Design of Fast Detection and Tracking System Based on FPGA

Abstract

Aiming at the problem that the classical Sobel edge detection algorithm is sensitive to noise and has a large amount of computation, an adaptive fast edge detection algorithm is proposed. Firstly, some background information which we are not interested in is removed through adaptive window to reduce the computational complexity. Then, Sobel operator is used to detect the real-time edge of the image in the window, and then morphological optimization is carried out. The fast detection algorithm in this paper is implemented on the FPGA. The results show that this method can effectively detect image edges and make the image edges clearer. At the same time, it can greatly reduce the noise and computational load in the image and reduce the power consumption.