Video foreground and background separation method based on improved low-rank sparse decomposition

Abstract

The invention discloses a video foreground and background separation method based on improved low-rank sparse decomposition (LRSD). In observed video data, backgrounds between frames have strong correlation and can be regarded as a low-rank matrix, and a foreground target presents a motion form different from that of the backgrounds, can be regarded as an abnormal point in the low-rank matrix andgenerally only occupies a small part of the whole background, and conforms to the sparse characteristic. Accordingly, in LRSD, it is considered that video data is composed of a background having a lowrank characteristic and a foreground having a sparse characteristic. According to the method, a generalized nuclear norm and Laplace scale mixture is adopted to construct a low-rank sparse decomposition model, and then an alternating direction multiplier method is adopted to solve the model to obtain a low-rank matrix and a sparse matrix, so that foreground and background separation of the videois completed. According to the method, the problem of inaccurate approximate expression of the rank function and the sparsity function in the existing low-rank sparse decomposition method is solved, and the performance of the video foreground and background separation method based on low-rank sparse decomposition is improved.

Description

technical field [0001] The invention belongs to the field of computer vision, in particular to a video foreground and background separation method based on improved low-rank sparse decomposition. Background technique [0002] Video foreground and background separation is one of the most important and challenging tasks in the field of computer vision, which aims to preserve the background model and separate the foreground and background of the current video. This is one of the key steps in video processing, because clean foreground and background are very important for many tasks, and their performance directly affects subsequent processing such as object tracking, object recognition, etc. In practical applications, the background is inevitably affected by environmental factors such as noise, lighting changes, and camera shake. Even in environments such as water waves and swaying trees, the background itself is dynamic. Therefore, video front-background separation has always ...

AI Technical Summary

Problems solved by technology

Therefore, the accuracy of the foreground and background separation of these methods is not high. In addition, the influence of the regularization parameter is very large, and this parameter is difficult to adjust

These all affect the accuracy of video foreground and background separation to a certain extent.

Images