自适应实时边缘检测系统设计

The design of adaptive real-time edge detection system

针对传统边缘检测系统缺乏自适应及实时性差的问题,利用FPGA的高速并行处理能力和改进的Canny算子设计了一种自适应实时边缘检测系统。系统根据采集图像的幅值梯度直方图特征,自主提取Canny算子所需的高低阈值,并辅以双阈值化处理来提取图像边缘信息;同时采用多级流水和并行处理的方式,将浮点数转化为整数计算,加快处理速度。实验结果表明,该系统提取边缘完整,实时性好,自适应能力强。

Aiming at the problem that the traditional edge detection system is lack of adaptive and poor real-time performance, an adaptive real-time edge detection system is designed by using the high speed parallel processing ability of FPGA and the improved Canny operator. The system extracts the high threshold value of Canny operator according to the characteristic of the amplitude gradient histogram, and extracts edge information by double threshold processing. At the same time,it converts floating point number into integer to improve the processing speed by multilevel pipelining and parallel processing. The experimental results show that the system can extract complete edge and has the characters of good real-time performance and strong adaptive ability.