基于离散平稳小波变换和Bayes估计的红外图像去噪方法

提出一种基于离散平稳小波变换与Bayes估计的红外图像去噪方法。针对红外图像信号与噪声的不同特点，对红外图像进行离散平稳小波变换后，采用Bayes估计方法自适应地从带噪信号的小波系数观测值中估计信号的小波系数值。该方法不仅有效克服了传统离散正交小波变换去噪时容易产生Gibbs现象的问题，也避免了常用小波去噪算法对小波系数截断式处理可能造成的图像细节模糊和信息丢失。最后，通过对实际电力设备红外图像的去噪实验，验证了方法的有效性。

De-noising of infrared images based on stationary wavelet transform and Bayes estimation

The de-noising method based on the stationary wavelet transform and Bayes estimation is proposed for infrared image.According to the trait of frequency distributing of infrared image signal and noise,this paper used Bayes method to estimate the wavelet coefficients on different orientations of different wavelet scaling spaces adaptively.The experiment shows that this method not only avoiding the Gibbs phenomena,but also keeping the information of the original image and excellent on noise removing.