Proceedings Article | 18 December 2019
KEYWORDS: Night vision, Cameras, Imaging systems, Visible radiation, Color imaging, Image processing, Semiconductors, Photodiodes, Panoramic photography, CMOS sensors
Research of starlight-level wide-spectrum full-color imaging technology is aimed at the current demand of obtaining the real color image in the low ambient light. because traditional visible light imaging, laser imaging, thermal imaging and other technical methods can only obtain the target grayscale imaging information, but not obtain the target true color information. According to statistics, color images contain 30 times more information than grayscale images, and spectral information is extremely critical in target recognition. This paper intends to adopt an imaging method that integrates super large pixels, ultra-thin process microlens collection, and large-diameter light transmission to increase the sensitivity of the device by 50~100 times. The technical difficulty at this stage is based on large pixel, large area array, high pass light rate, low noise imaging device preparation technology. The key to achieving full-color imaging is the fabrication of large-pixel, large-area imaging devices and high-transmission, low-noise process technology. On this basis, we also designed deep learning image processing algorithms, conducted color brightening and enhancement of devices’ physical true color, and increased the signal-to-background ratio of images, which laid a technological foundation for the follow-up target detection and identification, and industrialized application. lastly, the minimum illumination of color night vision is 0.0001Lux@25Hz, F1.0, thereby achieving wide-spectrum full-color imaging.
