基于视频的运动车辆阴影检测与去除-苏州职业大学学报.pdfVIP

第2 1卷 第3期 苏州市职业大学学报 Vol .2 1,No .3 2010年9月 Journal of Suzhou Vocational University Sep . ,2010 基于视频的运动车辆阴影检测与去除 1 2 王 磊 , 胡 霞 (1.苏州市职业大学 计算机工程系，江苏 苏州  215104；2.苏州工业职业技术学院 信息工程系，江苏 苏州  215104) 摘 要：对交通目标监控系统中运动车辆阴影的检测与去除的关键步骤进行研究,利用车辆阴影 所固有的特征进行阴影抑制,针对车辆颜色与背景颜色相近时分割容易丢失目标车辆部分像素点 的问题,提出了在提取的过程中将图像阈值分割和边缘检测相结合的算法,较好地把目标车辆完 整的提取出来,便于车辆的准确定位. 关键词：视频图像；阴影检测；阴影去除；运动车辆 中图分类号：TP391 文献标志码：A 文章编号：1008-5475(2010)03-0043-04 Shadow Detection and Elimination of Moving Vehicles Based on Video Processing 1 2 , WANG Lei HU Xia (1.Department of Computer Engineering, Suzhou Vocational University, Suzhou 215104, China; 2.Department of Information Engineering, Suzhou Institute of Industrial Technology, Suzhou 215104, China) Abstract: This paper investigates the crucial steps of the shadow detection and elimination of moving vehicles in the traffic monitoring system. The essential characteristics of vehicle shadows are used to carry out shadow repression. In addition, to solve the problem of partial loss of target vehicle pixels in segmentation when the vehicle color is similar to the background color, a new algorithm integrating the shreshold-based segmentation and edge detection is presented, It will effectively extract full target vehicles and accurately position them. Key words: video image; shadow detection; shadow elimination; moving vehicle 智能交通系统在缓解交通拥挤、提高交通效率等方面有着显著的作用 ,而基于视频的交通 目标监控 则是智能交通系统中的重要组成部分 .视频图像 中的运动车辆往往含有影子 .影子受 日照和灯光等因素 的影响