应用小波域三维Context模型的视频图像去噪

提出了一种基于三维小波变换和分块Context模型的视频去噪新方法（3DWTBCM）。视频图像序列的各帧之间具有较强的相关性，在三维小波变换域内去噪可以很好地将这种相关性加以利用。根据视频图像三维小波分解域内系数和噪声分布的特征，利用小波系数具有局部相关性对小波系数进行分块，将系数分解成各个局部区域。再将Context模型用于局部块中，按照能量分布将块内的小波系数分成多个子块。对各部分进行能量估计和多阈值估计，获得去噪声最佳阈值，有效地消除噪声。实验结果表明，3DWTBCM的噪声抑制效果明显优于各种2D去噪声方法，和常用的3D去噪声方法，PSNR平均提高1.5dB以上。从视觉效果来看，本文算法在去除噪声的同时，能较好的保留运动图像细节，运动物体显得比较平滑，不存在传统算法中的拖影、闪烁等现象。

Video de-noising method based on 3D wavelet transform and block context model

In this paper, a video de-noising method based on 3D wavelet transform and block Context model(3DWTBCM) is proposed. There is a strong correlation between the frames of video sequence, and this correlation can be utilized by de-noising in 3D wavelet domain. In order to take advantage of the characteristic of the coefficients in 3D wavelet domain and noise distribution, wavelet coefficients are partitioned into subblocks firstly according to local relativity of these coefficients, then Context model is used in the corresponding subblock. The wavelet coefficients in each block are divided into several parts according to their energy distribution by 3D Context model and each part is estimated by its independent energy distribution and a suitable threshold is got. Experimental results show that 3DWTBCM achieves better de-noising performance than hierarchical 2D de-noising methods and its PSNRs improves more than 1.5dB averagely compared with common 3D de-noising methods. In terms of visual quality, 3DWTBCM can effectively preserve the video detail when de-noising the wavelet coefficients and is especially adequate to process the video frames with acute movement and plenty of texture.