基于小波去噪和图像分割技术的改进NAS-RIF盲图像复原算法

An improved NAS-RIF algorithm based on wavelet denoising and image segmentation

针对非负和支持域受限递归逆滤波算法(NAS RIF)的缺点,提出了一种改进方案。首先,应用小波去噪技术,保持了退化图像边缘特征,抑止噪声,提高退化图像的信噪比;其次,在每次迭代中,利用图像分割技术找到准确的目标支持域,并用背景的均值取代非均匀背景;还采用重置共轭梯度法加快了算法的收敛速度。实验结果表明,改进后的算法具有更好的复原效果和更快的收敛速度。

An improved method is presented to overcome the drawbacks of the original NAS-RIF algorithm and the denoising technique is used to preserve the edge feature of the degraded image, to restrain the noise amplification and to increase the signal-to-noise ratio(SNR). The image segmentation technique is applied in each iteration to find the precise object support region where the non-uniform background is replaced by means of the background. The algorithm resetting of the convergence of the conjugate gradient is also employed to speed up the convergence rate. The improved algorithm is experimentally to have better restoration effect and faster convergence rate.