A Novel Image Denoising Algorithm Based on Non-subsampled Contourlet Transform and Modified NLM
Autor: | Xiaoli Lin, Hua-yong Yang |
---|---|
Rok vydání: | 2018 |
Předmět: |
Computer science
Noise reduction 020206 networking & telecommunications 02 engineering and technology Function (mathematics) Contourlet Image (mathematics) symbols.namesake Computer Science::Computer Vision and Pattern Recognition 0202 electrical engineering electronic engineering information engineering Gaussian function symbols 020201 artificial intelligence & image processing Noise (video) Image denoising Algorithm |
Zdroj: | Intelligent Computing Methodologies ISBN: 9783319959566 ICIC (3) |
DOI: | 10.1007/978-3-319-95957-3_71 |
Popis: | A novel image denoising algorithm based on non-subsampled contourlet transform (NSCT) and modified non-local mean (NLM) is proposed. First, we utilize NSCT to decompose the images to obtain the high frequency coefficients. Second, the high frequency coefficients are used for modified NLM denoising. Finally, the NLM weight values are calculated by modified bisquare function instead of Gaussian kernel function of the traditional NLM, and each noise coefficient is corrected to get the denoised image. According to results of the simulation experiment, the denoising results of the proposed algorithm obtain higher peak signal-to-noise ratio (PSNR) and better retains structural information of image in subjective vision. |
Databáze: | OpenAIRE |
Externí odkaz: |