car industry will also leave a batch to be eliminated. Lantu is also walking on thin ice every day. We hope that we can survive and cannot stop or slow down, so we must participate in the cruel competition as competitors in this process.

LANTU builds cars for users, not for capital. Sales are important, but LANTU pays more attention to the user experience. LANTU and Wei Xiaoli are the brothers of the new Chinese car-making forces. The competitor of the new force is Tesla, and the competitor of LANTU is the brand of high-end foreign group companies. Let’s transform together to create a healthy development environment for China’s high-end electric brands.

Q. Now the new forces, including Wei Xiaoli, are basically in the midst of losses. The investment in the early stage of the enterprise is naturally important, but when the sales volume is still climbing, the continuous loss will inevitably bring financial pressure. Will it have an impact on the future development of the enterprise? What are the plans of Landmap in terms of profitability?

A: From a business perspective, we definitely hope to achieve profitability as soon as possible. At present, Landmap is constantly increasing sales while also controlling costs. For new energy, you still need to be patient. The auto industry is a long-term thing, and building a new brand is also a long-term thing. If you just take a small time stage of the company to judge and compare, and then make a conclusion, the result will definitely be unfair. In fact, it took a long time for Tesla to make a profit. I believe that all excellent companies will eventually make a profit.

Q. What are the greatest achievements of Landmap in 2022? What are the plans for 2023?

A: In 2022, Lantu completed a series of major actions, completed the product layout of three major categories in three years, went to Europe, completed the A round of financing, released the world’s first ESSA native intelligent electric architecture and centralized SOA electronic and electrical architecture, and was selected for the State-owned Assets Supervision and Administration Commission of the State Council "Double Hundred Enterprises" list. It can be said that as a brand established for only two years, Lantu continued to maintain rapid development in 2022. But there are also some regrets, such as the impact of the epidemic on the automotive supply chain, and the impact on the factory production end.

In 2023, Lantu will continue to maintain rapid development. At the brand level, Lantu will continue to adhere to the brand positioning of high-end new energy vehicles, face the competition of foreign luxury brands, and promote the improvement of national independent car brands; in terms of products, the electric car Lantu Chasing Light will be released soon, and it will be delivered to users in 2023. Lantu will have three sub-products to serve users. Lantu will also continue to adhere to the strategic layout of one new car a year, and lay out more market segments in the fields of SUVs, MPVs, and cars to meet differentiated user requests. In terms of international market development, the first batch of Lantu FREE to go to Norway has arrived in Norway. After the Norwegian market, Lantu Dreamer will also be officially launched into the European market in 2023. In the future, Landmap cars will also enter four countries: Sweden, the Netherlands, Denmark, and Israel.

Light | deep learning empowered optical metrology

Writing | Zuo Chao Qian Jiaming

In March 2016, DeepMind, a Google-owned artificial intelligence (AI) company, defeated Go world champion Lee Sedol 4:1 with its AlphaGo artificial intelligence system, triggering a new wave of artificial intelligence – deep learning technology. Since then, people have witnessed the rapid rise and wide application of deep learning technology – it has solved many problems and challenges in computer vision, computational imaging, and computer-aided diagnostics with unprecedented performance. At the same time, Google, Facebook, Microsoft, Apple, and Amazon, the five tech giants without exception, are investing more and more resources to seize the artificial intelligence market, and even transforming into artificial intelligence-driven companies as a whole. They have begun to "ignite" the "art" of data mining and developed easy-to-use open-source deep learning frameworks. These deep learning frameworks enable us to use pre-built and optimized component sets to build complex, large-scale deep learning models in a clearer, concise, and user-friendly way without having to delve into the details of the underlying algorithms. Domestic "BAT" also regards deep learning technology as a key strategic direction, and actively deploys the field of artificial intelligence with its own advantages. Deep learning has rapidly left the halls of academia and is beginning to reshape industry.

Optical metrology, on the other hand, is a type of measurement science and technology that uses optical signals as the standard/information carrier. It is an ancient discipline, because the development of physics has been driven by optical metrology from the very beginning. But in turn, optical metrology has also undergone major changes with the invention of lasers, charge-coupled devices (CCDs), and computers. It has now developed into a broad interdisciplinary field and is closely related to disciplines such as photometry, optical engineering, computer vision, and computational imaging. Given the great success of deep learning in these related fields, researchers in optical metrology cannot suppress their curiosity and have begun to actively participate in this rapidly developing and emerging field. Unlike traditional methods based on "physics a priori", "data-driven" deep learning technology offers new possibilities for solving many challenging problems in the field of optical metrology, and shows great application potential.

In this context, in March 202,Nanjing University of Science and TechnologywithNanyang Technological University, SingaporeThe research team published in the top international optical journal "Lighting: Science & Applications"A joint statement entitled"Deep learning in optical metrology: a review"The first author of the review article is Nanjing University of Science and TechnologyZuo ChaoProfessor, PhD student at Nanjing University of Science and TechnologyQian JiamingCo-first author, Nanjing University of Science and TechnologyZuo Chao,Chen QianProfessor, Nanyang Technological University, SingaporeChandler SpearProfessor is the co-corresponding author of the paper, and Nanjing University of Science and Technology is the first unit of the paper.

This paper systematically summarizes the classical techniques and image processing algorithms in optical metrology, briefly describes the development history, network structure and technical advantages of deep learning, and comprehensively reviews its specific applications in various optical metrology tasks (such as fringe denoising, phase demodulation and phase unwrapping). By comparing the similarities and differences in principle and thought between deep learning methods and traditional image processing algorithms, the unique advantages of deep learning in solving "problem reconstruction" and "actual performance" in various optical metrology tasks are demonstrated. Finally, the paper points out the challenges faced by deep learning technology in the field of optical metrology, and looks forward to its potential future development direction.

Traditional optical metrology

Image generation model and image processing algorithm

Optical metrology technology cleverly uses the basic properties of light (such as amplitude, phase, wavelength, direction, frequency, speed, polarization and coherence, etc.) as the information carrier of the measured object to realize the acquisition of various characteristic data of the measured object (such as distance, displacement, size, morphology, roughness, strain and stress, etc.). Optical metrology has been increasingly widely used in CAD /CAE, reverse engineering, online detection, quality control, medical diagnosis, cultural relics protection, human-interaction machine and other fields due to its advantages of non-contact, high speed, high sensitivity, high resolution and high accuracy.In optical metrology, the most common information carriers are "streaks" and "speckles."For example, the images processed by most interferometry methods (classical interference, photoelasticity, digital speckle, digital holography, etc.) are interference fringes formed by the coherent superposition of object light and reference light, and the measured physical quantity is modulated in the phase information of the interference fringes. In addition, the fringe pattern can also be generated in a non-interferometric way, such as fringe projection profilometry (FPP) directly projecting the fringe pattern of structured light to the surface of the measured object to measure the three-dimensional surface shape of the object. In digital image correlation (DIC), the captured image is the speckle pattern before and after the deformation of the sample surface, from which the total field displacement and deformation distribution of the measured object can be obtained. Combining DIC with stereo vision or photogrammetry, the depth information of the measured scene can also be obtained based on multi-view speckle images. Figure 1 summarizes the image generation process of these techniques and their corresponding mathematical models.

Figure 1 The image generation process and corresponding mathematical model in traditional optical metrology technology

Traditional optical metrology is inseparable from image processing technologyImage processing of fringe/speckleIt can be understood as a process of inverting the required physical quantity to be measured from the captured original intensity image. Usually, this process is not "Instead, it consists of three logically hierarchical image processing steps – pre-processing, analysis, and post-processing.Each step involves a series of image processing algorithms, which are layered on top of each other to form a "pipeline" structure [Figure 2], where each algorithm corresponds to a "map"Operation, which converts the matrix input of an image/similar image into the output of the corresponding dimension (or resampling)."

