自适应全变分图像去噪模型及其快速求解*

在联合冲击滤波器和非线性各向异性扩散滤波器对含噪图像做预处理的基础上,利用边缘检测算子选取自适应参数,构建能同时兼顾图像平滑去噪与边缘保留的自适应全变分模型,并基于Bregman迭代正则化方法设计了其快速迭代求解算法.实验结果表明,自适应去噪模型及其求解算法在快速去除噪声的同时保留了图像的边缘轮廓和纹理等细节信息,得到的复原图像在客观评价标准和主观视觉效果方面均有所提高.

Adaptive total variation model for image denoising with fast solving algorithm

This paper combined shock filter with anisotropic diffusion to preprocess the noisy images, and used the edge detection filters to choose the parameters adaptively based on the preprocessed images. Then introduced an adaptive total variation regularization model for image denoising based on the chosen parameters. The proposed model could keep the balance between noises smoothing and edges preserving adaptively. Furthermore, it proposed a fast iterative algorithm to solve the proposed adaptive model based on Bregman iteration regularization method. The numerical results show that the proposed model and fast algorithm can smooth the noises and preserve the edge and fine detail information properly with fast solving convergence rate, while the peak signal to noise ratio, mean structural similarity and subjective visual effect of the denoised images are improved obviously.