scispace - formally typeset
Search or ask a question
Proceedings ArticleDOI

Image de-noising by non-local means algorithm

15 Apr 2013-pp 275-277
TL;DR: A non local method called as Non-Local Means estimates a noise-free pixel intensity as a weighted average of all pixel intensities in the image, and the weights are proportional to the similarity between the local neighbourhood of the pixel being processed and local neighbourhoods of surrounding pixels.
Abstract: Images are often corrupted with noise during acquisition, transmission and retrieval from storage media. So the need for efficient image de-noising methods has grown with the massive and easy production of digital images and movies. Furthermore, de-noising is often necessary as a pre-processing step in image compression, segmentation, recognition etc. Therefore, de-noising has been an important and widely studied problem in image processing and computer vision. Basically, the image de-noising methods are divided into two types: local and non-local. The methods that only exploit the spatial redundancy in local neighborhoods are referred as Local methods. The methods that estimate pixel intensity based on information from the whole image and thereby exploiting the presence of similar patterns and features in an image are referred as Non-Local. A non local method called as Non-Local Means[3] estimates a noise-free pixel intensity as a weighted average of all pixel intensities in the image, and the weights are proportional to the similarity between the local neighbourhood of the pixel being processed and local neighbourhoods of surrounding pixels. The method is quite spontaneous and powerful that results in comparable PSNR and visual quality to other non-local methods.
Citations
More filters
Journal ArticleDOI
TL;DR: A ripplet formulation of the total variation method for denoising images is proposed, known as a developed version of the curvelet transform and proposes a new tight frame with sparse representation for images with discontinuities along any type of boundaries.
Abstract: The main goals of denoising are to improve the signal-to-noise-ratio (SNR) and to preserve the informative features such as edges and textures. Aiming at reducing Gibbs-type artifacts, several researchers have combined wavelet-like transforms such as curvelets with total variation or diffusion methods. In this paper, a ripplet formulation of the total variation method for denoising images is proposed. The ripplet is known as a developed version of the curvelet transform and proposes a new tight frame with sparse representation for images with discontinuities along any type of boundaries. We manipulate the cost function of the total variation method, such that instead of minimizing the total variation of the noisy image, we minimize the total variation of a new image obtained from non-textured regions of ripplet subbands. To obtain these regions, ripplet coefficients are divided into textured regions and smooth ones using the twin support vector machine classifier. Numerical examples demonstrate that the proposed approach improves the image quality in terms of both subjective and objective inspections, compared with some other state-of-the-art denoising techniques.

26 citations


Cites methods from "Image de-noising by non-local means..."

  • ...Nonlocal methods exploit the presence of similar patches and features in an image [4, 12], and consequently every pixel intensity is estimated based on information from the whole image in these methods....

    [...]

Journal ArticleDOI
TL;DR: It is revealed that the oversegmentation is not only because of the irregular shapes of the particle images, which people are familiar with, but also because of some particles, such as ellipses, with more than one centre.
Abstract: Summary Oversegmentation is a major drawback of the morphological watershed algorithm. Here, we study and reveal that the oversegmentation is not only because of the irregular shapes of the particle images, which people are familiar with, but also because of some particles, such as ellipses, with more than one centre. A new parameter, the striping level, is introduced and the criterion for striping parameter is built to help find the right markers prior to segmentation. An adaptive striping watershed algorithm is established by applying a procedure, called the marker searching algorithm, to find the markers, which can effectively suppress the oversegmentation. The effectiveness of the proposed method is validated by analysing some typical particle images including the images of gold nanorod ensembles. Lay Description The morphological watershed algorithm is often used to process microscopic binary (black-and-white) images of overlapping particles, by which to automatically segment the connected particles into separated individual ones. However, oversegmentation is a major drawback of this method. That is, some individual particles may be further segmented into pieces unexpectedly. We find that the oversegmentation is not only due to the irregular shapes of the particle images, but also due to the existence of some particles (such as ellipses) with more than one center. In this work, we introduce a new parameter, called the striping level, and establish the Criterion for Striping Parameter to help find the right markers prior to segmentation. Then an adaptive striping watershed algorithm is established by applying a procedure, called the Marker Searching Algorithm, to find the markers, which can effectively suppress the oversegmentation. The proposed method is used to analyze some typical images, showing the effectiveness and reliability of the method.

5 citations


Cites methods from "Image de-noising by non-local means..."

  • ...…image can always be converted into binary image by some preprocessing techniques such as the image de-noising method (Kushwaha et al., 2012; Dixit & Phadke, 2013) and the image binaryzation process (Tsai & Lee, 2002; He et al., 2005) and those techniques are not directly related to the…...

    [...]

