基于层间相关性的小波边缘检测

Wavelet Edge Detection Based on Inter-scale Indenpdency

在计算机视觉系统当中,图像边缘检测占据着重要位置.图像成像过程中,噪声、畸变等因数引起图像细节模糊,如何有效提取图像边缘成为一个不断探索的课题.众所周知,采用Sobel、Laplacian边缘检测算子尽管能够有效地去除噪声的干扰,但是破坏了图像边缘细节.基于此,本文提出了一种基于层间相关性的小波边缘检测.在不同层间,如果低分辨率下的父系数绝对值大,则在高分辨率下四个相应的子系数的绝对值可能性也大,则其所代表地图像边缘信息可能性越大.因此,采用基于层间相关性的小波边缘检测算法既有效地去除噪声的干扰,又保留了图像边缘细节.

In the system of theory on computer vision, edge detection of image plays an important role.Because noise or aberrance blurs edge of image, how to extract edge of image becomes a researching subject..Though Sobel or Laplacian operator effectively resolve noisy disturbance ,it lose edge of image. Therefore, this paper takes advantage of algorithm of wavelet transformation based on inter-scale dependency to extract edge of image.Its principle is that if absolute value of parent coefficient in coarse layer are larger magnitude, absolute value of son coefficients are possibly larger. Therefore , it presents edge more possibly.So, this new algorithm can effectively not only resolve noisy disturbance but also preserve edge of image.