摘要
运动目标的实时检测与跟踪是智能监控和视频活动识别应用的基本步骤.运动目标检测把场景分割为前景目标和背景区域,但是在这个过程中运动目标投射的阴影很容易被误分类为前景目标,这种误分类会造成多个目标的合并或目标形状的改变.为了改善运动目标分割的效果,提出一个基于光强、色度和反射率的实时阴影检测和消除的算法,该算法不需要目标的特征、场景的照明条件等先验知识.仿真结果表明该算法比其它方法有更好的表现.
Real-time detection and track of moving objects are the fundamental steps of the smart surveillance and visual event recognition applications. Moving object detection segments the scene into foreground and background regions, but moving cast shadows can easily be misclassified as foreground in this process. This misclassification may lead to drastic changes in the shapes of objects or merging of multiple objects. To improve the performance of moving object detection, in this paper, we present an algorithm based on intensity, chromaticity, and reflectance ratio to detect moving cast shadows. No a prior knowledge about scene illumination and object characteristics are required. Obtained results show a significant improvement of performance compared to other works.
出处
《小型微型计算机系统》
CSCD
北大核心
2011年第2期361-364,共4页
Journal of Chinese Computer Systems
基金
国家"八六三"高技术研究发展计划基金项目(2007AA0955)资助
关键词
运动目标检测
阴影检测
阴影消除
反射率
moving objects detection
shadow detection
shadow removal
reflectance ratio