Interpretation of Anson's image sensor strategy-long accumulation and comprehensive innovation

　　Interpretation of Anson's image sensor strategy-long accumulation and comprehensive innovation

　　As the main enabler of intelligent sensing technology, what is the layout of onsemi in image sensor? A few days ago, at the Shanghai Electronics Show in Munich, the author was fortunate to interview Tao Zhi, Marketing Manager of Industrial and Consumer Applications Greater China of Anson's Intelligent Perception Department, and took stock of some products and layouts of Anson's intelligent perception field.

　　Long history and accumulation

　　Since the acquisition, Anson is the oldest CMOS image sensor company in the industry, which has been more than 45 years.

　　There are obviously several main lines here, including Cypress, Truesense, Aptina, Sensl and their corresponding predecessors, all of which once had glory in the sensor industry. For example, Cypress's custom CMOS sensor is used in ARRI professional digital camera, and movies shot with ARRI camera, lens and lighting system have won many awards such as best film and best photography.

　　Needless to say, Kodak has the most patents and inventions in the field of image sensors. For example, Bayer array, one of the foundations of image sensor technology, was invented by Bryce Bayer, a Kodak scientist.

　　The same is true for Aptina. Eric Fossum, the inventor of CMOS image sensor, founded Photobit, the first CMOS commercialization company, which was acquired by Micron, which was later Aptina.

　　Through successive acquisitions, Anson has the largest patent portfolio in the industry, with more than 2,300 items, and also has the most comprehensive product portfolio in the industry. At present, in addition to the mobile phone market, Anson's products are all over the fields including industry, automobile, consumption and IoT, and it is the number one CMOS sensor supplier in the global industrial and automotive markets.

　　Always at the forefront of the times

　　"What Anson is good at is innovative work." Tao Zhi emphasized.

　　The latest innovative example of Anson is the AR8022 image sensor. The device has embedded high dynamic range (eHDR) function and optimized near infrared (NIR) response, which is very important for applications with poor lighting conditions, such as security monitoring, camcorders, doorbell cameras and robots. The low-power architecture and motion wake-up function of the sensor are designed to greatly reduce the system power consumption.

　　The main function of HDR technology is to improve image quality. By using HDR technology, the color contrast and brightness in the image can be enhanced, thus making the image more vivid, vivid and real.

　　At present, there are three main HDR technologies, including multiple exposures, large and small pixels and super exposures.

　　Firstly, aiming at multiple exposure technology, the details at low brightness are captured by long exposure, and the images at medium and high brightness are captured by medium and short exposure. In the case of conventional pixel design, the dynamic range of 120dB(3 exposures) or even 140dB(4 exposures) can be achieved by multiple exposures. However, the general multi-exposure technology has shortcomings: there will be motion artifacts for moving objects, especially for LED light sources, sometimes it can be captured by short exposure time, and sometimes it can't, which leads to the problem of unclear LED flicker. Typical pulsed light sources are LED traffic lights, traffic signs and automobile taillights.

　　The second is the large and small pixel technology. There are both large pixels and small pixels on the chip of this design. Large pixels capture low light information and small pixels capture high light information. Because large and small pixels are exposed at the same time, this design avoids the problem of motion artifacts. However, this technology has the loss of large pixel QE, and the color problem caused by the influence of different pixel QE and crosstalk, which requires users to spend a lot of resources and costs to correct, which brings great challenges to users. The technology of large and small pixels uses the ratio of light induced by large pixels to small pixels to realize HDR. At present, the semiconductor manufacturing technology does not allow arbitrary expansion or contraction of the physical size of pixels, so the dynamic range of large and small pixels technology is limited.

　　At present, under the pressure of cost and miniaturization, it is necessary to use smaller pixels to make chips with the same resolution smaller, and it is more difficult to reduce the pixel size in large and small pixel schemes than in single pixel schemes. Note: The large and small pixel technology is a patent of Ansenmei, which was acquired by Micron in 2009.

　　The third is super exposure, which Anson adopted in Hyperlux series. By adding an extra circuit, when the pixel capacitor is saturated, the overflowing charge will continue to be transferred to the external capacitor storage, thus achieving a super-large full well charge, with a dynamic range of 110dB for a single exposure and 150dB for two exposures. This method has no LED flicker problem and motion artifact problem, and also has very good low light performance.

