For improving Kirsch operator, this paper presents an edge detection algorithm with multi-threshold selection and edge connection, which is used to detect the edge of images, especially medical images. The method first denoises the original image by Gaussian filter and calculates the gradient with the traditional Kirsch operator, then sets a lower threshold and adaptive thresholds to protect the weak edge and pick up the true edge, and finally detects the disconnected edge of the image by use of the modified edge connection algorithm. The experiment results show the efficiency of this method.%论文针对Kirsch边缘检测算子的缺点,提出一种多阈值选取与改进的边缘连接算法相结合的方法,对图像,尤其是医学图像进行边缘检测.该方法首先用高斯滤波器对原始图像去噪,使用传统Kirsch算子计算梯度,然后设定低阈值和自适应阈值来保护图像弱边缘和提取真正边缘,最后通过改进的边缘连接算法连接边缘.实验结果表明,论文的方法具有保持弱边缘、连接性好的边缘检测效果.
展开▼