Biometric-based Intelligent Video Surveillance Technology

Surveillance cameras frequently used derivative the intelligent video surveillance system.The intelligent video surveillance system can identify a series of meaningful units fromstreaming video. The method of face detection is normally used in the automatic monitoring andcontrol system to detect and track specific targets. Face detection belongs to theresource-intensive detection. Some soft biological characteristics are used in the intelligent videosurveillance system, such as the height and the color and so on.For a monitoring system, the influence of the shadow and non-character objects in theexternal environment makes identifying and tracking a particular character become a challenge.In the research work of this paper, we use a contour tracing method. In the reference of thecamera calibration height, a character in the shooting screen can be divided into three parts,namely the head, body and legs. The color of each part can be estimated from the average valueof all the pixels’ value of the R, G and B of this part. Therefore, we can get four softbiological indicators of human recognition factors. They are the height of the body, head, bodycolor, and the color of the legs.Using the proposed biological recognition factors to detect and track the specific characterin the streaming video. The experimental results show that the method is effective. In the bigbackground detection and track of the little people, and in the occasion of requiring only coarseposition of the character in the picture, the biological recognition factor method in this paperproved to be useful.