Adaptive ViBe Algorithm Based on Pearson Correlation Coefficient
Jing Liu, Yong Zhang, Qinjun Zhao · 2019
The ViBe algorithm is a foreground detection algorithm based on background update. Its effect is better than most mainstream background modeling methods. It is robust to video jitter to a certain extent, but it will still affect the background when the video jitter is too large. When video jitter occurs due to external factors, we propose an algorithm based on the Pearson coefficient, which is an improvement on the traditional ViBe algorithm. Firstly, the offset of the video is calculated, and then the ViBe modeling is changed according to the offset of the video. The center point, which changes the range of ViBe pixel set, changes the sensitivity of the model, and gives a better background modeling effect when the video is jittered.