Design of Edge Detection System Based on FPGA

• Published in: 2021 International Conference on Artificial Intelligence and Electromechanical Automation (AIEA) pp. 190 - 193
• Main Authors: Miao, Yuhao, Gao, Song, Liu, Jiaming, Xu, Jiaotao
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.05.2021
• Subjects: Filtering algorithms; FPGA; Image edge detection; image recognition; Licenses; Logic gates; Machine vision; Object detection; Roads; Sobel
• DOI: 10.1109/AIEA53260.2021.00048

As an image processing method, edge detection is widely used in object detection, image recognition, and machine vision. At present, the research on edge detection algorithms is still a hot spot in the field of image processing. In order to solve the problem that the large amount of salt and pepper noise in the image makes it difficult to identify the license plate of the vehicle entering and leaving the parking lot or on the road, and to identify the edges of important information more effectively, a traditional sobel operator edge detection algorithm is adopted and deployed on the Field programmable gate array(FPGA) platform with Xilinx's xc7a35tcsg324 as the main control chip. In this paper, the original image is converted into gray image by FPGA-based gray-scale conversion algorithm, and then the mean filtering algorithm and median filtering algorithm based on FPAG are adopted. By comparing the experimental results under 0.02 salt and pepper noise, it is concluded that the median filtering method can effectively filter out salt and pepper noise. After median filtering, the sobel edge detection algorithm is used. It is found that the sobel after median filtering is more obvious than the sobel before median filtering, and a lot of salt and pepper noise is filtered out.