Research and Restoration Technology of Video Motion Target Detection Based on Kernel Method

Abstract

Abstract In recent years, due to the video surveillance applications more and more widely, people are not satisfied with the goal of monitoring, and the video monitoring technology of intelligent video moving object detection and tracking technology has received extensive attention. The research work in this paper is in the field, the moving target detection spatiotemporal correlation and difference contour tracking algorithm based on a fixed background. The algorithm in the background under the condition of fixed to pay a smaller time complexity, the target detection and tracking has a good effect, so it has higher application value. In this paper, the prospect of caused motion detection of occlusion background foreground correlation problem, put forward the video moving object detection method based on kernel independent component analysis, canonical correlation to minimize the component in the high dimensional feature space in order to separate the foreground nuclear background. Independent component analysis assumes that the foreground and background independent, avoid the correlation problem. The two objective functions based on Kernel Independent Component Analysis: analysis of kernel independent component analysis based on kernel canonical component (KCCA) and kernel independent component analysis (KGV) based on the generalized variance. KCCA is the application of canonical correlation analysis in the kernel method, discuss is the first canonical correlation separation component of high dimensional map, and KGV are typical correlation between the components in the high dimensional space of the whole spectrum. Both KCCA and KGV improved the accuracy of motion detection.