基于小波变换图像边缘检测的研究  

The research of image edge detection based on wavelet-transform

作　　者：杜航行 黄梓瑜[1] 何佳蓉[1] 

机构地区：[1]西南石油大学电气信息学院,四川成都610500

出　　处：《自动化与仪器仪表》2016年第1期13-14,共2页Automation & Instrumentation

摘　　要：基于小波变换的图像边缘检测是一种新的检测图像边缘的方法,具有多分辨率,多尺度的特性。本文采用基于小波变换的模极大值原理,利用不同尺度小波变换后的不同方向上的特性,获取高频信息。再通过小波系数的模极值点与过零点,检测出图像在四个方向上的模极大值,并记录下来,得到图像边缘的位置后,进行比较,得到该位置模的局部最大值。研究结果表明,基于小波变换的图像边缘检测可以较好地检测图像边缘的细节特征,取得了很好的效果。Image edge detection based on wavelet transform is a new method for detecting image edges,with multi-resolution,multiscale features. In this paper,the wavelet transform modulus maximum principle based on the use of properties in different directions at different scales on wavelet transform to obtain high frequency information. And through the modulus maxima of wavelet coefficient and the crossing points to detect the image in four directions of modulus maxima,record down to get the position of image edge,then compare,obtain the local maximum of the location model. The research results show that the image edge detection based on wavelet transform can better detect the image edge details of the characteristics of good results have been achieved.

关 键 词：小波变换 边缘检测 多分辨率 多尺度 模极大值 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]