中文摘要：

介绍一种基于小波变换的图像去噪方法。这一新的方法是基于阈值的小波变换。并且此方法更适用于非平稳信号下的图像去噪。该算法根据图像与噪声在小波域的分布特性以及它们的小波变换模极大值随尺度的变化大小不同,运用迭代算法得到不同尺度小波域中噪声的具体位置以及小波系数大小,完成了图像去噪。实验结果表明,对峰值信噪比较低的图像,该方法去噪后峰值信噪比比传统方法的高,并且保留了较多的图像细节,同时对平稳和非平稳的噪声都能进行较好地去噪。

英文摘要：

A new method of signal denoising is introduced in this article based on analysis of wavelet transform.The algorithm is wavelet transform based on threshold,and is more suitable for image denoising under non-stationary condition.According to the distributing characteristic of image and noise in wavelet domain,and their variation of wavelet transform mode maximum value with amplitude,the algorithm can obtain the location of noise in different scale wavelet domain and the wavelet coefficient by iterative algorithm for fulfilling the image denoising.Experiment results confirm that,for the image with lower PSNR,the PSNR after denoising with the proposed algorithm is higher than that of the conventional algorithm,the algorithm can keep more details of a image,and is effective to reduce the noise under both stationary and non-stationary environments.