Research On The Techniques Of Infrared Target Detection And Tracking In Large Field Of View Complex Background

Since the photoelectric target detection technology is an important part of the precision guidance and target positioning system,a lot of research work has been carried out around the target detection and tracking technology.In the infrared images collected by a large field of view(FOV)photoelectric detection system,there are a lot of complicated situations such as occlusion of smoke,unclear target details and high brightness interference,which makes it difficult for low-altitude target detection and confirmation and ground target tracking.This paper focues on the several key technologies of infrared target detection and tracking in the large FOV complex background.The work of this paper is organized as follows:In the first part,the research on dynamic compression algorithm of high dynamic range(HDR)infrared images is carried out,and the target characteristics of this paper are analyzed.Firstly,the composition of the photoelectric subsystem is introduced.Then a HDR image compression algorithm based on guided image filtering(GIF)and layered processing is designed,and the measured data is used to verify that the algorithm can enhance the edge detail of the target while maintaining contrast.Finally,the characteristics of the large FOV low-altitude flying target and the moving target are analyzed.In the second part,the research on single-frame detection and target confirming algorithm for low-altitude flying targets is carried out.First of all,the commonly used target detection algorithm is introduced.Secondly,a single-frame target detection algorithm based on double-scale filtering is proposed.The experimental results show that the algorithm can accurately segment the target and reduce the number of false alarms.Based on the single frame detection result,a phased target confirming method based on discriminant logic design is proposed,and the effectiveness of the algorithm is verified by the measured data under different conditions.The third part studies the tracking algorithm of moving targets in complex background.To start with,the basic principles of the Kernelized Correlation Filter(KCF)algorithm are introduced.Then,aiming at the shortcomings of KCF,which is easy to lose target after camera shake and occlusion,and without scale adaptability,a moving target tracking algorithm based on improved KCF(IKCF)is designed in combination with GIF and SAMF algorithm.Finally,experiments verify that the algorithm can cope with the blur,scale changes and short-term disappearance of the target during the movement.