一种基于小波变换的边缘检测算法

A Wavelet-Based Edge Detection Algorithm

目的利用小波变换,完成图像边缘的检测,提取能表征图像概貌的重要特征。方法学习并比较几种边缘检测方法,结合小波变换的多尺度特性,对小波变换分解的近似系数和高频系数进行处理,构建模值和幅角图像,提出一种基于小波变换的图像边缘检测算法。结果在MATLAB R2018b环境下,首先以屋顶图像为例,将算法与Roberts算子、Sobel算子、Canny算子、LOG算子进行对比,通过添加椒盐噪声与高斯白噪声,验证算法在边缘定位和抗噪方面取得了较好性能;然后将算法应用于其他图像,边缘检测效果依然稳定。结论基于小波变换的边缘检测算法具有较好的抗噪性与适应性。

Objective To detect the image edge by wavelet transform and extract the important features representing the image profile.Methods By learning and comparing several edge detection methods,in combination of the multi-scale characteristics of wavelet transform,the approximate coefficients and high frequency coefficients decomposed by wavelet transform were processed,and modulus and argument images were constructed.Then an image edge detection algorithm based on wavelet transform was put forward.Results With the aid of MATLAB R2018b,the roof image was taken as an example to compare the algorithm with Roberts operator,Sobel operator,Canny operator,and LOG operator.By adding salt and pepper noise and Gaussian white noise,it was verified that the algorithm achieved good performance in edge location and anti-noise;when the algorithm was applied to other images,the edge detection effect was still stable.Conclusion The edge detection algorithm based on wavelet transform has good anti-noise and adaptability.