(1) PretreatmentImage preprocessing improves image quality by suppressing or minimizing unnecessary interference signals (such as noise, aliasing, distortion, etc.). Representative image preprocessing algorithms in optical metrology include image denoising, image enhancement, color channel separation, and image registration and correction.

(2) Analysis: Image analysis is the core step of image processing algorithms, which is used to extract important information carriers related to the physical quantities to be measured from the input image. In phase measurement technology, the main task of image analysis is to reconstruct phase information from fringe images. The basic algorithms include phase demodulation and phase unfolding. For stereo matching technology, image analysis refers to determining the displacement vector between points corresponding to the speckle image (the speckle pattern before and after the deformation of the sample surface/the multi-view speckle image), which generally includes two steps of subset matching and sub-pixel optimization.

(3) Post-processing:The purpose of image post-processing is to further optimize the measured phase data or speckle displacement fields and eventually convert them into physical quantities to be measured. Common post-processing algorithms in optical metrology include noise removal, error compensation, digital refocus, and parameter conversion. Figure 3 provides an overview of the image processing hierarchy of optical metrology and various image processing algorithms distributed in different layers.

A typical image processing process for optical metrology (e.g. fringe projection profiling) can be divided into three main steps: preprocessing (e.g. denoising, image enhancement), analysis (e.g. phase demodulation, phase unwrapping), and post-processing (e.g. phase-depth mapping).

Figure 3 Overview of the optical metrology image processing hierarchy and various image processing algorithms distributed in different layers

Deep learning technology

Principle, development and convolutional neural networks

Deep learning is an important branch in the field of machine learning. It builds neural structures that simulate the information processing of the human brainArtificial neural networks (ANN), enabling machines to perform bottom-up feature extraction from large amounts of historical data, thus enabling intelligent decision-making on future/unknown samples. ANN originated from a simplified mathematical model of biological nerve cells established by McCulloch and Pitts in 1943 2 ?? [Fig. 4a]. In 1958, Rosenblatt et al 2 ??, inspired by the biological nerve cell model, first proposed a machine that could simulate human perceptual abilities – a single-layer perceptron. As shown in Fig. 4b, a single-layer perceptron consists of a single nerve cell. The nerve cell maps the input to the output through a non-linear activation function with bias (b) and weight (w) as parameters. The proposal of perceptrons has aroused the interest of a large number of researchers in ANNs, which is a milestone in the development of neural networks. However, the limitation that single-layer perceptrons can only handle linear classification problems has caused the development of neural networks to stagnate for nearly 20 years. In the 1980s, the proposal of backpropagation (BP) algorithm made it possible to train multi-layer neural networks efficiently. It continuously adjusts the weights between nerve cells based on the chain rule to reduce the output error of multi-layer networks, effectively solving the problem of nonlinear classification and learning, triggering a boom in "shallow learning" 2 2. In 1989, LeCun et al. 2 3 proposed the idea of convolutional neural networks (CNNs) inspired by the structure of mammalian visual cortex, which laid the foundation for deep learning for modern computer vision and image processing. Subsequently, as the number of layers of neural networks increased, the problem of layer disappearance/explosion of BP algorithm became increasingly prominent, which caused the development of ANN to stagnate in the mid-1990s. In 2006, Hinton et al. proposed a deep belief network (DBN) training method to deal with the problem of layer disappearance; at the same time, with the development of computer hardware performance, GPU acceleration technology, and the emergence of a large number of labeled datasets, neural networks entered the third development climax, from the "shallow learning" stage to the "deep learning" stage. In 2012, AlexNet based on CNN architecture won the ImageNet image recognition competition in one fell swoop, making CNN one of the mainstream frameworks for deep learning after more than 20 years of silence. At the same time, some new deep learning network architectures and training methods (such as ReLU 2 and Dropout 2)It was proposed to further solve the problem of layer disappearance, which promoted the explosive growth of deep learning. In 2016, AlphaGo, an artificial intelligence system developed by Google’s AI company DeepMind, defeated Lee Sedol, the world champion of Go, which aroused widespread attention to deep learning technology among all mankind 2. Figure 4 shows the development process of artificial neural networks and deep learning technologies and the structural diagram of typical neural networks.

Figure 4 The development process of deep learning and artificial neural networks and the structural diagram of typical neural networks

Figure 5 Typical CNN structure for image classification tasks  

A) A typical CNN consists of an input layer, a convolutional layer, a fully connected layer, and an output layer b) a convolutional operation c) a pooling operation

The single-layer perceptron described above is the simplest ANN structure and consists of only a single nerve cell [Fig. 4b]. Deep neural networks (DNNs) are formed by connecting multiple layers of nerve cells to each other, with nerve cells between adjacent layers typically stacked in a fully connected form [Fig. 4g]. During network training, the nerve cell multiplies the corresponding input by a weight coefficient and adds it to the bias value, outputting it to the next layer through a non-linear activation function, while network losses are computed and backpropagated to update network parameters. Unlike conventional fully connected layers, CNNs use convolutional layers to perform feature extraction on the input data 2 [Fig. 5a]. In each layer, the input image is convoluted with a set of convolutional filters and added biases to generate a new output image [Fig. 5b]. Pooling layers in CNNs take advantage of the local correlation principle of the image to subsample the image, reducing the amount of data processing while preserving useful information [Fig. 5c]. These features make CNNs widely used in tasks of computer vision, such as object detection and motion tracking. Traditional CNN architectures are mostly oriented towards "classification" tasks, discarding spatial information at the output and producing an output in the form of a "vector". However, for image processing tasks in optical metrology techniques, neural networks must be able to produce an output with the same (or even higher) full resolution as the input. For this purpose, a fully convolutional network architecture without a fully connected layer should be used. Such a network architecture accepts input of any size, is trained with regression loss, and produces pixel-level matrix output. Networks with such characteristics are called "fully convolutional network architectures" CNNs, and their network architectures mainly include the following three categories:

(1) SRCNN:Dong et al. 3 2 skip the pooling layer in the traditional CNN structure and use a simple stacking of several convolutional layers to preserve the input dimension at the output [Fig. 6a]. SRCNN constructed using this idea has become one of the mainstream network frameworks for image super-resolution tasks.

(2) FCN:A fully convolutional network (FCN) is a network framework for semantic segmentation tasks proposed by Long et al. As shown in Figure 6b, FCN uses the convolutional layer of a traditional CNN [Fig. 5] as the network coding module and replaces the fully connected layer with a deconvolutional layer as the decoding module. The deconvolutional layer is able to upsample the feature map of the last convolutional layer so that it recovers to an output of the same size as the input image. In addition, FCN combines coarse high-level features with detailed low-level features through a skip structure, allowing the network to better recover detailed information while preserving pixel-level output.

(3) U-Net:Ronneberger et al. made improvements to FCN and proposed U-Net network 3. As shown in Figure 6c, the basic structure of U-Net includes a compressed path and an extended path. The compressed path acts as the encoder of the network, using four convolutional blocks (each convolutional block is composed of three convolutional layers and a pooling layer) to downsample the input image and obtain the compressed feature representation; the extended path acts as the network decoder using the upsampling method of transposed convolution to output the prediction result of the same size as the input. U-Net uses jump connection to perform feature fusion on the compressed path and the extended path, so that the network can freely choose between shallow features and deep features, which is more advantageous for semantic segmentation tasks.

The above-mentioned fully convolutional network structure CNN can convert input images of any size into pixel-level matrix output, which is completely consistent with the input and output characteristics of the "mapping" operation corresponding to the image processing algorithm in the optical metrology task, so it can be very convenient to "deep learning replacement" for traditional image processing tasks, which laid the foundation for the rapid rise of deep learning in the field of optical metrology.

Fig.6 Three representative fully convolutional network architectures of CNNs capable of generating pixel-level image output for image processing tasks

A) SRCNN b) FCN c) U-Net.