　　EHDR technology is called embedded High Dynamic Range, which can be understood as HDR to increase processing function. Usually, the sensor with multiple exposures will send the data of each exposure to the back-end processor for synthesis, while three exposures need to send three frames of data. According to the calculation, if the traditional HDR data transmission method is adopted, 4K 3 exposures need to exceed 10Gbps, which is a great challenge to SERDES and cables.

　　The multi-exposure synthesis of eHDR is completed in the sensor. Through the internal intelligent fitting and compression function, the 20-bit data of three exposures can be compressed into 12-bit data, and then transmitted to the back-end platform, which greatly reduces the data bandwidth of the system. If the traditional HDR data transmission method is adopted, 4K 3 exposures need to exceed 10Gbps, which is a great challenge to SERDES and cables.

　　AR0822 adopts eHDR(embedded High Dynamic Range), and the multi-exposure synthesis is completed inside the sensor. Through the internal intelligent fitting and compression function, the 20-bit data of three exposures can be compressed into 12-bit data, and then transmitted to the back-end platform, which greatly reduces the data bandwidth of the system.

　　AR0822 supports a variety of linear fitting functions of multiple exposure synthesis, such as DLO(Digital Lateral Overflow) and SCMAX(Smooth Combination Max), which are called intelligent fitting. This mode reduces the noise in the critical area of brightness during multiple exposure synthesis.

　　In addition, there are motion artifacts and LED flicker problems in multiple exposures, and the eHDR of AR0822 also adds the function of Motion compensation to alleviate these problems.

　　What's more, in addition to eHDRTM, AR0822 also has WOM (Wake on Motion) function: AR0822 can run at a very low frame rate. When motion is detected in the ROI area of the picture, it will wake up the SOC and restore the sensor to normal mode.

　　Extensive platform product portfolio

　　Tao Zhi said that CMOS sensors are used in different fields, and their requirements for various characteristics are completely different. According to the broad categories, it is divided into human eye recognition and machine vision, and the products corresponding to the two applications are completely different. For machine vision, what is needed is an objective, accurate, undistorted and well-recognized imaging that serves the algorithm. However, to serve people's eyes, we need to consider color, image quality and so on, which is more subjective.

　　However, Tao Zhi also said that there is no complete boundary between these requirements. For example, the global shutter technology is mostly used in industrial applications, but it may also be needed in the process of human eye recognition.

　　For industrial applications, the requirements for sensors are also different, even for the same application, the different requirements of the external environment are also very different. "The CMOS market is very complex, and it needs to consider various factors including pixels, low illumination effect, external environment light intensity, noise control, cost, power consumption, temperature, vehicle regulations, intrusion detection and so on. This requires a deep understanding of products and markets, thus defining products with dual advantages of architecture and cost." Tao Zhi emphasized.

　　To this end, Anson's solution is to launch a series of products based on a certain technical architecture through a platform, and customers can make choices according to actual requirements.

　　A strong ecosystem

　　Due to the complexity and division of labor of optical system, it is necessary to build an ecosystem and cooperate closely with partners for debugging. At the Muzhan site, Anson presented a variety of programs jointly developed with partners.

　　In addition, in order to lower the development threshold, Anson America and the third party have developed reference design or modular products respectively, and combined with Anson America's high-efficiency SDK, thus accelerating customers' innovation in the field of intelligent perception.

　　Fab Right strategy

　　In the field of image sensors, Anson has always adopted the OEM method. Now, with the help of the company's new Fab Right strategy, a Fab for producing image sensors will be established.

　　Tao Zhi said that the shortage of semiconductors in the past few years has also affected CMOS image sensors, and now several friends are actively expanding their production capacity to cope with future demand growth. As Ansenmei, it will adopt the way of OEM+own Fabrika, and walk on two legs to better balance the return on capital investment, risk level, supply cycle and cost. In addition, after being out of stock a few years ago, the supply of goods has also become an important consideration for customers when selecting models.

　　summary

　　Image sensor is the most important and intuitive channel to communicate the physical world and the digital world. As AI begins to blossom and bear fruit in various fields, the demand for image sensors is also increasing.

　　As Tao Zhi has always emphasized, the selection of image sensors can't be divorced from practical applications. For this reason, Ansenmei uses its deep cultivation and accumulation in the field to achieve innovation in many aspects such as underlying technology, product definition and scheme integration, thus fully empowering intelligent sensing technology.