The Research On Detection Of Target Invasion Algorithm In Video Sequences

The detection of target invasion has been an increasingly important research direction in video analysis area and has been investigated extensively by researchers in the world, which is the important high-tech measures and technology to guarantee the public security in modern society. The core includes two parts: one part is moving object detection and classification, and the other is moving object tracking. And to improve the the real-time and accuracy of target invasion detection algorithm for the successful completion of the target detection and higher levels have a very important significance and the purpose of thesis is mainly concentrated on this point.In this thesis, it have made computer can discover the abnormity by processing and analyzing the video sequences, send the alarm and useful information without human action.In this paper, we place the emphasis on invasion target detection and tracking based on background models. The main contributions of this thesis include the followings:(1) This paper studied traditional moving target detection algorithm,and proposed an improved object tdetection algorithm.A new motion detection and background updating method was developed based on the background differencing using Otsu dynamic threshold and three frame differencing. Firstly a background model is based on improved Surendra algorithm, then the binary image is gained using background differencing based on Otsu dynamic threshold, and then background differencing is combined with the three frame differencing to gain more reliable moving target region and update background real-timely. The results of experiment show that the method is accurate and real-time.(2) Extraction characteristics and informations, define the criteria to the detected moving target feature according to the combining method of target point presentation and the rectangular box, with the multi-feature tree classifier, making the classification based on video sequence more clear in order to distinguish people and vehicles. For object tracking, we have chosen an object firstly and used color histogram as the feture to track it based on Camshif. The experimental results show that the algorithm can reliably predict the movement and track the target tray ectory.(3) A demo system of Intrusion Detection System based on VC++ and OpenCV have been developed. the platform can complete the target invasion detection of video sequences , and solve the phenomenon of omission due to static objects when access to control area and distortion due to background update delay when one object leaves.