Research On Video Target Tracking Under Complicated Background

The so-called video moving target tracking technology,is the core in the field of computer calculation is applied to the intelligent video analysis and retrieval,in the field of intelligent video monitoring,precision has a wide range of applications.However,the complicated background conditions of target tracking in video often becomes difficult,therefore,video target tracking in practical application and theory there are many problems to be solved.In practical application,for example,the target rotation and scaling,dynamic factors such as background,video brightness changes,obscured will seriously affect the effect of target tracking.Therefore,study the problem of great help to promote the development of target tracking technology.In this paper,through the study of gray-scale compressed tracking algorithm and classical template matching tracking algorithm,the ant colony optimization algorithm is optimized,thus improve the tracking robustness,reduce the time complexity,under the background of the complicated target tracking method was studied,the main work has the following several aspects:(1)Focused on the issue that the tracking efficiency of the tracking algorithm based on compressed sensing theory is not high andthe tracking algorithm can't anti-occlusion easily,the compressive tracking based on ant colony optimization was proposed.Theant colony optimization algorithm incorporated into the compressiontracking,proposed use ant colony optimization instead of the method that every other pixel to select a candidate of targets,when the target is occluded,searching the hole image by the ant colony optimization.Experimental results on 20 video framesdatabase show that the proposed algorithm greatly improves the tracking efficiency,and there is a strong anti-target occlusion and deformation thus improving the ability to track the success rate.(2)Focused on the issue that the tracking algorithm based on template matching has low speed and low success rate.We propose d a template matching tracking algorithm based on ant colony optimization.The algorithm use the ant colony optimization as the search strategy of the candidate template.Firstly,Selectsome candidate templates in a setting scope and work out the best of individual and the best of global;Then work out the best candidate template through the ant colony optimization;Finally,update the target template self-adaption.After analysis and experiments,the computation of the template matching tracking algorithm based on ant colony optimization can produce a well real-time tracking the accuracy of tracking has been greatly improved.