Research And Implementation Of Infrared Image Tracking Based On DSP

In recent years,with the continuous development of science and technology,people's research on image recognition and tracking has entered a new stage.Among them,infrared images are widely used in the military field due to their local invariant features,such as sea surface ship target recognition and air target recognition.In the research process of this paper,the tracking algorithm of infrared image is selected by TMS320DM642 chip.This paper first introduces the background and research significance of infrared image tracking based on DSP.Secondly,it introduces the characteristics of infrared image and the software and hardware design ideas of the system.It is aimed at infrared image preprocessing,image segmentation,moving target recognition and moving target tracking.The theoretical basis of different algorithms is introduced and the experimental simulation is carried out.The main work of this paper is as follows:(1)The infrared image preprocessing algorithms such as mean filtering,median filtering,morphological operation,edge detection and threshold segmentation are analyzed by simulation.The results show that the threshold segmentation algorithm works best,thus determining the suitable infrared ship for Haitian background.The best processing algorithm for the target.(2)Comprehensive simulation analysis of adjacent frame difference method and template matching method in moving target recognition algorithm and related tracking,predictive tracking,Cam-Shift and other algorithms in moving target tracking algorithm.The Cam-Shift algorithm is based on the fact that the target information in the video changes its color information during the movement to achieve fast tracking of the target.The algorithm is less affected by the change of the target shape,and can effectively solve the problem that the target moves or the target part is occluded.Because the traditional Cam-Shift algorithm has poor tracking performance when the lighting conditions are changed,this paper improves on the traditional Cam-Shift algorithm,and can adjust the algorithm threshold according to different saturation and brightness.A more stablerecognition tracking effect.(3)The system learns the convolutional neural network related algorithm,and uses the algorithm of convolutional neural network to realize the classification and recognition of infrared targets,and obtains the correct classification results and high confidence;based on DM642 hardware platform,comprehensive utilization of infrared Image preprocessing and segmentation related algorithms,moving target recognition and tracking related algorithms,preliminary realization of DSP-based infrared image tracking research and implementation system functions,completed the project requirements.