Optical metrology in deep learning

Changes in thinking and methodology

In optical metrology, the mapping between the original fringe/speckle image and the measured physical quantity can be described as a combination of forward physical model and measurement noise from parameter space to image space, which can explain the generation process of almost all original images in optical metrology. However, extracting the physical quantity to be measured from the original image is a typical "inverse problem". Solving such inverse problems faces many challenges, such as: unknown or imprecise forward physical model, error accumulation and local optimal solution, and pathology of inverse problems. In the field of computer vision and computational imaging, the classic method for solving inverse problems is to define the solution space by introducing the prior of the measured object as a regularization means to make it well-conditioned [Figure 7]. In the field of optical metrology, the idea of solving the inverse problem is quite different. The fundamental reason is that optical metrology is usually carried out in a "highly controllable" environment, so it is more inclined to "actively adjust" the image acquisition process through a series of "active strategies", such as lighting modulation, object regulation, multiple exposures, etc., which can reshape the original "sick inverse problem" into a "well-conditioned and sufficiently stable regression problem". For example, demodulating the absolute phase from a single fringe image: the inverse problem is ill-conditioned due to the lack of sufficient information in the forward physical model to solve the corresponding inverse problem uniquely and stably. For researchers in optical metrology, the solution to this problem is very simple: we can make multiple measurements, and by acquiring additional multi-frequency phase-shifted fringe images, the absolute phase acquisition problem evolves into a good-state regression problem. We can easily recover the absolute phase information of the measured object from these fringe images by multi-step phase-shifting and time-phase unwrapping [Figure 8].

Figure 7 In computer vision (e.g. image deblurring), the inverse problem is ill-conditioned because the forward physical model mapped from the parameter space to the image space is not ideal. A typical solution is to reformulate the original ill-conditioned problem as a well-conditioned optimization problem by adding some prior assumptions (smoothing) that aid regularization

Fig. 8 Optical metrology transforms a ill-conditioned inverse problem into a well-conditioned regression problem by actively controlling the image acquisition process. For example, in fringe projection profilometry, by acquiring additional phase-shifted fringe images of different frequencies, the absolute phase can be easily obtained by multi-frequency phase-shift method and temporal phase expansion method

However, when we step out of the laboratory and into the complex environment of the real world, the situation can be very different. The above active strategies often impose strict restrictions on the measurement conditions and the object being measured, such as:Stable measurement system, minimal environmental disturbance, static rigid objects, etcHowever, for many challenging applications, such as harsh operating environments and fast-moving objects, the above active strategy may become a "Luxury"Even impractical requirements. In this case, traditional optical metrology methods will face serious physical and technical limitations, such as limited data volume and uncertainty of forward models.How to extract high-precision absolute (unambiguous) phase information from minimal (preferably single-frame) fringe patterns remains one of the most challenging problems in optical metrology today.Therefore, we look forward to innovations and breakthroughs in the principles and methods of optical metrology, which are of great significance for its future development.

As a "data-driven" technology that has emerged in recent years, deep learning has received more and more attention in the field of optical metrology and has achieved fruitful results in recent years. Different from traditional physical model-driven methods,The deep learning method creates a set of training datasets composed of real target parameters and corresponding original measurement data, establishes their mapping relationships using ANN, and learns network parameters from the training dataset to solve the inverse problem in optical metrology[Figure 9]. Compared to traditional optical metrology techniques, deep learning moves active strategies from the actual measurement phase to the network training phase, gaining three unprecedented advantages:

1) From "model-driven" to "data-driven"Deep learning overturns the traditional "physical model-driven" approach and opens up a new paradigm based on "data-driven". Reconstruction algorithms (inverse mappings) can learn from experimental data without prior knowledge of physical models. If the training dataset is collected based on active strategies in a real experimental environment (including measurement systems, sample types, measurement environments, etc.), and the amount of data is sufficient (diversity), then the trained model should be able to reflect the real situation more accurately and comprehensively, so it usually produces more accurate reconstruction results than traditional physical model-based methods.

(2) From "divide and conquer" to "end-to-end learning":Deep learning allows for "end-to-end" learning of structures in which neural networks can learn a direct mapping relationship between raw image data and the desired sample parameters in one step, as shown in Figure 10, compared to traditional optical metrology methods of independently solving sequences of tasks. The "end-to-end" learning method has the advantage of synergy compared to "step-by-step divide-and-conquer" schemes: it is able to share information (features) between parts of the network performing different tasks, contributing to better overall performance compared to solving each task independently.

(3) From "solving linear inverse problems" to "directly learning pseudo-inverse maps": Deep learning uses complex neural networks and nonlinear activation functions to extract high-dimensional features of sample data, and directly learns a nonlinear pseudo-inverse mapping model ("reconstruction algorithm") that can fully describe the entire measurement process (from the original image to the physical quantity to be measured). For regularization functions or specified priors than traditional methods, the prior information learned by deep learning is statistically tailored to real experimental data, which in principle provides stronger and more reasonable regularization for solving inverse problems. Therefore, it bypasses the obstacles of solving nonlinear ill-conditioned inverse problems and can directly establish the pseudo-inverse mapping relationship between the input and the desired output.

Fig. 9 Optical metrology based on deep learning  

A) In deep learning-based optical metrology, the mapping of image space to parameter space is learned from a dataset by building a deep neural network b) The process of obtaining a training dataset through experimentation or simulation.

Figure 10 Comparison of deep learning and traditional algorithms in the field of fringe projection

A) The basic principle of fringe projection profiling is 3D reconstruction based on optical triangulation (left). Its steps generally include fringe projection, phase recovery, phase unwrapping, and phase-height mapping b) Deep learning-based fringe projection profiling is driven by a large amount of training data, and the trained network model can directly predict the encoded depth information from a single frame of fringes

Application of deep learning in optical metrology

A complete revolution in image processing algorithms

Due to the above advantages, deep learning has received more and more attention in optical metrology, bringing a subversive change to the concept of optical metrology technology. Deep learning abandons the strict reliance on traditional "forward physical models" and "reverse reconstruction algorithms", and reshapes the basic tasks of digital image processing in almost all optical metrology technologies in a "sample data-driven" way. Breaking the functional/performance boundaries of traditional optical metrology technologies, mining more essential information of scenes from very little raw image data, significantly improving information acquisition capabilities, and opening a new door for optical metrology technology.Figure 11 reviews typical research efforts using deep learning techniques in the field of optical metrology. Below are specific application cases of deep learning in optical metrology according to the image processing level of traditional optical metrology techniques.

Figure 11 Deep learning in optical metrology: Since deep learning has brought significant conceptual changes to optical metrology, the implementation of almost all tasks in optical metrology has been revolutionized by deep learning

(1) Image preprocessing:Early work on applying deep learning to optical metrology focused on image preprocessing tasks such as image denoising and image enhancement. Yan et al. constructed a CNN composed of 20 convolutional layers to achieve fringe image denoising [Fig. 12a]. Since noise-free ideal fringe images are difficult to obtain experimentally, they simulated a large number of fringe images with Gaussian noise added (network input) and corresponding noise-free data (true value) as training datasets for neural networks. Figures 12d-12e show the denoising results of traditional denoising methods – windowed Fourier transform (WFT 3) and deep learning methods. It can be seen from the results that the deep learning-based method overcomes the edge artifacts of traditional WFT and exhibits better denoising performance. Shi et al. proposed a deep learning-based method for fringe information enhancement [Fig. 13a]. They used the fringe images captured in real scenes and the corresponding quality-enhanced images (acquired by subtracting two fringe images with a phase shift of π) as a dataset to train neural networks to achieve a direct mapping between the fringe images to the quality-enhanced fringe information. Fig. 13b-Fig. 13d shows the results of the 3D reconstruction of the moving hand by the traditional Fourier transform (FT) 3 and deep learning methods. From this, it can be seen that the deep learning method is significantly better than the traditional method in imaging quality.

Figure 12 The denoising method of fringe image based on deep learning and the denoising results of different methods.

A) The process of fringe denoising using depth learning: the fringe image with noise is used as the input of neural networks to directly predict the denoised image b) the input noise image c) the true phase distribution d) the denoising result of deep learning e) the denoising result of WFT 3

Fig. 13 Fringe information enhancement method based on deep learning and 3D reconstruction results under different methods.

