The Rare Behavior Event Detection within the Compressed Video Domain

• Main Authors: Chen-Yu Hong, 洪振宇
• Other Authors: Cheng-Chang Lien
• Format: Others
• Language: en_US
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/37484033488905642361

碩士 === 中華大學 === 資訊工程學系碩士班 === 93 === The human behaviors are frequently analyzed by using the uncompressed video data (raw data). However, the amount of uncompressed video data is too huge to be transmitted via the network with limited bandwidth. Hence, the rare behavior detection within the compressed video may not only extract the visual features directly from MPEG compressed video but also overcome the limited network bandwidth problem. In this paper, several visual features extracted from the compressed video, e.g., the motion vectors and color features, are applied to develop new action feature descriptor. Based on the action feature descriptor the human actions are detected and then the rare behaviors may be identified. The proposed rare behavior detection system consists of four characteristics. (1) Each GOP is regarded as the smallest processing unit in which the features of motion and color are extracted. Then, the method of recursive shortest spanning tree is applied to partition the region of human body and a new motion feature called object-based accumulative motion vector (OAMV) is generated to describe the human actions. (2) The polar histogram for the OAMV motion feature is used for each GOP to describe the human action characteristics. (3) The human actions are detected by using the Hidden Markov Models. (4) According to the state transition diagram of abandoned object event and faint event respectively, the rare events are identified.