An auto exposure control algorithm based on lane recognition for on-board camera
Tong Li, Song Yan, Tao Mei · 2015
In order to obtain the accurate exposure time in real-time for an on-board camera in different urban environments, the proposed algorithm divides captured image into 5×5 sub-areas, calculates each sub-area's average brightness value to get a histogram, analyzes the peak value distribution in histogram to determine what environments the automobile is in. According to the environment the automobile works in, the algorithm includes two modes: normal lit condition and high-contrast lit condition. It executes the appropriate exposure adjustment mechanism for two modes by analyzing their brightness distribution in different environments. To avoid the interference of other factors like the sky, the algorithm marks the road surface be the regions of interest in real time. Besides, the optimization goal is not mid-tone at all but the maximum difference between lane and background in the region of interest. The experimental results show that the algorithm can rapidly and stably switch exposure mode when the automobile is traveling in different road conditions, and it can get accurate exposure time in both modes fast and accurately.