A) using depth learning for fringe information addition process: the original fringe image and the acquired quality enhancement image are used to train DNN to learn the mapping between the input fringe image and the output quality enhancement fringe information b) input fringe image c) conventional FT method 38 3D reconstruction results d) 3D reconstruction results of deep learning method

(2) Image analysis:Image analysis is the most core image processing link in optical metrology technology, so most deep learning techniques applied to optical metrology are for processing tasks related to image analysis. For phase measurement technology, deep learning has been widely explored in phase demodulation and phase unwrapping. Zuo et al. applied deep learning technology to fringe analysis for the first time, and effectively improved the three-dimensional measurement accuracy of FPP. The idea of this method is to use only one fringe image as input, and use CNN to simulate the phase demodulation process of the traditional phase shift method. As shown in Figure 14a, two convolutional neural networks (CNN1 and CNN 2) are constructed, where CNN1 is responsible for processing the fringe image from the input (IExtract background information (ACNN 2 then uses the extracted background image and the sinusoidal portion of the desired phase of the original input image generation.M) and the cosine part (D); Finally, the output sine-cosine result is substituted into the arctangent function to calculate the final phase distribution. Compared with the traditional single-frame phase demodulation methods (FT 3 and WFT 3), the deep learning-based method can extract phase information more accurately, especially for the surface of objects with rich details, and the phase accuracy can be improved by more than 50%. Only one input fringe image is used, but the overall measurement effect is close to the 12-step phase shift method [Fig. 14b]. This technology has been successfully applied to high-speed 3D imaging, achieving high-precision 3D surface shape measurement up to 20000Hz [Fig. 14c]. Zuo et al. further generalized deep learning from phase demodulation to phase unwrapping, and proposed a deep learning-based geometric phase unwrapping method for single-frame 3D topography measurement. As shown in Figure 15a, the stereo fringe image pairs and reference plane information captured under the multi-view geometric system are fed into the CNN to determine the fringe order. Figures 15b-15e show the 3D reconstruction results obtained by the traditional geometric phase unwrapping method and the deep learning method. These results show that the deep learning-based method can achieve phase unwrapping of dense fringe images in a larger measurement volume and more robustly under the premise of projecting only a single frame of fringe images.

Fig. 14 Fringe analysis method based on deep learning and three-dimensional reconstruction results under different methods 3 a) Fringe analysis method flow based on deep learning: First, the background image A is predicted from the single frame fringe image I by CNN1; then CNN2 is used to realize the fringe pattern I, The mapping between the background image A and the sinusoidal part M and the cosine part D that generate the desired phase; finally, the phase information can be wrapped with high accuracy through the tangent function b) Comparison of three-dimensional reconstruction of different phase demodulation methods (FT 3, WFT 3, deep learning-based method and 12-step phase shift method 3 3) c) Deep reconstruction results of a high-speed rotating table fan using depth learning method

Fig. 15 Geometric phase unwrapping method based on deep learning and 3D reconstruction results under different methods < unk > a) Flow of geometric phase unwrapping method assisted by deep learning: CNN1 predicts the wrapping phase information from the stereo fringe image pair, CNN2 predicts the fringe order from the stereo fringe image pair and reference information. The absolute phase can be recovered by the predicted wrapping phase and fringe order, and then 3D reconstruction is performed b) 3D reconstruction results obtained by combining phase shift method, three-camera geometric phase expansion technique, and adaptive depth constraint method, c) 3D reconstruction results obtained by combining phase shift method, two-camera geometric phase expansion technique, d) 3D reconstruction results obtained by geometric constraint method based on reference surface, e) 3D reconstruction results obtained by deep learning method

Deep learning is also widely used for stereo matching and achieves better performance than traditional subset matching and sub-pixel optimization methods. Zbontar and LeCun ?? propose a deep learning method for stereo image disparity estimation [Fig. 16]. They constructed a Siamese-type CNN to solve the matching cost calculation problem by learning similarity metrics from two image blocks. The output of the CNN is used to initialize the stereo matching cost, and then to achieve disparity map estimation by refining the initial cost through cross-based cost aggregation and semi-global matching. Fig. 16d-Fig. 16h are disparity images obtained by traditional Census transformation and deep learning methods. From this, it can be seen that the deep learning-based method achieves lower error rates and better prediction results. Pang et al. propose a cascaded CNN architecture for sub-pixel matching. As shown in Figure 17a, the initial disparity estimation is first predicted from the input stereo image pair by DispFulNet with upsampling module, and then the multi-scale residual signal is generated by the hourglass-structured DispResNet, which synthesizes the output of the two networks and finally obtains the disparity map with sub-pixel accuracy. Figures 17d-17g show the disparity map and error distribution predicted by DispfulNet and DispResNet. It can be seen from the experimental results that the quality of the disparity map has been significantly improved after the optimization of DispResNet in the second stage.

Figure 16 The disparity estimation results of the subset matching method based on deep learning and the disparity estimation results of different methods ?? a) The algorithm flow of disparity map estimation using depth learning: Siamese CNN is constructed to learn similarity metrics from two image blocks to solve the matching cost calculation problem, and finally realizes the disparity estimation through a series of post-processing b-c) The input stereo image d) true value e, g) Census and the disparity estimation results obtained by CNN

Figure 17 a) Sub-pixel matching method based on deep learning: First, the initial disparity estimation is predicted from the input stereo image pair through DispFulNet, and then the multi-scale residual signal is generated through the hourglass structure DispResNet, and the final output of the two networks is obtained. The disparity map with sub-pixel accuracy b) the left viewing angle of the input stereo image c) true value d-g) the disparity map and error distribution predicted by DispfulNet and DispResNet

(3) Post-processing: Deep learning also plays an important role in the post-processing phase of optical metrology (phase denoising, error compensation, digital refocus, phase-height mapping, etc.). As shown in Figure 18a, Montresor et al. input the sine and cosine components of the noise phase image into the CNN to predict the noise-removed high-quality phase image, and the predicted phase is fed back to the CNN for iterative refining to achieve better denoising effect. Figures 18b-18e show the phase denoising results of the traditional WFT 3 method and the deep learning method. Experimental results show that the CNN can achieve lower denoising performance than the WFT peak-valley phase error.

Figure 18 Phase denoising method based on deep learning and phase denoising results of different methods a) The process of phase denoising using depth learning: the sine and cosine components of the noise phase image are input to the CNN to predict the high-quality phase image with noise removal, and the predicted phase is fed back to the CNN again for iterative refining to achieve better denoising effect b) input noise phase image c) denoising result of WTF 3 d) denoising result of deep learning e) Comparison of WTF and deep learning method denoising results

Li et al. proposed a phase-height mapping method for fringe projection profilometry based on shallow BP neural networks. As shown in Figure 19a, the camera image coordinates and the corresponding projector image horizontal coordinates are used as network inputs to predict the three-dimensional information of the measured object. To obtain training data, the dot calibration plate is fixed on a high-precision displacement table and stripe images of the calibration plate are captured at different depth positions. By extracting the sub-pixel centers of the calibration plate dots, and using the absolute phase, the matching points of the camera and projector images corresponding to each marker center are calculated. Figures 19c and 19d show the error distribution of the three-dimensional surface shape results of the stepped standard parts obtained by the traditional phase height conversion method < unk > < unk > and the neural networks method. The results show that the neural networks-based method can learn more accurate phase height models from a large amount of data.

