基于形态多结构元和小波变换的边缘检测算法

An Edge Detection Algorithm of Noisy Image Based on Wavelet Transform and Multiple Structuring Elements Morphology

提出了一种小波变换与多结构元形态学相结合的抗噪边缘检测方法。通过改进的小波边缘提取方法选择噪声图像的突变点,同时滤除部分噪声;针对图像中噪声和边缘形态的不同,建立了多个结构元素,采用多结构元形态检测算子对选取的突变点进行形态操作,在抑制噪声的同时,较好地提取了边缘。基于实验结果,指出对含有不同类型噪声(如椒盐噪声、高斯噪声等)的图像,该方法都可以较好地抑制噪声,提取边缘,且效果优于经典的边缘检测算法。

This paper puts forward an edge detection method based on wavelet transform and multi-structuring elements morphology, selects the coarse edge points and deleting part of noise by an improved edge detection method of wavelet transform, aiming at the morphology difference of noise and edge, constructs the multiple structuring elements, by using multiple structuring elements morphological operators detects the edge at the selected coarsc edge points, which detects the edge very well and restrain the noise also very well,and based on the experimental results, points out that for different kinds noisy images, such as salt and pepper noise and Gaussian noise and so on, the method can detect edge and restrain noise very well, and the effect of edge detection is better than classical edge detection.