尺度独立MAS小波算法检测目标图像边缘

通过分析图像中不同类型的奇异点，进而区分目标图像的不同类型边缘，由此来解决带有噪声干扰条件下的图像边缘检测问题。小波理论在时频域的紧支性和对非平稳信号（噪声）良好的处理效果，在图像处理的研究中大大优于传统的Fourier变换，并广泛应用于图像去噪和边缘提取方面。在小波理论中，Lipschitz指数可以刻画不同的奇异点，从而区分出图像的边缘和背景噪声。在此基础上，提出了一种独立于尺度的有效算法，并应用于模角分离（modular—angle—separated，MAS）的小波函数。用该算法处理二维图像的边缘，效果优于同类方法，且特别适用于有噪声背景的图像边缘提取。

Algorithm of Scale-Independent MAS Wavelet for Image Edge Detection

This paper analysed different kind of odd points and distinguished different kind of edges in image,and solved the detection of image edge in the noisy background.Wavelet theory is widely used in image denoising and edge extraction because of the character of time-frequency compact support and the nice effect of denoising,so the wavelet transformation is much better than Fourier transformation in the field of image processing.In the wavelet theory,the Lipschitz index can interpret different kind of odd points and distinguish edges from the noise of the image.Based on this,a scale-independent algorithm,which applied in the modular-angle-separated(MAS) wavelet transformation is proposed.The algorithm is used in processing edges of 2-D image,especially in the edge extraction of noisy background.