Fast nonlocal means for image denoising

Bilcu, Radu Ciprian; Vehviläinen, Markku

doi:10.1117/12.695079

Cited by 20 publications

(13 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to overcome these limits, various attempts have been proposed in the literature. For example, some works make NL-means faster by reducing the search domain eliminating insignificant pixels or by optimizing the computation of the similarity measure [1,13,23,24]. Most of them also use a different domain where to catch similarities.…”

Section: Introductionmentioning

confidence: 97%

Time-Scale Similarities for Robust Image De-noising

Bruni

Vitulano

2011

J Math Imaging Vis

View full text Add to dashboard Cite

This paper presents a novel image denoising algorithm, namely Atomic Non Local Means (ANL-means), that looks for similarities in the time-scale domain. To this aim, wavelet details are approximated by linear combinations of predefined atoms, whose centers of mass trace trajectories in the time-scale plane (from fine to coarse scales). These trajectories depend on the mutual distance between not isolated singularities, their different decay along scales and their amplitude ratio. These three parameters have proved to be useful in catching image self-similarities and in the implementation of a robust NL-means based denoising algorithm. ANL-means is able to reach and often outperform the most powerful and recent NL-means based de-noising schemes in terms of both mean square error and visual quality

show abstract

Section: Introductionmentioning

confidence: 97%

Time-Scale Similarities for Robust Image De-noising

Bruni

Vitulano

2011

J Math Imaging Vis

View full text Add to dashboard Cite

show abstract

“…Although the algorithm can achieve very good denoising effect, but a large amount of calculation and high time complexity make the application of this algorithm limited. The literatures [9][10][11] reduce the number of weighted average pixels through the pre-classification algorithm that can effectively reduce the computation, but probably create artifacts in the image if the similarity parameters are selected improperly.…”

Section: Introductionmentioning

confidence: 99%

Non-local means denoising algorithm accelerated by GPU

2009

View full text Add to dashboard Cite

On the basis of studying Non-Local Means (NLM) denoising algorithm and its pixel-wise processing algorithm in Graphics Processing Unit (GPU), a whole image accumulation algorithm of NLM denoising algorithm based on GPU is proposed. The number of dynamic instructions of fragment shader is effectively reduced by redesigning the data structure and processing flow, that make the algorithm suitable to the graphic cards supported Shader Model 3.0 and/or Shader Model 4.0, and so enhance the versatility of the algorithm. Then the continuous and parallel processing method for 4 gray images based on Multiple Render Target (MRT) and double Frame Buffer Object (FBO) is proposed, and the whole processing flow with GPU is presented. The experimental results of both simulative and practical gray images show that the proposed method can achieve a speedup of 45 times while remaining the same accuracy.

show abstract

“…Although the algorithm can achieve very good denoising results, but there are the existence of two shortcomings: 1) Similarity calculation of local areas for each pixel would take the high computational complexity; 2) Denoising effect and the choice of filtering parameters are closely related. Computing speed for the first problem, the literatures [11][12][13] made the pre-classification to reduce the number of weighted average pixels that can effectively reduce the computation, but how to improve the denoising effect of NLM denoising algorithm in practical applications is still need to be solved.…”

Section: Introduction (Heading 1)mentioning

confidence: 99%

Adaptive Non-local Means Denoising Algorithm for Cone-Beam Computed Tomography Projection Images

Huang

Zhang

Wang

et al. 2009

2009 Fifth International Conference on Image and Graphics

View full text Add to dashboard Cite

Aiming at the difficulty of choosing filtering parameters of Non-Local Means (NLM) denoising algorithm for Cone-Beam Computed Tomography (CBCT) projection images, an adaptive NLM denoising algorithm is proposed. First, a simplified noise estimation method based on subblock division is proposed to replace the approach of designating the background region, which can adaptively identify the background region and estimate the noise level when the background region is changed obviously. And then an algorithm of adaptively obtaining filter strengths of projection images is proposed by analyzing the impact of filter strength to filtering result in NLM denoising algorithm, in which the filter strength is adjusted according to the characteristics of current image and the noises in all projection images are suppressed to a similar level. The experimental result shows that the proposed algorithm can accurately estimate the noise level of projection image and markedly improve the slice image quality.

show abstract

Fast nonlocal means for image denoising

Cited by 20 publications

References 12 publications

Time-Scale Similarities for Robust Image De-noising

Time-Scale Similarities for Robust Image De-noising

Non-local means denoising algorithm accelerated by GPU

Adaptive Non-local Means Denoising Algorithm for Cone-Beam Computed Tomography Projection Images

Contact Info

Product

Resources

About