基于背景擦除的视频监控图像超分辨率重建  

Super Resolution Reconstruction Algorithm of Video Surveillance Image Based on Background Erasure

作　　者：俞文静[1] 张明军[1] 王影[1] 

机构地区：[1]广州大学华软软件学院,广州510990

出　　处：《计算机与数字工程》2016年第4期730-734,共5页Computer & Digital Engineering

基　　金：广东高校优秀青年创新人才培养计划项目(编号:2013LYM_0114);广州大学华软软件学院2014年教学研究;科学研究资助立项项目(编号:ky201410)资助

摘　　要：视频图像的超分辨率重建在视频监控领域有非常重要的应用,论文采用背景擦除的方法研究了一种应用于视频监控的视频图像超分辨率重建算法。首先建立了超分辨率重建算法的一般模型,并阐明了视频图像的超分辨率重建的过程,重点研究了视频监控图像重建的优化算法,该算法采用背景差分算法先对视频图像背景进行擦除提取运动目标,再使用改进DS算法对目标进行运动估算,最后使用改进的POCS对图像进行重建。实验表明该文算法与一般的POCS相比能明显地提高视频监控的重建图像的清晰度和效率。Super resolution reconstruction of video image has very important application in the field of video surveillance,this paper uses background erasure to research one of super-resolution reconstruction algorithm applications in video surveillance image.Firstly,the model of the super-resolution reconstruction algorithm is established,and the super-resolution reconstruction of the video image is clarified,and the optimized algorithm of the video surveillance image reconstruction is studied.In this algorithm,background subtraction is used to extract the moving target from video image,and the target image is estimated by using the improved DS（Diamond Search）algorithm,and then the improved POCS（Projection Onto Convex Sets）is used to reconstruct the image.Experiments show that the proposed algorithm compared with the general POCS can obviously improve the definition of the video surveillance images and its efficiency.

关 键 词：视频监控 超分辨率重构 凸集投影法 背景擦除 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构] TP391.9[自动化与计算机技术—计算机科学与技术]