Research On Target Tracking Algorithm In Complex Scenes

Precision guided weapon is a high-tech weapon developed on the basis of automatic control technology,involving microelectronics control technology,computer technology,photoelectric conversion technology etc.It has the advantages of long range,high speed,high precision,high power and so on.With the continuous improvement of the technical level of weapons and equipment,the role and status of precision-guided weapons in modern high-tech warfare are also continuously improving.Based on the image guidance technology in precision-guided weapons,this paper makes an in-depth study on the tracking of moving targets in the image guidance system.The specific work is as follows:Firstly,in order to solve the problems of low utilization rate of target background information and weak ability of target feature expression in related filtering tracking algorithms,this paper proposes a multi-feature compression tracking algorithm that integrates background information.In this algorithm,spatial context structure constraints are introduced for model optimization,and various optimization schemes are adopted.In order to further enhance the capability of feature expression,model response fusion is performed adaptively according to the maximum response value and APCE index,and feature extraction methods of position filter and scale filter are also compressed.Secondly,in order to solve the problem of in-plane rotation of the target in the tracking process,this paper adds a rotation detection module to the tracking algorithm based on kernel correlation filtering,extracts rotation angle information from the target images between adjacent frames.When the target is found to rotate in-plane,on the basis of obtaining the position filter detection coordinates,the foreground extraction method is used to correct the detection result of the position filter,and the confidence filter is set to make the tracking result more accurate,thus realizing stable tracking of the target under rotation.Finally,based on VS2013 and OpenCV machine vision library,the algorithm is compiled and tested on JN-IGH57 X hardware platform.Simulation results show that the proposed algorithm can achieve stable tracking of moving targets in dynamic background.