基于形态学的边缘检测系统与FPGA实现

Design of real-time edge detectionsystem based on the improved Sobel operator and its FPGA implementation

针对传统的Sobel算子边缘定位不够精确,对图像中的噪声响应较大的问题.提出一种基于形态学的边缘检测系统.采用4个方向的Sobel算子分离出图像的边缘点与非边缘点,将边缘点在梯度方向上进行形态学的膨胀操作;将非边缘点视为噪声点,采用腐蚀操作.为了便于工程应用,采用FPGA芯片对算法进行硬件加速.实验结果显示,该方法边缘定位更加精确,对噪声的响应更小.

The problem that the edge location of the traditional Sobel operator is not accurate enough to respond to the noise in the image. An edge detection system based on morphology is proposed. The 4 directions of Sobel operator are used to separate the edge points and the non-edge points of the image, dilation the edge points on the gradient direetion, and regard the non-edge points as the noise points, and adopt the corrosion operation. In order to facilitate the engineering application, the hardware acceleration of the algorithm is carried out by using FPGA chip. The experimental results show that the edge location of the method is more accurate and the response to noise is smaller.