Design And Implementation Of Image Enhancement And Template Matching System Based On FPGA

Target tracking technology has been widely used in military and civilian applications,and play an indispensable role.However,with the haze,cloudy and other low-light bad weather,making the overall image brightness is too low,the details are not clear,the contrast is low,picture quality decline and so on.This kind of image mainly displays in the range of the histogram of the pixel mainly in the low grayscale interval and does not disperse,the change range is also small,when the video image is being transmitted,processed and other operations,the quality of video image will be further reduced,which leads to the increase of mismatching rate in video image target tracking.So,in order to achieve better tracking effect,it is very important to study the image enhancement method and template matching algorithm of video image.Based on the characteristics of imaging system with low contrast and low illuminance,this paper combines the conditions of hardware FPGA system,and researches the design scheme of image enhancement and template matching system at home and abroad,and designs an image based on FPGA Enhanced and template matching systems.The main characteristic of low contrast and low brightness image is that the gray level is concentrated on the side of low gray level,the target is not easy for people to observe and for machine to process.It has certain negative influence on the realization of template matching to achieve target tracking.So we need to enhance the image first,so that the image is easy to human visual observation and subsequent template matching.In this paper,the image enhancement method is improved on the basis of gray-scale stretching,making the gray-scale stretching more uniform,more obvious texture information,image information richer,easy to observe the target.Template matching is carried out on the enhanced image,Analysis of the existing MAD,NCC and SSDA algorithms,compare their characteristics,And an improved template matching algorithm is proposed.In order to reduce the influence of the gray-scale value change,using the method of removing the DC component.and in order to reduce the computational complexity,a matching method of extracting row and column is adoptedIn this paper,a hardware image processing system based on FPGA is been disigend,builds the hardware platform,and implements the corresponding algorithm module on the platform.The hardware logic language of Verilog HDL is realized in the Quartus II development environment and downloaded to FPGA to achieve.The FPGA chip chosen in this paper is EP3SE80 Stratix III series of ALTERA company.It is proved that the designed system can realize the image enhancement and matching of video image,and meet the real-time performance.