Video stabilization algorithm based on Pearson correlation coefficient
Jing Liu, Yong Zhang, Qinjun Zhao · 2019
In the video image acquisition, the monitoring equipment is very easy to cause abnormal lens jitter due to external factors. This kind of jitter has a great influence on the background modeling effect. In view of the above situation, this paper proposes a video jitter correction algorithm based on Pearson correlation coefficient to remove abnormal lens shake. The initial frame of the video is removed from the top, bottom, left and right by 50 pixels, and the middle portion is taken as the reference frame. The reference frame and other frames of the video are projected horizontally and vertically. The Pearson correlation coefficient is used to describe the inter-frame correlation, and the other frames are found and the reference frame is found. The most relevant area, repairing video anomalies and getting a stable video.