FPGA-based Improved Canny Edge Detection System

摘要

In order to address the issues associated with the classic Canny operator, such as high computational load, incomplete edge information capture, and poor real-time performance, an improved multi-directional Canny edge detection method is proposed for using in service robot navigation systems. This method collects image information through the OV5640 camera, preprocesses the image after grayscale processing combined with mixed multiple filtering, and then performs eight directional adaptive threshold edge detection based on the classical Canny operator. After image enhancement, it runs on the FPGA chip. It can be seen from the experimental results that compared with the classic edge detection algorithm, the improved Canny operator has significantly improved the detection effect and can efficiently process the edges of various complex images in real time, providing a feasible solution for real-time image processing in the FPGA field.