Fig. 19 a) Learning-based phase-depth mapping method: camera image coordinates and the horizontal coordinates of the corresponding projector image are used as network inputs to predict the three-dimensional information of the measured object b) The three-dimensional results of the step-shaped standard obtained by the learning-based method c, d) Error distribution of the three-dimensional surface shape results of the step-shaped standard obtained by the traditional phase height conversion method ?? and neural networks method e, f) Input phase images and output three-dimensional information of complex workpieces

Challenges and opportunities of deep learning in optical metrology

At present, deep learning has gradually "penetrated" into the discipline of computational imaging and optical measurement, and has shown amazing performance and strong application potential in fringe analysis, phase recovery, phase unfolding, etc. However, deep learning still faces many challenges in the field of optical metrology:

(1) As a data-driven technology, the performance of deep learning network output largely depends on a large number of labeled training data. The data collection process of most optical metrology experiments is complicated and time-consuming, and often the ideal true value cannot be obtained accurately and reliably after data collection [Figure 20].

Fig. 20 The challenge of deep learning in optical metrology – the high cost of acquiring and labeling training data. Taking fringe projection profilometry as an example, the multi-frequency time phase unwrapping method is used to obtain high-quality training data at the cost of projecting a large number of fringe images. However, in practice, hardware errors, ambient light interference, calibration errors and other factors make it difficult to obtain the ideal true value through traditional algorithms

(2) So far, there is still no theory that clearly explains what structure of neural networks is most suitable for specific imaging needs [Figure 21]?

(3) The success of deep learning usually depends on the "common" features learned and extracted from the training examples as prior information. Therefore, when artificial neural networks are faced with "rare examples", it is very easy to give a wrong prediction without realizing it.

(4) Unlike the traditional "transparent" deduction process based on physical model methods, most current deep learning-based decision-making processes are generally regarded as "black boxes" driven by training data. In optical metrology, interpretability is often crucial, as it ensures traceability of errors.

(5) Since information is not "created out of nothing", the results obtained by deep learning cannot always be accurate and reliable. This is often fatal for many application fields of optical measurement, such as reverse engineering, automatic control, defect detection, etc. In these cases, the accuracy, reliability, repeatability and traceability of the measurement results are the primary considerations.

Figure 21 The challenge of deep learning in optical metrology – empiricism in model design and algorithm selection. Taking phase extraction in fringe projection profilometry as an example, the same task can be achieved by different neural networks models with different strategies: The fringe image can be directly mapped to the corresponding phase map via DNN1; The numerator and denominator terms of the tangent function used to calculate the phase information can also be output from the fringe image and the corresponding background image via DNN2; The numerator and denominator can be predicted directly from the fringe image using a more powerful DNN

Although the above challenges have not been fully addressed, with the further development of computer science and artificial intelligence technology, it can be expected that deep learning will play an increasingly prominent role in optical metrology in the future through the following three aspects:

(1) The application of emerging technologies (such as adversarial learning, transfer learning, automated machine learning, etc.) to the field of optical metrology can promote the wide acceptance and recognition of deep learning in the field of optical metrology.

(2) Combining Bayesian statistics with deep neural networks to estimate and quantify the uncertainty of the estimate results, based on which it is possible to evaluate when neural networks produce unreliable predictions. This gives researchers another possible choice between "blind trust" and "blanket negation", namely "selective" adoption.

(3) The synergy between prior knowledge of image generation and physical models and data-driven models learned from experimental data can bring more expertise in optical metrology into deep learning frameworks, providing more efficient and "physically sound" solutions to specific optical metrology problems [Figure 22].

Figure 22 Introducing a physical model into deep learning can provide a more "reasonable" solution to a specific optical metrology problem. A) Directly predict the wrapped phase from the fringe image based on the end-to-end network structure (DNN1) b) It is difficult for the end-to-end strategy to accurately reproduce the 2π phase truncation, resulting in the loss function of the network not converging during training c) Incorporate the physical model of the traditional phase shift method into deep learning to predict the molecular and denominator terms of the tangent function used to calculate the phase information from the fringe image 39 d) The loss function of the deep learning network combined with the physical model can be stably converged during training

Summary and Outlook

There is no doubt that deep learning technology offers powerful and promising new solutions to many challenging problems in the field of optical metrology, and promotes the transformation of optical metrology from "physics and knowledge-based modeling" to "data-driven learning" paradigm. A large number of published literature results show that methods based on deep learning for specific problems can provide better performance than traditional knowledge-based or physical model methods, especially for many optical metrology tasks where physical models are complex and the amount of information available is limited.

But it has to be admitted that deep learning technology is still in the early stage of development in the field of optical measurement. A considerable number of researchers in this field are rigorous and rational. They are skeptical of the "black box" deep learning solutions that lack explainability at this stage, and are hesitant to see their applications in industrial testing and biomedicine. Should we accept deep learning as our "killer" solution to the problem, or reject such a "black box" solution? This is a highly controversial issue in the current optical metrology community.

From a positive perspective, the emergence of deep learning has brought new "vitality" to the "traditional" field of optical metrology. Its "comprehensive penetration" in the field of optical metrology also shows us the possibility of artificial intelligence technology bringing huge changes to the field of optical metrology. On the contrary, we should not overestimate the power of deep learning and regard it as a "master key" to solve every challenge encountered in the future development of optical metrology. In practice, we should rationally evaluate whether the large amount of data resources, computing resources, and time costs required to use deep learning for specific tasks are worth it. Especially for many applications that are not so "rigorous", when traditional physical model-based and "active policy" techniques can achieve better results with lower complexity and higher interpretability, we have the courage to say "no" to deep learning!

Will deep learning take over the role of traditional technology in optical metrology and play a disruptive role in the next few years? Obviously, no one can predict the future, but we can participate in it. Whether you are a "veteran" in the field of optical metrology who loves traditional technology, or a "newbie" who has not been involved in the field for a long time, we encourage you to take this "ride" – go and try deep learning boldly! Because it is really simple and often works!

Note: This article comes with a deep learning sample program for single-frame fringe analysis (Supplemental Material File #1) and its detailed step guide (Supplementary Information) to facilitate readers’ learning and understanding. For more details related to this article, please click https://www.nature.com/articles/s41377-022-00714-x Come and read the body of the 54-page paper.

Paper information

Zuo, C., Qian, J., Feng, S. et al. Deep learning in optical metrology: a review. Light Sci Appl 11, 39 (2022). 

https://doi.org/10.1038/s41377-022-00714-x

The first author of this article is Professor Zuo Chao of Nanjing University of Science and Technology, and PhD student Qian Jiaming of Nanjing University of Science and Technology is co-author. Co-authors include Associate Professor Feng Shijie of Nanjing University of Science and Technology, PhD student Yin Wei of Nanjing University of Science and Technology, PhD student Li Yixuan of Nanjing University of Science and Technology, PhD student Fan Pengfei of Queen Mary University of London, UK, Associate Professor Han Jing of Nanjing University of Science and Technology, Professor Qian Kemao of Nanyang University of Technology in Singapore, and Professor Chen Qian of Nanjing University of Science and Technology.

references

1. G?svik, K.J. Optical metrology. (John Wiley & Sons, 2003).

2. Creath, K. V phase-measurement interferometry techniques. In Progress in optics vol. 26 349 – 393 (Elsevier, 1988).

3. Aben, H. & Guillemet, C. Integrated photoelasticity in Photoelasticity of Glass 86 – 101 (Springer, 1993).

4. Gabor, D. A new microscopic principle. Nature 161, 777 – 778 (1948).

5. Schnars, U., Falldorf, C., Watson, J. & Jüptner, W. Digital holography in Digital Holography and Wavefront Sensing 39 – 68 (Springer, 2015).

6. Gorthi, S. S. & Rastogi, P. Fringe projection techniques: whence we are? Opts. Lasers Eng. 48, 133 – 140 (2010).

7. Pan, B., Qian, K., Xie, H. & Asundi, A. Two-dimensional digital image correlation for in-plane displacement and strain measurement: a review. Meas. Sci. Technol. 20, 062001 (2009).

8. Marr, D. & Poggio, T. A computational theory of human stereo vision. Proc. R. Soc. Lond. B Biol. Sci. 204, 301 – 328 (1979).

