图像的正则变换与快速自适应滤波.pdfVIP

电子学报990815 电子学报 ACTA ELECTRONICA SINICA 1999年　第27卷 第8期 Vo1.27 No.8 1999 图像的正则变换与快速自适应滤波* 李翠华　郑南宁 　　【提要】　本文给出了图像的一类卷积逼近公式，称之为图像的正则变换，其变 换核是无穷次连续可微的二元函数，并具有局部支撑性质.以此变换为基础，以最小二 乘为准则，构造了基于低松弛迭代格式的反卷积快速自适应滤波算法.对于一个N×N 输入图像，由于变换核的局部支撑性质以及低松弛迭代算法的引入，使得最小二乘滤 2 3 波算法的计算复杂度降为O(N )，比不动点(FP)迭代算法的O(N)及预处理共轭梯度 2. (PCG)算法或小波重构算法的O(N logN)都好，从而使最小二乘滤波算法真正成为高维 信号处理中的一类实用有效的自适应滤波算法.大量的实验结果表明，该滤波算法对 Gauss白噪声及均匀分布的噪声都有良好的抑制特性. 　　关键词:正则变换，去噪，图像处理，自适应滤波，最小二乘算法 Regular Transformation of Image and Fast Adaptive Filtering Li Cuihua,Zheng Nanning (Institute of Artificial Intelligence and Robotics,Xian Jiaotong University,Xian 710049) 　　Abstract:　In this paper,an approximate formula for image transformation is proposed. The formula referred to as a Regular-Transformation(RT) is derived from a convolution sum with a locally supported and infinitely differentiable kernel.According to the law of least squares,a RT-based fast adaptive filter with the underralaxation iterative scheme is developed.For a N×N image,the computational complexity of the filtering algorithm is O (N2 3 ) ，which is significantly better than O(N ) of the fixed point iterative method for handling LS problem and O(N2 log N) of both the preconditioned conjugate gradient iterative algorithm and wavelet-transform-based denoising algorithms.Consequently,the filter may be used in computer vision and real-time signal processing.The numerical results show that the filter is suitable for the reduction of both Gaussian noise and noise with uniform distribution. 　　Key words:　Regular-Transformation,Denoising,Image processing,Adaptive filtering, Least squares method 一、引　言 　　以最小二乘(LS)为准则