Design Of Target Recognition And Tracking System Based On FPGA

The real-time recognition and tracking of moving targets is an important research direction in the fields of machine vision and intelligent monitoring.This technology has been widely used in many fields of military and civilian.Nowadays,the image processing algorithms based on target recognition and tracking has been increasingly perfect,but most of these algorithms are based on the software platform on personal computer,and their convenience and cost are difficult to meet the requirements in some fields and places.Therefore,the hardware implementation of target recognition and tracking algorithm has become a popular research direction.FPGA has been widely used in the field of image processing because of its fast processing speed and parallel processing.In addition,compared with the traditional image processing system based on personal computer,it has great advantages in system cost,product maintenance and program upgrade.With the improvement of modern integration technology,its integration will be higher and higher,the performance will be more prominent,and the applicable fields and occasions will be more and more.Therefore,it has a high academic prospect and practical value to use FPGA as processor to detect and process images in real time and quickly.In this paper,the self-made FPGA development board is used as the experimental platform to realize the target recognition and tracking in the dynamic background.This paper first introduces the development software of FPGA and its development flow,then introduces the hardware resources of the development board in detail,then introduces several common target detection and tracking algorithms in detail,and combined with its advantages and disadvantages and the development characteristics of FPGA,finally adopts the SAD template matching tracking algorithm based on color feature detection to realize the target recognition and tracking in the dynamic background.Finally,the FPGA implementation process of the algorithm is introduced in detail,and the function verification and sequence verification of the algorithm are carried out by Matlab and Modelsim software.The system is divided into five parts: image acquisition module,image storage module,image processing module,image display module and hardware tracking module.Firstly,the functions of each module are realized by Verilog hardware description language,and then its functions are simulated and verified by Modelsim software.Finally,the code is burned into FPGA chip to achieve the purpose of target recognition and tracking.Through the comprehensive verification,the system can realize the recognition and tracking of the target in the dynamic background.