| Infrared target recognition and tracking is a hot research direction in the field of computer vision.Scholars at home and abroad have done a lot of research on the characteristics of infrared images such as strong concealment,strong anti-interference and strong environmental adaptability.Relevant academic achievements have been widely used in military fields,industrial application fields and civil fields such as transportation and video surveillance.However,in the military field,with the improvement of system integration and the requirements of micro-integrated system,the processor of system equipment should not only meet the high performance requirements of image processing algorithms,but also have the characteristics of small size and low power consumption.Therefore,it is particularly important to complete the hardware system design and prototype development of infrared target tracking and recognition with small size,light weight and low power consumption.Based on the principle of infrared image recognition and tracking of vehicles,tanks and other targets in real battlefield environment,and through comprehensive comparison and analysis of the advantages and disadvantages of various hardware platforms,this paper proposes a scheme of infrared vehicle target recognition and tracking system based on single DSP processor.Aiming at miniaturization and low power consumption,the DSP hardware system platform is designed based on the idea of modularization.DSP hardware platform mainly includes video acquisition unit,image processing and control unit and video output unit.The video acquisition unit is mainly used to complete the image acquisition of target infrared information and output the acquired infrared image information to the image processing and control unit.Image processing and control unit is mainly used for infrared image preprocessing,target recognition and target tracking processing,and controls the video acquisition unit and video output unit to acquire and output target images.The video output unit is used for receiving the image output by the image processing and control unit,and transmitting the image to the display for display.The Gaussian filter+threshold segmentation+sequential similarity detection algorithm proposed in this paper is simple and effective,and is suitable for single DSP processor.Gaussian filter and threshold segmentation algorithm can reduce the background noise of the image and accurately segment and extract the target image.Sequential similarity detection algorithm can better realize the target image recognition and tracking.Based on VS and Open CV,the effectiveness and practicability of Gaussian filter,threshold segmentation and sequential similarity detection algorithm for infrared vehicle target recognition and tracking are simulated and verified.The results show that the algorithm meets the needs of infrared vehicle target recognition and tracking.Combined with the hardware resource characteristics of hardware system platform,the image processing algorithm is transplanted,and the optimization design to improve the image processing efficiency is carried out,and the hardware and software integration and application of infrared vehicle target recognition and tracking system are completed.The results show that the overall dimensions of the system are 5cm*6cm*1cm,and the system power consumption is not more than 3w,which is very suitable for weapon systems such as small unmanned aerial vehicles or portable ground equipment with strict power supply requirements and small space in the field,air and sea. |