9. Pitas, I. Digital image processing algorithms and applications. (John Wiley & Sons, 2000).

10. Trusiak, M., Patorski, K. & Wielgi, M. Adaptive enhancement of optical fringe patterns by selective reconstruction using FABEMD algorithm and Hilbert spiral transform. Opts. Express 20, 23463 – 23479 (2012).

11. Awatsuji, Y. et al. Single-shot phase-shifting color digital holography. In LEOS 2007-IEEE Lasers and Electro-Optics Society Annual Meeting Conference Proceedings 84 – 85 (IEEE, 2007).

12. Fusiello, A., Trucco, E. & Verri, A. A compact algorithm for rectification of stereo pairs. Mach. Vis. Appl. 12, 16 ? 22 (2000).

13. Zuo, C. et al. Phase shifting algorithms for fringe projection profilometry: A review. Opts. Lasers Eng. 109, 23 – 59 (2018).

14. Zuo, C., Huang, L., Zhang, M., Chen, Q. & Asundi, A. Temporal phase unwrapping algorithms for fringe projection profilometry: A comparative review. Opt. Lasers Eng. 85, 84 -103 (2016).

15. Konolige, K. Small vision systems: Hardware and implementation. in Robotics research 203 – 212 (Springer, 1998).

16. Hong, C.K., Ryu, H.S. & Lim, H.C. Least-squares fittings of the phase map obtained in phase-shifting electronic speckle pattern interferometry. Opt. Lett. 20, 931 -933 (1995).

17. Zuo, C., Chen, Q., Qu, W. & Asundi, A. Phase aberration compensation in digital holographic microscopy based on principal component analysis. Opt. Lett. 38, 1724 – 1726 (2013).

18. Langehanenberg, P., Kemper, B., Dirksen, D. & Von Bally, G. Autofocusing in digital holographic phase contrast microscopy on pure phase objects for live cell imaging. Appl. Opt. 47, D176 – D182 (2008).

19.Wang, Y. & Zhang, S. Optimal fringe angle selection for digital fringe projection technique. Appl. Opt. 52, 7094 – 7098 (2013).

20. McCulloch, W. S. & Pitts, W. A logical calculus of the ideas imperceptible in nervous activity. Bull. Math. Biophys. 5, 115 – 133 (1943).

21. Rosenblatt, F. The perceptron: a probabilistic model for information storage and organization in the brain. Psychol. Rev. 65, 386 (1958).

22. Rumelhart, D.E., Hinton, G.E. & Williams, R.J. Learning representations by back-propagating errors. nature 323, 533 – 536 (1986).

23. LeCun, Y. et al. Backpropagation applied to handwritten zip code recognition. Neural Comput. 1, 541 – 551 (1989).

24. Hinton, G.E., Osindero, S. & Teh, Y.-W. A fast learning algorithm for deep belief nets. Neural Comput. 18, 1527 – 1554 (2006).

25. Krizhevsky, A., Sutskever, I. & Hinton, G. E. ImageNet classification with deep convolutional neural networks. Commun. ACM 60, 84 -90 (2017).

26. Nair, V. & Hinton, G.E. Rectified linear units improve restricted boltzmann machines.in ICML (2010).

27. Hinton, G.E., Srivastava, N., Krizhevsky, A., Sutskever, I. & Salakhutdinov, R.R. Improving neural networks by preventing co-adaptation of feature detectors. ArXiv Prepr. ArXiv12070580 (2012).

28. Chen, J.X. The evolution of computing: AlphaGo. Comput. Sci. Eng. 18, 4 ? 7 (2016).

29. LeCun, Y., Bengio, Y. & Hinton, G. Deep learning. Nature 521, 436 – 444 (2015).

30. Ouyang, W. et al. DeepID-Net: Object detection with deformable parts based convolutional neural networks. IEEE Trans. Pattern Anal. Mach. Intell. 39, 1320 – 1334 (2016).

31. Doulamis, N. & Voulodimos, A. FAST-MDL: Fast Adaptive Supervised Training of multi-layered deep learning models for consistent object tracking and classification. in 2016 IEEE International Conference on Imaging Systems and Techniques (IST) 318-323 (IEEE, 2016).

32. Dong, C., Loy, C.C., He, K. & Tang, X. Image super-resolution using deep convolutional networks. IEEE Trans. Pattern Anal. Mach. Intell. 38, 295 – 307 (2015).

33. Long, J., Shelhamer, E. & Darrell, T. Fully convolutional networks for semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition 3431 – 3440 (2015).

34. Ronneberger, O., Fischer, P. & Brox, T. U-net: Convolutional networks for biomedical image segmentation. In International Conference on Medical image computing and computer-assisted intervention 234 – 241 (Springer, 2015).

35. Yan, K. et al. Fringe pattern denoising based on deep learning. Op. Commun. 437, 148 – 152 (2019).

36. Shi, J., Zhu, X., Wang, H., Song, L. & Guo, Q. Label enhanced and patch based deep learning for phase retrieval from single frame fringe pattern in fringe projection 3D measurement. Opt. Express 27, 28929 (2019).

37. Kemao, Q. Windowed Fourier transform for fringe pattern analysis. Appl. Opt. 43, 2695 – 2702 (2004).

38. Takeda, M., Ina, H. & Kobayashi, S. Fourier-transform method of fringe-pattern analysis for computer-based topography and interferometry. JosA 72, 156 – 160 (1982).

39. Feng, S. et al. Fringe pattern analysis using deep learning. Adv. Photonics 1, 025001 (2019).

40. Feng, S., Zuo, C., Yin, W., Gu, G. & Chen, Q. Micro deep learning profilometry for high-speed 3D surface imaging. Opts. Lasers Eng. 121, 416 – 427 (2019).

41. Qian, J. et al. Deep-learning-enabled geometric constraints and phase unwrapping for single-shot absolute 3D shape measurement. APL Photonics 5, 046105 (2020).

42. Tao, T. et al. Real-time 3-D shape measurement with composite phase-shifting fringes and multi-view systems. Opt. Express 24, 20253 (2016).

43. An, Y., Hyun, J.-S. & Zhang, S. Pixel-wise absolute phase unwrapping using geometric constraints of structured light systems. Opt. Express 24, 18445 – 18459 (2016).

44. Tao, T. et al. High-speed real-time 3D shape measurement based on adaptive depth constraints. Opt. Express 26, 22440 (2018).

45. Z’bontar, J. & LeCun, Y. Stereo matching by training a convolutional neural network to compare image patches. 32.

46. Mei, X. et al. On building an accurate stereo matching system on graphics hardware. In 2011 IEEE International Conference on Computer Vision Workshops (ICCV Workshops) 467 – 474 (IEEE, 2011).

47. Pang, J., Sun, W., Ren, J. SJ., Yang, C. & Yan, Q. Cascade Residual Learning: A Two-Stage Convolutional Neural Network for Stereo Matching. in 2017 IEEE International Conference on Computer Vision Workshops (ICCVW) 878 – 886 (IEEE, 2017). do i: 10.1109/ICCVW.2017.108.

48. Montresor, S., Tahon, M., Laurent, A. & Picart, P. Computational de-noising based on deep learning for phase data in digital holographic interferometry. APL Photonics 5, 030802 (2020).

49.Li, Z., Shi, Y., Wang, C., Qin, D. & Huang, K. Complex object 3D measurement based on phase-shifting and a neural network. Opt. Commun. 282, 2699 – 2706 (2009).

50.Li, Z., Shi, Y., Wang, C. & Wang, Y. Accurate calibration method for a structured lighting system. Opt. Eng. 47, 053604 (2008).

51. Feng, S., Zuo, C., Hu, Y., Li, Y. & Chen, Q. Deep-learning-based fringe-pattern analysis with uncertainty estimation. Optica 8, 1507 – 1510 (2021).

Article reprint/business cooperation/research group submission, WeChat:447882024

Let’s read one article a day!Join >Light reading club

Industrial Consumer Finance: Value Symbiosis, Good and Far-reaching.

