An Repairing Algorithm for Infrared Video Brightness Mutation Based on Information Entropy

Images with abnormal brightness changes in videos pose a serious obstacle to video analysis tasks. In this paper, an information entropy-based brightness variation repair algorithm for infrared video images is proposed. In this paper, image average pixel value and image information entropy are used to find image frames with abnormal brightness changes. The adaptive Gamma transform determined by the information entropy value is used to repair the video image with brightness variation, and the "dead zone" pixels are replaced with the image pixels at the moment before the brightness change. The experimental results show that the algorithm can effectively repair the sudden change of video brightness caused by moving objects approaching the camera, and provides more reliable data for subsequent video analysis tasks.