针对Video Block-Matching and 3-D Filtering(VBM3D)算法耗时高且去噪视频存在块效应的问题进行了改进。在对变换域系数进行收缩时,采用连续阶导数阈值法,代替原算法中的硬阈值法,减少块效应;在帧内匹配时,采用基于积分图思想的图像块...针对Video Block-Matching and 3-D Filtering(VBM3D)算法耗时高且去噪视频存在块效应的问题进行了改进。在对变换域系数进行收缩时,采用连续阶导数阈值法,代替原算法中的硬阈值法,减少块效应;在帧内匹配时,采用基于积分图思想的图像块距离计算加速方法;在帧间匹配时,使用帧间预测性匹配方法,减少计算量,提高算法效率。理论分析和实验结果表明,改进后的算法不仅能有效改善原VBM3D算法中的块效应,且算法复杂度大大降低。展开更多
Block matching based 3D filtering methods have achieved great success in image denoising tasks. However the manually set filtering operation could not well describe a good model to transform noisy images to clean imag...Block matching based 3D filtering methods have achieved great success in image denoising tasks. However the manually set filtering operation could not well describe a good model to transform noisy images to clean images. In this paper, we introduce convolutional neural network （CNN） for the 3D filtering step to learn a well fitted model for denoising. With a trainable model, prior knowledge is utilized for better mapping from noisy images to clean images. This block matching and CNN joint model （BMCNN） could denoise images with different sizes and different noise intensity well, especially images with high noise levels. The experimental results demonstrate that among all competing methods, this method achieves the highest peak signal to noise ratio （PSNR） when denoising images with high noise levels （σ 〉 40）, and the best visual quality when denoising images with all the tested noise levels.展开更多
目的图像在获取和传输的过程中很容易受到噪声的干扰,图像降噪作为众多图像处理系统的预处理模块在过去数十年中得到了广泛的研究。在已提出的降噪算法中,往往采用加性高斯白噪声模型AWGN(additive white Gaussian noise)为噪声建模,噪...目的图像在获取和传输的过程中很容易受到噪声的干扰,图像降噪作为众多图像处理系统的预处理模块在过去数十年中得到了广泛的研究。在已提出的降噪算法中,往往采用加性高斯白噪声模型AWGN(additive white Gaussian noise)为噪声建模,噪声水平(严重程度)由方差参数控制。经典的BM3D 3维滤波算法属于非盲降噪(non-blind denoising algorithm)算法,在实际使用中需要由人工评估图像噪声水平并设置参数,存在着噪声评估值随机性大而导致无法获得最佳降噪效果的问题。为此,提出了一种新的局部均值噪声估计(LME)算法并作为BM3D算法的前置预处理模块。方法本文专注于利用基于自然统计规律(NSS)的图像质量感知特征和局部均值估计技术构建图像噪声水平预测器,并通过它高效地获得噪声图像中准确的噪声水平值。关于自然场景统计方面的研究表明,无失真的自然场景图像在空域或者频率域上具有显著的统计规律,一旦受到噪声干扰会产生规律性的偏移,可以提取这些特征值作为反映图像质量好坏的图像质量感知特征。另外,局部均值估计因其简单而高效率的预测特性被采用。具体实现上,在具有广泛代表性且未受噪声干扰图像集合上添加不同噪声水平的高斯噪声构建失真图像集合,然后利用小波变换对这些失真图像进行不同尺度和不同方向的分解,再用广义高斯分布模型(GGD)提取子带滤波系数的统计信息构成描述图像失真程度的特征矢量,最后用每幅失真图像上所提取的特征矢量及对其所施加的高斯噪声水平值构成了失真特征矢量库。在降噪阶段,用相同的特征提取方法提取待降噪的图像的特征矢量并在失真特征矢量库中检索出与之类似的若干特征矢量及它们所对应的噪声水平值,然后用局部均值法估计出待降噪图像中高斯噪声大小作为经典BM3D算法的输入参数。�展开更多
文摘针对Video Block-Matching and 3-D Filtering(VBM3D)算法耗时高且去噪视频存在块效应的问题进行了改进。在对变换域系数进行收缩时,采用连续阶导数阈值法,代替原算法中的硬阈值法,减少块效应;在帧内匹配时,采用基于积分图思想的图像块距离计算加速方法;在帧间匹配时,使用帧间预测性匹配方法,减少计算量,提高算法效率。理论分析和实验结果表明,改进后的算法不仅能有效改善原VBM3D算法中的块效应,且算法复杂度大大降低。
基金This research was supported by the National Natural Science Foundation of China under Grant Nos. 61573380 and 61672542, and Fundamental Research Funds for the Central Universities of China under Grant No. 2016zzts055.
文摘Block matching based 3D filtering methods have achieved great success in image denoising tasks. However the manually set filtering operation could not well describe a good model to transform noisy images to clean images. In this paper, we introduce convolutional neural network （CNN） for the 3D filtering step to learn a well fitted model for denoising. With a trainable model, prior knowledge is utilized for better mapping from noisy images to clean images. This block matching and CNN joint model （BMCNN） could denoise images with different sizes and different noise intensity well, especially images with high noise levels. The experimental results demonstrate that among all competing methods, this method achieves the highest peak signal to noise ratio （PSNR） when denoising images with high noise levels （σ 〉 40）, and the best visual quality when denoising images with all the tested noise levels.
