Adaptive Video Background Estimation and Moving Object Detection Using QX-VSS LMS Algorithm

摘要：

In this paper, we propose a novel adaptive background estimation system using QX-VSS LMS algorithm. Especially, we focus on the challenges of illumination changes and small motions of background, shadows of objects and presence of new objects in background scene. We design a simple state machine with two-states that model the background of video. The state machine utilizes the QX-VSS LMS algorithm to detect the moving objects adaptively. Also, we use a modification of NCC algorithm to detect and eliminate the shadows of moving objects. Finally, we report empirical evaluations of the proposed on background in the different traffic datasets.

关键词： Background estimation Object detection shadow elimination QX-VSS LMS algorithm.

作者: Mostafa Mahi Ali Amiri

作者单位: Payam Noor University-Computer Branch Tehran, Iran Iran University of Sciene & Technologh Tehran, Iran

会议类型: 国际会议

会议名称: 2011 3rd International Conference on Computer and Automation Engineering(ICCAE 2011)(2011年第三届IEEE计算机与自动化工程国际会议)

会议地点: 重庆

会议语种:英文

页码: 410-414

在线出版日期: 2011-01-21（万方平台首次上网日期，不代表论文的发表时间）