Research On Tracking And Detection Algorithm Of Infrared Weak And Small Target Image

With the development of infrared detection equipment and infrared imaging technology,the study of infrared target tracking and detection has gradually become a hotspot in the field of signal processing.This paper focuses on the research of infrared detection of low-altitude flying objects in complex ground backgrounds and aims to detect and track weak and extended infrared targets in images.The detection algorithm for weak targets in single-frame images is studied and improved,and the multi-frame image tracking algorithm for moving targets is enhanced,addressing the detection of extended targets as well.The improved algorithms are experimentally validated,and the results of the experiments are analyzed and evaluated.The main research content of this paper is as follows:To address the challenging problem of detecting weak targets in a complex background in a single frame due to environmental interference,this study proposes an object detection algorithm called high-boost weighted tri-layer local contrast measure(HB-WTLLCM).An image preprocessing algorithm,improved high boost filter(IHBF),is introduced to tackle the issue of generating a large number of high responses during the preprocessing stage in complex infrared backgrounds.The tri-layer window approach is utilized to calculate local contrast,addressing the difficulty of detecting infrared targets that are not prominent or clear using a dual-layer window.Additionally,a weight function called enhanced regional intensity level(ERIL)is proposed to evaluate the complexity of processed images,further enhancing the detection of weak targets.Experimental analysis is conducted to evaluate the algorithm,and improvements in evaluation metrics such as SCRG and BSF demonstrate that the proposed algorithm effectively enhances the ability to detect targets compared to mainstream detection algorithms,thereby improving the accuracy of target detection.To address the tracking challenges of weak and extended infrared targets in complex backgrounds,this study proposes an adaptive genetic algorithm particle filter(AGAPF).A residual mutation algorithm is introduced to overcome the issue of particle degeneracy caused by standard resampling,thus enhancing the diversity of particles and the accuracy of target tracking.Additionally,an improved adaptive genetic algorithm particle filter(IAGAPF)is proposed,incorporating the improved Metropolis Hastings(IMH)algorithm to enhance the mutation process and improve the algorithm’s computational efficiency.Experimental analysis confirms that the proposed algorithms enhance the conformity of target tracking trajectories.Compared to the standard particle filter algorithm,the proposed algorithms demonstrate an increase in the effective number of tracked frames,with an average NRMSE gain of approximately 40%,thus demonstrating their accuracy in infrared target tracking.