Automatic Face Detection and Tracking Based on Adaboost with Camshift Algorithm

摘要：

With the development of information technology, video surveillance is widely used in security monitoring and identity recognition. For most of pure face tracking algorithms are hard to specify the initial location and scale of face automatically, this paper proposes a fast and robust method to detect and track face by combining adaboost with camshift algorithm. At first, the location and scale of face is specified by adaboost algorithm based on Haar-like features and it will be conveyed to the initia! search window automatically. Then, we apply camshift algorithm to track face. The experimental results based on OpenCV software yield good results, even in some special circumstances, such as light changing and face rapid movement. Besides, by drawing out the tracking trajectory of face movement, some abnormal behavior events can be analyzed. Keywords-Video surveillance; adaboost algorithm; camshift algorithm; OpenCV; face detection and tracking

作者: Hui Lin JianFeng Long

作者单位: College of Electrical and Information Engineering Hunan University Changsha, China

会议类型: 国际会议

会议名称: 2010 International Conference on Signal and Information Processing(2010年IEEE信号与信息处理国际会议 ICSIP2010)

会议地点: 长沙

会议语种:英文

页码: 118-122

在线出版日期: 2010-12-14（万方平台首次上网日期，不代表论文的发表时间）