To address the problem of difficulty in recognizing and tracking targets using infrared thermal imaging cam-eras in a complex and rapidly changing battlefield environment,a moving target recognition and tracking algorithm based on infrared image processing is proposedin this paper.The algorithm mainly consists of two parts:detection and tracking of mobile targets.Firstly,the Gaussian background modeling method is used to extract moving targets.Second-ly,Kalman filtering is applied to locate and track moving targets,and finally,the Hungarian algorithm is used to deter-mine the attribution of target points.The experimental results show that compared with the SORT tracker,the proposed algorithm improves MOTA by 1.4％,MT by 11.9％,reduces ML by 6.5％,IDs by 33.3％,and FN by 13.6％,which improves the tracking accuracy and successfully realizes the tracking of multiple moving targets.The algorithm in this paper has broad application prospects in military fields.
维普
万方数据