| Edge detection is an important part in digital image processing,which is widely used in machine vision,image diagnosis,target recognition,industrial monitoring and other fields.Therefore,the research of edge detection becomes an important subject in the field of image processing.Edge detection has produced many classical algorithms in theory.With the development of modern high technology,edge detection has made great breakthroughs in anti-noise performance and image processing effect.In order to solve the characteristics of real-time and fast image transmission,an image edge detection method based on FPGA is proposed.Firstly,this paper introduces the research background and significance of edge detection as well as the development status at home and abroad.By comparing the common methods for edge detection,Canny operator is used as the edge detection algorithm in this design.From smoothing filter,gradient calculation,non-maximal suppression,adaptive threshold detection and edge connection,the corresponding algorithms are proposed.Canny algorithm for the key technology,the use of the second derivative gradient histogram threshold adaptive,in MATLAB simulation verification.CORDIC algorithm was used to calculate the gradient,simplifying the complexity and greatly simplifying the FPGAoperation.In edge connection,the connector is optimized to achieve edge integrity.According to the characteristics of FPGA and the selection of appropriate logic,this paper introduces the design and implementation scheme of each module in FPGA,tests and verifies the feasibility of each module on the FPGA hardware platform.The design of pipeline and parallel algorithm is used to optimize the circuit.The original image data is used as a MIF file to qualitatively and quantitatively analyze the edge of the tested image.The test results show that the corresponding function of edge detection algorithm is realized on FPGA,which achieves the design goal and is an exploration in the field of image. |
PDF Full Text Request