Journal ArticleDOI
TL;DR: Simulation results show that the proposed method "local spayed and optimized center pixel weights (LSOCPW) with non local means" has given the better performance when compared to the existing algorithms in terms of peak signal to noise ratio (PSNR) and mean square error (MSE).
Abstract: Now a day’s digital image processing applications are widely used in various fields such as medical, military, satellite, remote sensing and even web applications also. In any application image denoising is a challenging task because noise removal will increase the digital quality of an image and will improve the perceptual visual quality. In this paper we proposed a new method “local spayed and optimized center pixel weights (LSOCPW) with non local means” to improve the denoising performance of digital color image sequences. Simulation results show that the proposed method has given the better performance when compared to the existing algorithms in terms of peak signal to noise ratio (PSNR) and mean square error (MSE). DOI: http://dx.doi.org/10.11591/ijece.v4i5.6624

5 citations


Cites methods from "Image de-noising by non-local means..."

  • ...More recently, noise reduction techniques based on the “NON-LOCAL MEANS (NLM) had developed to improve the performance of denoising mechanism [1] [4] [5] [9] [15]....

    [...]

Journal ArticleDOI
TL;DR: The improved preclassification non local-means (IPNLM) is proposed for filtering grayscale images degraded with additive white Gaussian noise (AWGN) and reveals good results outperforming other filters based on NL-mean by balancing the tradeoff between the noise suppression, detail preservation, and processing time.
Abstract: In this paper, we develop an extensive research on different types of grayscale images applying standard non local (NL)-means algorithm on different search and patch windows sizes to obtain optimal parameters where the values of criteria peak signal-to-noise ratio (PSNR), mean absolute error (MAE), and structural similarity index (SSIM) would be the best possible. The research shows quantitatively the importance on the appropriate selection of the windows sizes used during the filtering process. Based on the optimal parameters of the standard NL-means, we propose the improved preclassification non local-means (IPNLM) for filtering grayscale images degraded with additive white Gaussian noise (AWGN). The proposal uses a descriptors evaluation for each search window in the noisy image to apply statistical neighborhood preclassification respect to the homogeneity of each window to distinguish whether the current noisy pixel is in a homogeneous region or it is in an edge object region. Also, two thresholds based on the standard deviation of the local region in the noisy image are proposed to classify the pixels and perform a filtering level degree providing a commitment between the image denoising and the processing time. The proposal IPNLM reveals good results outperforming other filters based on NL-means by balancing the tradeoff between the noise suppression, detail preservation, and processing time. Experimental results demonstrate that IPNLM algorithm can reduce considerably the processing time from 8 through 15 times in comparison with the standard NL-means and other analyzed filters.

3 citations

Book ChapterDOI
19 Jun 2015
TL;DR: A fast NLM denoising algorithm which can product comparable or better result with less computation time than the traditional NLM methods is proposed.
Abstract: Non-local means (NLM) is a powerful denoising algorithm that can protect texture effectively. However, the computational complexity of this method is so high that it is difficult to be widely applied in real-time systems. In this paper, we propose a fast NLM denoising algorithm which can product comparable or better result with less computation time than the traditional NLM methods. Some experimental results are provided to demonstrate the superiority of the proposed method.

1 citations

References
More filters
Proceedings ArticleDOI
20 Jun 2005
TL;DR: A new measure, the method noise, is proposed, to evaluate and compare the performance of digital image denoising methods, and a new algorithm, the nonlocal means (NL-means), based on a nonlocal averaging of all pixels in the image is proposed.
Abstract: We propose a new measure, the method noise, to evaluate and compare the performance of digital image denoising methods. We first compute and analyze this method noise for a wide class of denoising algorithms, namely the local smoothing filters. Second, we propose a new algorithm, the nonlocal means (NL-means), based on a nonlocal averaging of all pixels in the image. Finally, we present some experiments comparing the NL-means algorithm and the local smoothing filters.

6,804 citations


"Image de-noising by non-local means..." refers background in this paper

  • ...Due to the loss of image details, Baudes A., Coll B., Morel J.M. have developed the Non-Local Means (NLMeans) algorithm [3]....

    [...]

  • ...BASIC NL-MEANS ALGORITHM [3], PSEUDO CODE, NOISE MODEL AND POST PROCESSING FILTER [1] A. Basic NL-Means algorithm The self-similarity assumption can be exploited to de-noise an image....

    [...]

Proceedings ArticleDOI
03 Dec 2010
TL;DR: The proposed technique invokes twice, in the first and the third stages, the well-known Non-Local Means method for spatial and temporal denoising: it is well adapted to the application, leading to the definition of a novel NLM tool.
Abstract: The paper presents a novel three-stage algorithm for very-low-light video denoising and enhancement. The proposed technique invokes twice, in the first and the third stages, the well-known Non-Local Means (NLM) method for spatial and temporal denoising: it is well adapted to the application, leading to the definition of a novel NLM tool. The intermediate stage performs a custom tone adjustment specifically aimed at enlarging the dynamic range of very dark videos. The overall approach transforms very dark videos into more watchable ones, effectively reduces very high noise, and all in all, produces high quality restored image sequences outperforming the recent state-of-the-art results. Additionally, the first and third stages can be combined as a two-step filtering scheme for normal-light videos: the novel denoising solution achieves a heavy noise removal, while reducing motion blur artifacts and preserving image details.

27 citations