交替使用小波去噪和全变差正则化的盲图像恢复算法

盲图像恢复就是在点扩散函数未知情况下从降质观测图像恢复出原图像.该文提出了一种交替使用小波去噪和全变差正则化的盲图像恢复算法.观测模型首先被分解成两个相互关联的子模型,这种分解转化盲恢复问题成为图像去噪和图像恢复两个问题,可以交替采用图像去噪和图像恢复算法求解.模糊辨识阶段,使用全变差正则化算法估计点扩散函数;图像恢复阶段,使用小波去噪和全变差正则化相结合的算法恢复图像.实验结果和与其它方法的比较表明该文算法能够获得更好的恢复效果.

Blind Image Restoration Algorithm Iteratively Using Wavelet Denoising and Total Variation Regularization

Blind image restoration is to recover the original image form the observed degraded image with unknown the Point Spread Function (PSF). This paper proposes a blind image restoration algorithm iteratively using wavelet denoising and total variation regularization. The observation model is first divided into two mutually associated sub-models, and this representation converses the blind restoration into the two issues of image denoising and image restoration, which makes us to solve the problem by iteratively using image denoising and image restoration algorithms. The stage of the PSF identification uses Total Variation (TV) regularization and the stage of image restoration uses wavelet denoising and TV regularization. The experimental results show that the proposed algorithm achieves better performance than the existing algorithms.