Moving Object Detection in Video Sequences

摘要：

This paper discussed a popular method for moving object detection in stationary situation,and then proposed a flew method to decide whether the pixel is matched.By considering the pixel values from lots of frames as stochastic process.we used adaptive mixture of Gaussians to modeI the background.and then used the idea of maximizing likelihood to determine whether the pixel is matched.Compared with the popular matching method.there are more supporting evidences in mathematics.Finally the experimental results show that the new matching method is also effective.

关键词： moving object detection Gaussian mixture model maximizing likelihood pixel matching

作者: Xiaofeng Gu Hui Yao Yan Fu Ping Kuang Shixin Sun

会议类型: 国际会议

会议名称: The International Conference Information Computing and Automation(2007国际信息计算与自动化会议)

会议地点: 成都

会议语种:英文

页码: 810-812

在线出版日期: 2007-12-19（万方平台首次上网日期，不代表论文的发表时间）