一种改进的基于小波变换的图像边缘提取算法

A Revised Image Edge Detection Method Based on Wavelet

在对现有的基于小波变换的图像边缘提取方法进行简要分析的基础上,提出了图像边缘提取改进算法.该算法主要是对边缘点的判断准则进行了改进,通过将任一点的梯度方向与水平、垂直、45°、135°方向进行比较,将这4个方向中与该点梯度方向最接近的一个方向重新赋给该点的梯度方向,并检验二进小波变换的模在该点是否为新的梯度方向上的局部极大值,若是,该点就是边缘点,否则不是.最后通过模拟实验给出了该方法的实验结果,实验结果有力地证明了该方法的有效性.

This paper present the existing image edge detection method based on dyadic wavelet trans form first, and then analyzed the drawback of the method. Based on the aforementioned analysis, a revised image edge detection algorithm is suggested. The revised algorithm mainly modify the edge point judge cri terion, for any point of the image, we compare the gradient direction of the point with four directions, ho rizon, verticality, forty-five degree and one hundred and thirty-five degree, then assign the nearest direction to the gradient direction of the point, furthermore, check the module of dyadic wavelet transform at this point. If the module is the local maximum in the new gradient direction of the point, the point is edge point, otherwise not. In the end, the experimental results are presented. Experimental results show that the proposed algorithm is very reliable and efficient.