Recently, the policy on consumer finance has been continuous, and the development of the industry has attracted much attention. The tide is just the time to sail. As the first consumer finance company controlled by a joint-stock commercial bank in China, Industrial Consumer Finance has always adhered to the concept of the parent bank Industrial Bank, insisting on "lending with sincerity" and practicing Pratt & Whitney, rooting in Bamin and spreading its branches and leaves to all corners of the country.
Up to now, the company’s exhibition area has covered more than 50 major economically developed cities in China, with accumulated loans exceeding 330 billion and accumulated customers exceeding 21 million. The compliance foundation has been continuously built, the service quality and efficiency have been steadily improved, the digitalization level has advanced by leaps and bounds, and the corporate image and brand have been deeply rooted in the hearts of the people.
Consumer finance is for the people and promotes development from the perspective of the masses.
If you have a goal in your heart and a direction under your feet, the more you know, the more you will do. At present, China’s financial and economic structure has undergone tremendous changes, from the previous investment to the current consumption-driven. After vicissitudes and changes in the consumer market, consumer finance has emerged from the booming consumer market, and its position in economic development has become increasingly important. Following the Central Economic Work Conference, the General Administration of Financial Supervision once again stressed: "To promote the consolidation and enhancement of the economic recovery, we must accelerate the development of consumer finance."
To be broad and subtle. Small and micro groups are "capillaries" in economic development, so consumer finance should help the economy grow steadily, and in the final analysis, it should take meeting the people’s growing needs for a better life as the starting point and the end result. Xingye Consumer Finance has always been committed to improving the consumption power and quality of life of domestic residents and helping the development of the real economy. It adheres to inclusive finance’s initial intention, adheres to the service concept of "standing on credit and lending with sincerity", and features differentiated operations at the grassroots level. It cooperates with each other online and offline, making consumer finance deeper, more detailed and more refined, innovating consumer financial products and services in diversified consumption scenarios, and striving to make development "temperature" and people’s happiness "texture".
Taking serving the new citizens as an example, doing a good job in financial services for the new citizens is a necessary measure to satisfy people’s yearning for a better life and promote the common prosperity of all people. Under the guidance of the Action Plan of Xingye Consumer Finance Co., Ltd. on Serving the Financial Needs of "New Citizens", Xingye Consumer Finance focuses on the new citizen groups, based on the three product systems of "family consumption loan", "talent development plan" and "career development plan", fills the gap, strengthens the position and provides excellent service, and sends the mind to the "doorstep" and the warmth to the "heart". "New citizens generally have financial qualification defects such as insufficient collateral assets and lack of credit information. On the basis of adhering to offline pro-nuclear and pro-visit, we linked online big data portraits, accurately positioned demand, and always paid attention to the practical difficulties of new citizens. By covering diversified and universal consumer financial products such as life consumption, further education and employment, we provided them with pure credit, unsecured installment loan services and exclusive service programs to help new citizens start a better city life." Industrial consumer finance staff said.
A branch and a leaf always care about the situation, and show their responsibility with "value symbiosis"
Seek good and far-reaching, and both righteousness and benefit will last for a long time. Xingye Consumer Finance inherits the parent bank’s genes, and in combination with the requirements of the regulatory authorities, while striving for its own healthy, sustained and steady development, it takes the initiative to assume more social responsibilities, and pays attention to putting back social and humanistic care into the connotation of sustainable development of enterprises through every good deed within its power.
The "Xingcai Plan", which has been launched for three years, is the "sustainable development sample" of Xingye’s consumer finance exploration. The project gives priority to public welfare, and "let every aspiring young person go to school and study hard" is the initial intention of Xingcai Plan. In addition to providing loans for higher education, a total of 1.5 million yuan has been donated for the revitalization of rural talents, which is used to set up the "Xingcai Plan Scholarship and Scholarship". Among them, scholarships have been distributed to 195 college students with financial difficulties, of which rural registered students account for 68.6%; Scholarship funds are specially designed to reward educators who work in rural primary schools and make important contributions to rural revitalization.
In addition to scholarships, Xingye Consumer Finance also relies on the Xingcai Plan to unite with the government and schools to build a public welfare base for the inspirational growth of talents, give full play to the linkage role of the government, society and industry, make use of the existing resource advantages, pay attention to the shaping of students’ inspirational character and the cultivation of correct employment values, help college students become the human resources needed by the market and society, and further broaden and extend the rising channels of graduates, so that it is better to give them a sword than a stick.
Embroider Gankun at the head of the tide, and drive the engine with "digital intelligence technology"
In the era of digital intelligence, the business logic of consumer finance is undergoing fundamental changes, from the competition of capital and price to the competition of ecology and technology, and the role of science and technology is accelerating from supporting and ensuring to leading development. Under this background, Xingye Consumer Finance took the lead in formulating the information technology plan in the 14th Five-Year Plan, and according to the strategic goal of "building a financial technology weapon to enable the company to develop at a high speed and with high quality", it planned a blueprint for digital architecture, covering the whole life-cycle consumer credit service, optimizing financial supply by relying on digital tools, and fully empowering business development and management, and the picture of "Digital Xingye" slowly unfolded.
In promoting high-quality development, Industrial Consumer Finance has broken the geographical limitations of physical outlets with the help of digital technologies such as the Internet, big data and artificial intelligence, and its service radius and coverage have been greatly expanded; Build automated and standardized risk identification, anti-fraud, post-loan collection and operation systems, and gradually improve operational efficiency; Identify and quickly respond to customers’ personalized service needs, and realize the networking of the whole business process, the automation of loan approval decision-making, and the normalization of 7×24-hour uninterrupted service. In terms of consumer rights protection, Industrial Consumer Finance actively explored the cross-domain cooperation of "digital finance+smart justice", put into operation the blockchain electronic deposit system, realized the data docking between the court case handling and the financial business system, reduced the burden of consumers’ proof, effectively realized data protection and consumer rights protection, and built a more efficient post-loan disposal channel to effectively resolve financial risks. At present, through the blockchain electronic deposit system and the Quanzhou Intermediate People’s Court, the company has realized the electronic deposit of credit in the whole process and cycle, automatically generating and submitting litigation materials in batches with one button, and has handled more than a thousand cases cumulatively, forming a good demonstration effect.
With its strong scientific and technological innovation ability and high service level, Xingye Consumer Finance won the "Tianji Award of Outstanding Consumer Finance Company in 2023" and was selected as the case of "Outstanding Brand Value Consumer Finance Platform in 2023".
Take root down and grow up. In the future, Societe Generale’s consumer finance will actively integrate into and serve the new development pattern, and stride towards high-quality inclusive finance with a more high-spirited attitude.
Disclaimer: This article is for promotional information, and does not represent the views of this website. If there is any violation or you think that the content of this page infringes your rights, please contact us and we will deal with it in time.
Reporting/feedback

The launching ceremony of the movie "Lovely Home" was held grandly.

  Recently, the launching ceremony of the family inspirational film "Lovely Home", which has attracted much attention in China film industry, was successfully held in Xinchang, Zhejiang. This film is the second film written, directed and performed by director Chen Tianyi. Caifei He, winner of the Golden Rooster Award for Best Actress, Zhang Chenguang, a famous actress from Taiwan Province, Tian Niu, a Hong Kong actress, and Wang Wei, a young actress, joined in.

  Coincidentally, Chen Tianyi, Caifei He and Wang Weitong, both starring in this film, graduated from Zhejiang Art School, which makes people look forward to the wonderful cooperation of the three alumni in the same play.

  It is understood that the film is based on billiards, and the story focuses on elements such as affection, inspiration and family feelings. The film tells the touching story of Dele, a talented player who was abandoned by his mother and his father died young. After meeting He Xiaoyi, the daughter of a billiards coach, he experienced various practical obstacles and psychological obstacles, and finally changed from a used car dealer to a world billiards champion. The value of family affection and people’s courage and perseverance in the face of difficulties are further discussed.

