Comparative performance analysis of various wavelet and nonlocal means based approaches for image denoising

详细信息    查看全文

• 作者：Karamjeet Singh ; ^{karampup@gmail.com} ; Sukhjeet Kaur Ranade ^{mrsdjjr@gmail.com} ; Chandan Singh ^{chandan.csp@gmail.com}
• 关键词：Nonlocal means (NLM) ; Discrete cosine transform (DCT) ; Computational complexity ; Weighted denoising
• 刊名：Optik - International Journal for Light and Electron Optics
• 出版年：2017
• 出版时间：February 2017
• 年：2017
• 卷：131
• 期：Complete
• 页码：423-437
• 全文大小：5349 K
• 卷排序：131

文摘

This paper presents the comparative performance analysis of popular nonlocal means-based and wavelet-based image denoising approaches for Gaussian noise reduction and propose a novel method for computing weights for patch matching using DCT coefficients to enhance the performance of the existing nonlocal means-based approach for image denoising. Various quantitative and qualitative parameters such as peak signal-to-noise ratio, mean structural similarity index and method noise have been used for comparing the existing NLM and wavelet based denoising approaches and establish the preeminence of the proposed approach in terms of the quality of the denoised image produced and time taken.