A Novel Image Denosing Scheme via Combining Dual-tree Complex Wavelet Transform and Bandelets

In this paper, a novel image denosing scheme is proposed by applying 2D dual-tree complex wavelet transform (DTCWT) to the second bandelet transform. Compared with traditional discrete wavelet transform (DWT), the DTCWT has nearly shift invariant and directionally selective in two and higher dimensions important properties, which are suitable for image denoising. The bandelet transform has offer an asymptotically optimal representation for geometric images, but it lacks translation invariant. Firstly, the DTCWT is employed to obtain subands, and then bandeletization is implemented in each subband. During finding the best geometrical flow and optimal quadtree segment, the cost term of Lagrangian is confirmed by Bayes estimator. At last, Bayes soft-threshold shrinkage denoising in bandelet transform domain is implemented. The experiments show that the proposed denoising method has better results than several DTCWT-based methods for image with rich directional features. The denoised images recovered by the proposed scheme are visually more appealing.