视频图像中雪花的检测和去除.pdfVIP

光 电 子 ·激 光 第 18 卷 第 4 期 　2007 年 4 月 　　　　　 Vol . 18 No . 4 　Ap r . 2007 J our n a l o f Op t oelect r onics ·L aser 　　　　　　　 视频图像中雪花的检测和去除 许　亮 , 贾振红 , 覃锡忠 (新疆大学信息科学与工程学院,新疆 乌鲁木齐 830046) 摘要 :以影响视频图像的自然界降雪雪花为研究对象。考虑到雪花在空间中随机分布和具有一定降落速度 ,并 且目标或相机运动会给图像带来类似雪花运动的频率 ,用一种相关模型来捕获雪花的运动 ,用一种基于物理学 的光度测定模型 ,从亮度角度描述雪花对图像的污染。基于这些模型 ,提出了一种检测和去除视频图像中雪花 ( ) 的DRS detection and removal of snow 算法 ,并通过实验得到了验证。 关键词 :图像处理; 视频处理; 不良天气条件; 雪花 ( ) 中图分类号: TP391. 41 　　文献标识码 :A 　　文章编号 2007 Detection and Removal of Snow From Videos XU Liang , J IA Zhenhong , QIN Xizhong (College of Information Science and Engineering ,Xinjiang University ,Urumqi 830046 ,China) Abstract :The snowflake that appear s in videos is studied. The snowflake is randomly distributed in images and it has a cer tain velocity . The temporal frequencies due to obj ect and camera motions are similar to those produced by snow . A correlation model is used to capture t he dynamics of snow . A physicsbased motion blur model is used to explain the photometry of ( ) snow .Based on these models ,we develop an efficient algorithm DRS to detect and remove snowflake from videos. The ef fectiveness of our algorithm is demonstrated by experiment s. The algorithm described in this paper can be used in video sur veillance. vision based navigation ,tracking ,license plate recognition and video/