预光滑子正则化求解的图像修复策略

Pretreatment Smoothing Regularized Image Restoration Strategy

由于传感器噪声或者拍摄抖动,容易导致数字图像含有噪声,所以必须对模糊图像进行修复处理,本文针对正则化模型在图像修复中还存在的抗噪性能较差的问题,提出了一种预光滑子正则化求解的图像修复策略。首先采用软阈值对正则化去噪模型进行最优化求解,然后构建基于离散小波的多重网格,然后为了得到最优正则化参数,采用预光滑子策略对其最粗层进行优化,并采用软阈值方法消除残留的高频信息。算法仿真实验结果表明,本文提出的方法在大多数噪声水平下比其它方法表现更优秀,并且计算时间明显比其它方法更少。

Because of the noise or shake of the sensor,it is easy to cause the digital image to contain the noise,so the fuzzy image must be repaired. In this paper,aiming at the poor anti-noise performance of the regularization model in image restoration,Regularized image restoration strategy. Firstly,a soft threshold is used to solve the regularization denoising model. Then a multi-grid based on discrete wavelet is constructed. Then,in order to get the optimal regularization parameter,the pre-smoothing sub-optimal strategy is adopted to optimize the denoising model, Threshold method to eliminate residual highfrequency information. The simulation results show that the proposed method performs better than other methods at most noise levels,and the computational time is significantly less than that of other methods.