小波域三维块匹配图像去噪

Image denoising by mixing wavelet transformation with sparse 3D collaborative filtering

提出了一种关于图像去噪的三维块匹配算法(BM3D算法)的改进算法。它不仅保留了三维块匹配算法好的性质,而且最大的优点是能大大减少计算量,缩短运算时间。算法包括三个步骤:首先,对含噪图像进行小波分解;其次,对小波分解后的高频分量用三维块匹配(BM3D)算法进行去噪处理;最后,用处理后的结果进行小波重构得到去噪图像。给出了该算法的详细实现过程,并把它与以前的三维块匹配算法进行了比较。结果表明,改进后的算法,不但保留了三维块匹配算法在去噪方面好的性质,而且大大减少了运算量。

An improved algorithm of image denoising by sparse 3D transform-domain collaborative filtering（BM3D） is proposed.It not only keeps the good performance of the BM3D algorithm,but also can reduce the computation time significantly.This paper realizes it using the three successive steps：Wavelet decomposition of an image,process in the wavelet transformation domain by using the BM3D method,and wavelet reconstruction.An efficient implementation of this algorithm is presented in full detail.Also the comparison of this improved algorithm with the BM3D approach is given.The experimental results demonstrate that this improved method achieves excellent denoising performance in terms of even higher signal-to-noise ratio,subjective visual quality and much less computation amount.