Region-based fusion method for surveillance video

In order to improve the visual experience of different regions within an image or a video frame, in this paper, we propose a novel region-based illumination fusion method for nighttime video enhancement, in which is combine daytime background with high illumination and nighttime video frames. The proposed method firstly is that the input nighttime video is divided into two cases: (1) static scene region, and (2) moving objects region. Secondly, for static scene region, we propose a region-based illumination fusion method for enhancing illumination. Thirdly, for moving objects region, these regions are divided into two part(dark part and high illumination part), we use conventional method to enhance dark part. Finally color image reconstruction. Experimental results show that our method can obtain more appealing perceptual quality than the conventional algorithms. The proposed method also shows the effectiveness. Copyright © 2013 Binary Information Press.