边缘检测算子研究及其在医学图像中的应用

Analysis of Edge Detection Operators and Application in Medical Image

阐述了边缘检测在医学图像处理中的重要作用,分析了几种常用算子(Roberts,Sobel,Prewitt,Laplacian,Canny)基本原理及算法实现,并对比各算子的优缺点,并用VC++.NET工具编程实现出以上各种算子对一幅人体头部的MRI图进行边缘提取。对实验结果进行分析得出结论,各种不同的算子对同一幅图片进行边缘提取,会得到完全不同的结果,Canny算子检测出的边缘比一般微分算子更精确、更细,但也可能平滑掉一些有用的边缘。因此,在不同的条件下,应根据具体情况,选择最适合的边缘检测算子对图像进行处理,才能得到最佳效果。

Edge detection is crucial in the field of medical image processing. In this paper, some operators（Roberts, Sobel, Prewitt, Laplacian, Canny） which are frequently used are studied theoretically and arithmetically, and compared with the characteristic of these methods on image edge detection. Then the operators discussed above are realized and applied to an MRI image edge processing. And we conclude the different results gained by extracting the edge of the same image with different kinds of operators. For example, the operator of Canny will lead to more accurate edge but at the same time may miss some effective edges. So we should select appropriate operators according to specific conditions for optimal results.