Dual Way Residue Noise Thresholding along with feature preservation

作者：Goyal Bhawna^*; Dogra Ayush; Agrawal Sunil; Sohi B S

来源：Pattern Recognition Letters, 2017, 94: 194-201.

DOI：10.1016/j.patrec.2017.02.017

• NSST (Non-Sub Sampled Shearlet transform)
• Denoising
• MRI brain image
• NSP (Non-Separable pyramids)
• Thresholding
• BM3D
• Weighted bilateral filtering

It is extensively endorsed that preserving the intrinsic geometrical features of an image is essential while denoising it. With an aim to achieve this several directional image representations have been given in the recent literature. In this paper an efficient denoising scheme using an innovative method of calculating the residue image is being proposed. The residue image is further thresholded to remove excessive noise while recovering fine features and details. The recovered features are added to first stage of denoised image to enhance the information content and visual quality of denoised image. The proposed methodology DWRNT (Dual Way Residue Noise Thresholding) is a combination of various spatial and transforms domain methods. Extensive experimental results and investigations reveal that our rnethod can depict far better performance in terms of both subjective evaluation and objective evaluation than various other state-of-the-art image denoising techniques. In this way the proposed methodology is able to recover feature details of an image thereby reducing information loss along with efficient noise removal.

• 出版日期2017-7-15