论文部分内容阅读
针对监控视频在时间上存在冗余的问题,对ViBe(visual background extractor)算法进行改进,解决了ViBe算法存在噪声和易引入鬼影的问题,通过改进后的算法对视频进行背景建模,并对得到的背景掩模提取外轮廓以确定视频帧中是否存在前景对象。将存在前景对象的视频帧写入视频流中,达到视频浓缩的目的。经过试验验证,该方法可以有效地减少视频中的冗余信息,减小视频的体积,视频中的重要信息同时也得到了完整保留,满足实时性要求。
To solve the problem of video redundancy in time, this paper improves ViBe (visual background extractor) algorithm to solve the problem of noise and easy introduction of ghost in ViBe algorithm. The improved algorithm is used to background the video and The resulting background mask is extracted from the outer contour to determine whether there is a foreground object in the video frame. The foreground object video frame will be written into the video stream, to achieve the purpose of video condensation. After experiment, this method can effectively reduce the redundant information in the video, reduce the volume of the video, and the important information in the video is also completely preserved to meet the real-time requirements.