Design Of The Image Sensor And Edge Detection Based On FPGA

With the gradual development of image processing technology and the emergence of high-performance image sensors,image acquisition is widely used in those areas,including face recognition,security monitoring,image measurement etc.Usually,the target object which is collected by image acquisition needs to be processed,while the edge detection is a very important step in image processing.When faced with massive image data,the real-time property of edge detection algorithm needs to be guaranteed.Among the existing solutions,the implementation of edge detection,which is based on FPGA can meet the demand of equipment miniaturization and low cost in production,so it has a very good application prospect.In this paper,taking this solution into consideration,an improved edge detection scheme based on further analysis of the existing edge detection algorithm is proposed.The main work can be divided into the following aspects:First of all,on the basis of reading a large number of literature,the principle and significance of image sensor were introduced,the current situation and development trend of edge detection were expounded in detail,then some relevant FPGA technologies were also introduced.So a corresponding solution has been determined;Secondly,for the two types of image sensor about CCD and CMOS,we designed the respective peripheral drive circuit and the CMOS image sensor OV7725 was chosen to design a video image acquisition system based on FPGA.Using Verilog HDL to program the image acquisition module,SDRAM cache module,data processing module and image display module,then we use the verification language System Verilog to verify these modules;Thirdly,the collected images were processed in real time with the requirements of edge detection,and then were analyzed mainly from the perspectives of filtering and image boundary condition.Including denoising algorithm of the two kinds of noise,morphological filtering processing and threshold implementation,the effects and the application of the method were compared and analyzed.By analyzing the inadequacies of filtering and threshold,then the implementation of median filtering algorithm and threshold were improved on the basis of existing algorithm;Fourthly,an improved edge detection scheme is proposed.After comparing andanalyzing several kinds of traditional edge detection operators,then we choose the Kirsch operate as the benchmark to make improvement for the edge detection system of the image,besides,experiments on edge detection of the gray images with a character as the main body and colorful flowers are carried out.Through the analysis of the final experimental results,it is shown that the proposed improved edge detection scheme can help further improve the edge detection effect of the image which is to be detected.