Research On Improved Aglorithm Of Detection And Tracking Based On Three Frame Difference And Camshift

Intelligent monitoring system is a hot field of computer vision which has been widely used in the military, security and other fields. Moving target detection and tracking technology is the core of intelligent video surveillance system, and therefore the study of moving target detection and tracking technology of intelligent monitoring system performance and efficiency is of great significance. This thesis studies the moving target detection and tracking algorithms exist in some of the key issues and the corresponding algorithm has done a validation. The main researches in this thesis are as follows:First of all, This thesis introduces some classical moving target detection algorithms, and the according shortcomings, meanwhile, it also introduces some basic knowledge of image processing.Second, this thesis introduces inter-frame difference method, background subtraction and single Gaussian background model. Focus on comparison of two frame differential and three frame differential detection results, the experiments show that although the three finite difference method has been greatly improved in the test result than the two test results, but the detected movement of the target area is also far from complete, and there are a large number of empty existence. In this paper, the background subtraction method is added to the three frame differential method for testing. Compared through experiments of the original three differential methods and the improved algorithm detection, experiments show that the improved algorithm can extract the complete moving target, eliminating the phenomenon of empty.Third, in the moving target tracking part of this paper, Cam-shift algorithm was chosen to accomplish the task of tracking. To solve the problem that the effect of tracking fast moving target with Cam-shift algorithm was unsatisfactory, the traditional algorithm was improved by predicting the target location in next frame of video with Kalman filter. Improved Cam-shift algorithm got an iteration starting point which is closer to real location of target. Therefore, searched area was smaller, and iteration times were less, meanwhile, the tracking speed and tracking precision was improved, and reliability and robustness of object tracking system was improved too.At last, the results show that the algorithms in this dissertation are effective, which can be adopted in actual scenes to detect and track moving object.