High Efficiency Camera Hierarchy for Human Identification

• Main Authors: Shyang-Yuan Chen, 陳祥掾
• Other Authors: Ching-Wei Yeh
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/44591060478834899903

碩士 === 國立中正大學 === 電機工程所 === 98 === Most of today''s video surveillance system using a single tier system, it must maintain work state. So it is not efficient, anytime, anywhere in the consumption of power. In the application of wireless sensor networks (WSN) in the video surveillance system, and the image needed to be transmitted to the host sever at any time. The image must be compressed to save time or save bandwidth, and compression process must be through a complicated calculation and power consumption. To save power consumption of single tier system, this paper proposes two-tier system. The first tier (front-end system) is the use of low-resolution camera, and its role is to determine whether people pass or not. This paper presents a very low computational method to complete it. The second class (back-end system) is to use high-resolution camera. When the front-end system detects that people pass, then the camera of second tier will work and save images. Because we only save important image, so this system can save the back-end hard disk space.