SAR image denoising based on wedgelet and dual-tree complex wavelet transform

Image denoising based on wedgelet transform and dual-tree complex wavelet transform (WDT-CWT) is proposed in this paper. Wedgelet transform is a new method that has a good performance in approximating edges. The limitation of wedgelet transform is that it smoothes the flat region excessively, leading to the loss of some texture features. To reduce this limitation, we employed dual-tree complex wavelet transform (DT-CWT) to improve the detection of texture information. Through a combination of the wedgelet transform and DT-CWT, we develop a detector of texture, edge and direction information. The experimental results show that WDT-CWT outperforms many traditional approaches both visualization and in terms of evaluation values.