改进的小波变换模极大值检测图像边缘的方法

提出了一种改进的基于小波多尺度多分辨率特征的数字图像的边缘检测算法，分别利用不同尺度小波变换后的水平方向和垂直方向高频信息，根据李氏指数与小波变换关系，采用小波模极大值在不同尺度下传播的特性，检测出图像在2个方向的极大值，然后利用模糊算法构造相应的隶属函数，提取弱边缘信息，最后得到不同尺度下的边缘图像。本算法可以兼顾良好的边界定位、噪声抑制和弱边界检测等性能指标，可以有效解决传统边缘检测方法中存在的搞定为精确及强去噪能力之间的矛盾。

An improved method of detecting the image edge by the modulus maxima of wavelet transform

An improved algorithm of numerical image edge detection based on multi-scale and multi-resolution characterization of wavelet transform has been described.This algorithm separately utilizes high frequency information of horizontal and vertical directions of wavelet transform under different scales.According to the relationship of Lipschitz exponent and the wavelet transform,the properties propagation under different scales are utilized to determine local maxima in two directions.Membership functions are designed with a fuzzy algorithm to detect weak edge information.Finally,all images of different scales are gotten.The algorithm can resolve the inconsistency between high precision localization and high de-noise ability existed in traditional edge detection algorithm.