Shadow Elimination and Moving Object Detection in the Video Surveillance
Due to the complexity of the external environment ,moving target detected in video surveillance tends to possess contour defects, as well as shadow and noise problems. It proposed a method to improve the profile, eliminate the shadow and noise. Firstly, it used five frame gradient image to determine the changed region of the moving object and then or the results of background differencing method to obtain a complete outline of the moving target It analyzed the characteristics of HSV in the color space. so as to eliminatethe shadow. And in accordance with the distribution characteristics of noise it removed the noise. Processed results show that in the processing speed of 20ms/frame ,it achieves real-time requirements. The moving target obtained eliminates the shadow and noise, and has strong robustness.
moving target detection gradient image five frame difference method the background difference method shadow elimination de-noising
Liang Feng-mei Du Yong-feng Zhang Qi-gui
College of Information Engineering Taiyuan University of Technology Taiyuan, China
国际会议
2010 International Conference on Future Information Technology(2010年未来信息技术国际会议 ICFIT 2010)
长沙
英文
13-16
2010-12-14(万方平台首次上网日期,不代表论文的发表时间)