基于小波变换和稀疏表示的图像去噪

传统小波阈值去噪在对图像进行去噪时,并不能很好地保留图像的细节纹理等边缘信息部分.针对这一不足,结合了稀疏表示相关的理论,提出了一种基于小波变换和正交匹配算法相结合的图像去噪算法.首先选取小波函数对含噪图像进行处理,分离出图像的高频和低频小波系数,然后对高频系数结合正交匹配追踪算法,通过多次反复迭代求得高频稀疏分量,再结合低频分量,用逆小波变换得到恢复图像.实验结果表明,在相同的噪声条件下,该算法能取得较好的峰值信噪比(PSNR),获得更好的视觉效果.

Image Denoising Based on Wavelet Transform and Sparse Representation

The traditional wavelet threshold denoising method can not well preserve the detail texture part of image edge information in image denoising.To solve this problem,a new image denoising algorithm based on wavelet transform and orthogonal matching pursuit(OMP) is proposed.First,the wavelet transform is used for the noisy image,and the high-frequency and low-frequency wavelet coefficients of image are separated.Then the OMP algorithm is employed on the high-frequency wavelet coefficients,and repeated iteration to obtain the high-frequency sparse components.Finally,combined with the low-frequency components,the restored image is obtained by inverse wavelet transform.The experimental results show that under the same noise conditions,the algorithm can abtain a better PSNR and improve the visual effect of the recovered image.