  Director Chen Tianyi was born in Xinchang, Zhejiang, China. Since he was a child, he has shown a strong interest in art, and he has performed well in calligraphy, painting, writing and other disciplines. Director Chen Tianyi was admitted to the film and television dance major of Zhejiang Art School with excellent results. After graduation, he transferred to the Central Academy of Drama for further study, and systematically learned the knowledge related to film and television. He never gave up his dream of directing, and even tried to write plays after work. Finally, in 2016, his first play, Miss Petunia, was born, which was also a key step in his dream. In September of the same year, with limited funds, he successfully completed the film production with his own talent and team efforts. The film Miss Petunia was released in 2019. This work has won many awards at home and abroad, making him a high-profile film director.

  Teacher Caifei He, winner of the Golden Rooster Award for Best Actress, also came to the press conference. She said that she liked the script "Lovely Home" very much. The script was very cinematic and excellent. Taking billiards as the carrier, it shows the family story between the protagonist and his mother from a very novel perspective. At the same time, Mr. Caifei He thinks that no matter whether it is a big production or a small production, as long as the film story is good and well shot, it is a good work. I look forward to seeing this film very much.

  Teacher Zhang Chenguang said in an interview that when she first came to Xinchang, she felt that the local culture was profound, the mountains and rivers were beautiful, and the transportation was developed. Xinchang’s future tourism industry would be very good. Playing billiards coach for the first time is also very novel. The theme of billiards is very good. Now there are 100 million people playing billiards in China, and the audience is huge. He is full of confidence in this film and thinks that the future is promising.

  Teacher Tian Niu, who can’t come to the scene because she works overseas, made a special video to express her regret and give her best wishes. She also said that she was looking forward to participating in the film "Lovely Home" and hoped to perform with you as soon as possible.

  Wang Wei, the star of the recent hit drama "Taking Love as the Camp", plays the heroine Xiao Yi in this film. At the opening ceremony, she told reporters that this is a warm family realistic film, and she likes this role very much. She is looking forward to the wonderful billiards competition in the film, and revealed that there will be heavyweight billiards figures in the form of eggs.

  Director Chen Tianyi said that he can go back to his hometown Xinchang to shoot this film, and he can integrate the characters and stories of this film into the local customs and beautiful scenery of his hometown, which will have a resonance and strive to create a local cultural IP in Xinchang. When the reporter asked why the director wanted to play the role of Dele himself, the director admitted that he was an enthusiast of billiards and had high requirements for billiards scenes. It was difficult to find an actor who could play well, perform well and deeply understand the characters. At the same time, it was more convenient and accurate to control the performance of the characters’ personality details in artistic creation and expression. Director Chen Tianyi said that "lovely home" is just around the theme of home, so it is more meaningful to jointly create this film. Finally, director Chen Tianyi said that it is the foundation of a filmmaker to keep the original heart and establish correct values. This film’s deep emotional expression and positive theme will undoubtedly attract the attention and love of many audiences.

  This high-profile film can successfully hold the opening ceremony, thanks to the hard work and team cooperation of the whole crew. Lovely Home will be presented to the audience with a touching story and exquisite production. Let’s look forward to the successful release of this film and contribute to the progress and development of China film industry.

Weihai, No.1! Ma Honeycomb released the 2023 Tourism Big Data Report

Source: [Weihai News Network]

A few days ago, Ma Honeycomb released the 2023 Tourism Big Data Report, and Weihai was selected respectively.The popularity of "eco-travel" soared to the first place.8 th place in the popular city of "seaside island tour"19th place in the hottest city in 2023.. Besides,Haiyuan Park and NaxianghaiSelected respectively, the heat soared in 2023.1st and 8th places of "Seaside Island" scenic spot..

This article comes from [Weihai News Network] and only represents the author’s point of view. The national party media information public platform provides information dissemination and dissemination services.

ID:jrtt

Degea enjoys Premier League Golden Gloves Award! The reason for not renewing the contract is exposed, and it is necessary to wait for Tenghage to determine who to buy.

David Degea, a 32-year-old goalkeeper from Manchester United, won the Golden Glove Award in this season’s Premier League two rounds ahead of schedule. He kept a clean sheet against Bournemouth for the 17th time this season, and no other Premier League goalkeeper can catch up.

In fact, even if Degea concedes a goal in this game, it will not prevent him from enjoying the exclusive honor, because alysson, the closest competitor to Liverpool, was scored by Aston Villa at Anfield, which means that he has only one round and only 14 clean sheets. The winner of the Manchester United goalkeeper, who has scored 16 clean sheets before this round, has been determined in advance.

Manchester United only conceded 41 goals in 36 games this season, ranking fifth in the league’s defensive list, but some fans believe that two Manchester United teammates, lisandro Martinez and Vallanet, are the bigger reasons for the team’s not conceding goals. However, Degea also contributed. If it weren’t for his save, Manchester United wouldn’t have scored all three points against Bournemouth.

Degea himself agrees with the fans, stressing that the goalkeeper is not the only contributor to the Golden Gloves Award: "Der goldene Handschuh is not only about me, but our defense has been solid this season. Not only the defenders, but also the whole team played well. "Degea said.

Manchester United coach Tenhage said: "Don’t forget the whole team, because all our 11 players are involved in defense and have a good game plan. But in the end, you have to admit that we have a very mature goalkeeper who has solved many good opportunities for opponents, just like today, he performed well again. 」

After the game, Degea also encouraged his Manchester United teammates to continue their efforts and surpass Newcastle United to win the third place in the Premier League. "Bournemouth got some chances and you have to stay focused and get involved in the game all the time. We won the game, so I am satisfied with everyone, which is a great improvement compared with last season. He said, "We are on the right path, but the work is not finished yet. We still have to play two games at home. We will try to win, we only need one point, but we will try to win two games. Let’s see if we can finish third, not fourth. 」

Manchester United’s remaining two Premier League rounds are at home, and Degea added: "Playing at Old Trafford is always special. We have a chance to win the Champions League at home, so this is a perfect position and everyone must complete the task. 」

Degea has not yet decided to renew his contract, and Manchester United have offered a salary reduction contract, with the weekly salary reduced by nearly half compared with the current 375,000 pounds. The Spanish goalkeeper wants to stay at Manchester United, but no agreement has been reached yet. Tenghage supports Degea to stay, but players and agents are worried that he will not be able to secure the No.1 goalkeeper position next season.

In recent months, Manchester United have been linked with many goalkeepers, including David Laya of brentford, Jordan Pickford of Everton and diogo Costa of Porto. Therefore, Degea may wait until the club decides which famous player to bring in before deciding his future. If Manchester United introduces a substitute goalkeeper, he will naturally reduce his salary and stay in the team. If he buys a new starting goalkeeper, Dad Duck obviously doesn’t want to stay on the bench.

In addition, the Glazer family may sell Manchester United, which will also affect the players’ contract renewal. The negotiation of a new contract for English striker Rachford has also reached a deadlock.

Not afraid of the big Paris single-core operation, but afraid of the boss’s control variables.

I am not afraid of the single-core operation in Paris, but I am afraid that Boss Mei will control the variables. If it is an accident that the mother director was injured after missing a penalty in the opening stage, it is more like a silent demonstration that Messi led the team to complete the goal in the second half.

Remember the fans’ cynicism about Messi after losing in the last round in Paris? The team lost the game, but the fans pointed their finger at Messi. Some people doubted Messi’s ability to lead the team. Some even said that Messi had no desire to win after playing the World Cup and was completely flat in the club. However, in this game, without Mbappé, Messi played the level of single-core combat when playing in the World Cup. His excellent overall situation and superb passing repeatedly tore open the opponent’s defense line, and finally helped Barley win the game.

Boss Mei’s panting expression after the game shows that he didn’t lie flat as people say, but tried his best. No wonder some people say that they are not afraid of the single-core operation in Paris, but they are afraid that the coal boss may strengthen the control variables. Messi’s core position in the team is the key to winning in Greater Paris. Let Qiu Wang become Mbappé’s deputy, which will only destroy the last green glory of Boss Mei.