关键词: 非均匀剖分, V系统, 离散V变换, 图像消噪, 空域, 频域

Abstract: A new images denoising method based on non-uniform partition approximation and V-transform is proposed.The key point of non-uniform partition approximation is that the gray values of the pixels are regarded as fitting data，which are fitted with the least squares method，and then the digital image is expressed as a piecewise polynomial.It can retain the edge as well as detailed characteristics of image.The non-uniform partition approximation can realize image denoising by controlling partition precision value.V-system is a class of complete orthogonal function system on L2[0，1]，with the characteristics of the multi-resolution of multi-wavelet.It can transform a digital image from spatial space to frequency domain，and by processing the coefficients in high frequency and low frequency to realize image denoising.This paper combines the advantages of these two methods to process image deniosing.The experimental results show that the proposed method is efficient to improve the quality of denoising images.

Key words: non-uniformly partition, V-system, discrete V-transform, image denoising, spatial space, frequency domain