摘要
针对Canny边缘检测算法在伪边缘剔除方面的不足,结合洞室岩体裂隙图像成像特点,提出一种Canny算子与形态学相结合的边缘检测方法。先分别对原始图像进行Canny算子边缘检测和图像分割,得到两幅二值图像,然后参考图像裂隙的几何形态特征,对它们进行形态学运算,使最终得到的二值图像保持了较高的裂隙定位精度,并有效剔除了伪裂隙边缘,同时具有边缘闭合、平滑等优点。
Considering the drawbacks that Canny edge detection algorithm has in fake edge eliminating,a new method of edge detection that combines Canny operator and the morphology is proposed in view of imaging characteristics of tunnel rock fracture.Firstly,two binary images are generated from the original image after the Canny operator edge detection and image segmentation are conducted.Then,based on the geometric characteristics of image fracture,morphological operations are applied on these two images,which makes the binary image attained at last maintain a high accuracy in fracture positioning,and also has the fake edge eliminated effectively.Meanwhile,it has the virtue of close and smooth real edge.
出处
《计算机应用与软件》
CSCD
北大核心
2012年第3期262-264,共3页
Computer Applications and Software
