基于压缩感知的交通监控视频目标检测算法

针对公路交通监控的需求增大与网络带宽资源有限的矛盾,本文提出一种贝叶斯压缩感知的目标检测算法.该算法采用小波基对信号进行稀疏,用部分哈达玛测量矩阵进行观测,实现视频的压缩,为了得到更为准确的前景,提出在部分时间均衡自适应背景模型下,将背景分割思想和小波树结构的贝叶斯压缩感知结合的重构算法完成目标检测.通过对多个场景监控视频的试验,验证了该方法的准确性和有效性,并对光线变化具有一定的鲁棒性和减少视频传输的成本.

Object Detection Algorithm of Traffic Video Monitoring Based on Compressed Sensing

To adapt the contradiction between the increasing demands of highway traffic network monitoring and the limited network bandwidth resources,an object detection algorithm is proposed based on Bayesian compressed sensing.Videos are sparse in a wavelet base,and the Hadamard measurement matrix is adopted to compress the videos.To get more accurate foreground,an adaptive background model of part time average mean is proposed.An object detection method is also proposed that combines background difference and Bayesian compressed sensing of wavelet tree structure.The proposed method can robustly detect the targets under changing light and reduce the price of video transmission.Through many experiments in video monitoring of multiple scenes,the results show the accuracy and effectiveness of the proposed method.