An Automatic Multiple-Slope Integration Algorithm for CMOS Image Sensor Based on Image Brightness

Abstract

In deep space exploration, illumination of environment changes tempestuously and it happens often that the contrast of target and background is large. Under this circumstance, an automatic multiple-slope integration algorithm is proposed for CMOS sensor based on the distribution characteristics of image brightness. First, the image is divided into a background region, some transitional regions and a target region according to its features of gray scale’s distribution. The exposure slope and the proportion of multi-stage integral time length are obtained by calculating the percentage of number of pixels in total and average pixel brightness value of each region. Secondly, the weighted average brightness value of the whole image can be computed by assigning different weights to different regions. In the end, the total integral time length is obtained by comparing the weighted average brightness value with an expected brightness value to get an adjustment coefficient for integral time based on look up table. The algorithm extends dynamic range of CMOS sensor and enhances the image gray level. It can optimize the response curve of imaging system and improve the quality of image. Experiment results prove that this algorithm improves the dynamic range effectively and can be applied well not only in conventional light environment but also in space environment with greatly changing intensity of light.