基于矩阵补全的自适应去雨雪算法  

Adaptive deraining and desnowing algorithm based on matrix completion

作　　者：田文英 黄成泉 冉龙才 Tian Wenying;Huang Chengquan;Ran Longcai(School of Data Science&Information Engineering,Guizhou Minzu University,Guiyang 550025,China;Engineering Training Center,Guizhou Minzu University,Guiyang 550025,China)

机构地区：[1]贵州民族大学数据科学与信息工程学院,贵阳550025 [2]贵州民族大学工程技术人才实践训练中心,贵阳550025

出　　处：《计算机应用研究》2020年第5期1570-1573,共4页Application Research of Computers

基　　金：国家自然科学基金资助项目(61762020,61802082);贵州民族大学科研基金资助项目(2017YB073)。

摘　　要：传统去雨雪方法通常没有考虑参数的自适应问题。为了提高视频去雨雪的效果,在Kim方法的矩阵补全模型中添加了一个自适应参数并提出了基于矩阵补全的自适应去雨雪算法。首先,简要地描述Kim方法的主要工作;其次,把自适应参数添加到经典的Kim模型的第二项;最后,使用各种雨雪视频验证了该参数的有效性及优异性,并使用网格搜索法找到去雨效果最好的参数。实验结果表明,添加的自适应参数能够有效地去除视频中的雨雪。The traditional methods of deraining and desnowing usually have not considered the parameter adaptive problem.In order to improve the effect of video deraining and desnowing,this paper added an adaptive parameter to the matrix completion model of Kim method,and proposed an adaptive deraining and desnowing algorithm based on matrix completion.Firstly,this paper introduced the main work of Kim method.Secondly,the proposed algorithm added the adaptive parameter to the second term of the classic Kim model.Finally,this paper used various rain and snow videos to verify the validity and superiority of the parameter,and applied the grid-search method to find the parameter with the best rain removal effect.Experimental results show that the added adaptive parameter can effectively remove rain and snow from the videos.

关 键 词：视频去雨雪 雨线去除 自适应参数 稀疏表示 矩阵补全 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]