Research On Target Detection And Tracking Algorithm Based On Video Surveillance Image

The continuous development of intelligent system and the arrival of large data age,computer vision achieve the continuous progress.Extracting the moving objects is the hotpot of the most important contents of computer vision.The effect of computer vision is to simulate biological visual function approximately by dealing with images or videos and achieves a certain degree of engineering tasks.It is widely used in intelligent monitoring,traffic control,machine intelligence,medical diagnosis and other fields by combining advanced technology of many computer fields such as image processing,pattern recognition,automatic control and artificial intelligence.In the aspect of intelligent monitoring,while the increasing demand of various complex environments,the focus of the current target tracking field is improving the robustness and accuracy of the moving target detection and track.Based on the commonly research,this paper selects Gaussian mixture model(GMM)and Kalman filter algorithm to detect and track the dynamic target.And to improve the effects of Gaussian mixture model algorithm further.The concrete work of this paper is as follow:1.In a video surveillance system,moving target detection is the foundation,which determines the performance of the whole system to a large extent.Because the target is moving and continuous,this requires the real-time performance of the algorithm to be good.When GMM model algorithm in dynamic target detection,which has the advantages of simple operation and high efficiency.It satisfies the real-time performance of video image and is also a better method of comprehensive performance.According to the characteristics of video surveillance system,GMM is selected as the dynamic target detection algorithm.2.GMM has the disadvantages of incomplete detection object and fuzzy feature edge in detecting dynamic targets.In order to improve the detection quality of dynamic targets and overcome the shortcomings of the algorithm,this paper introduces a new algorithm in image processing that is Graph Cuts.By fusing GMM and Graph Cuts,the accurate real-time detection of dynamic video frames is realized.Because the Gauss mixed model is adapted to the multi complex background,and the Graph cuts focus on the correlation between adjacent pixels.3.In order to get the position of moving target and duration characteristics,this paperselects Kalman filter algorithm as tracking algorithm to achieve tracking of dynamic targets accurately and prepares for subsequent target identification and analysis.