Development trend of sensor modules

发表于2019-11-27

  1. Intelligent image sensor consists of image sensor and visual software that can capture and analysis of visual information, instead of the human eye to all sorts of equipment measurement and judgment, the application components - camera at present has been widely used in all kinds of such as mobile phones, computers and consumer electronics products, such as wearable devices, as the future of the ADAS system widely popular and driverless cars, feeling will get on-board camera field.


    Compared with cameras, 3D imaging of lidar is more accurate, which is the first choice of vision system of driverless cars and will become the hot spot of capital market. We believe that in-vehicle, unmanned aerial vehicle and AR/VR intelligent image sensors will be the new growth areas of demand in the next five years, and predict that the market size of in-vehicle cameras will grow from $5.8 billion in 2016 to $21.45 billion in 2020, with a compound annual growth rate of 38.6%. Conservatively, the automotive lidar market will grow from $600 million in 2016 to $8 billion in 2025, with a compound annual growth rate of 33%. Drone cameras will grow from $12 million in 2016 to $100 million in 2020, a compound annual growth rate of 35.4 percent. Because of the high cost of lidar, most imaging technologies are mainly used in cameras. However, with the reduction of the cost of lidar in the future, its role as a replacement camera in various fields will be more prominent.


    MEMS sensors with micron characteristics are gradually replacing the traditional mechanical sensors. Preliminary estimates indicate that MEMS smart sensors will grow from $11.5 billion in 2015 to $20 billion in 2021. The application of camera technology is relatively mature. 3D imaging, iris recognition and gesture recognition are the main trends of technology development. Lidar is expensive and not yet commercially available. In the future, the integrated method of eliminating the mechanical rotating structure to reduce the cost will be the breakthrough of the future technology.


    In the field of artificial intelligence, specialization and integration will become the development trend of sensor modules in the future. The core of specialization lies in the matching of algorithms and functions. The integration of different types of sensors enables them to complement each other in functions, thus promoting advantages and avoiding disadvantages. At present, foreign countries have monopolized the advanced algorithm, and the integration mode will become the main trend of China's intelligent image sensor market in the next 3-5 years. The development of mobile phone and PC industry has been quite mature, and AR and VR are the market hotspots. With the emphasis and promotion of ADAS system, intelligent image sensor will become a new growth point of the industry. In addition, drones, connected cars and smart cities will also be the future wind of the industry.


    Finally, from an algorithmic point of view, embedded technology is more targeted and has advantages in solving local problems. Deep learning in the field of artificial intelligence will become the mainstream algorithm in the industry, and the combination of big data and high-speed end-to-end transmission will promote the practical application of deep learning algorithm. We believe that vehicles, uavs and AR/VR intelligent image sensors will be the new demand points in the next five years. The camera market is estimated to reach $22.3 billion by 2020, with a compound annual growth rate of 37% from 2016 to 2020. The automotive lidar market is expected to reach $8 billion by 2025, with a compound annual growth rate of 70 percent. From 2016 to 2025, this number will increase by 33%. In addition, as the cost of lidar continues to fall, the role of lidar in replacing cameras will be highlighted.