An Analysis and Implementation of the BM3D Image Denoising Method

Lebrun, Marc

doi:10.5201/ipol.2012.l-bm3d

Cited by 310 publications

(204 citation statements)

References 13 publications

Supporting

Mentioning

201

Contrasting

Unclassified

Order By: Relevance

“…At the time of writing only two implementations of BM3D algorithm were publicly available, the original design by Dabov et al [1] and an implementation created using C++ by Lebrun [34]. The source code for the original design was not available, and therefore the presented implementations could not be fully compared.…”

Section: Resultsmentioning

confidence: 99%

“…Five different implementations were compared with each other having the same filter profile in use in each test case. These five implementations are referred as Lebrun [34], Original by Dabov et al [1], OpenCL, CUDA and Mobile CUDA. The test execution times were averaged from five sequential test runs.…”

Section: Resultsmentioning

confidence: 99%

“…Also removing the layering by serializing the write operations was not an option when seeking a true parallel solution. The second method is similar to the parallelization used in the reference implementation by Lebrun [34], and can be seen in Figure 12. In the method, the memory dependency pattern of many-to-many in the algorithm is redundantly reduced to be type of many-to-one, which eliminates the need of writing intermediate results thoroughly.…”

Section: Parallelizationmentioning

confidence: 98%

“…This program, created by Lebrun [34], is done using standard C++ and is compilable on various operating systems with minor changes. The program executes only on host machine CPU, but can be compiled to use multi-threading with OpenMP library [35].…”

Section: Reference Implementationsmentioning

confidence: 99%

See 3 more Smart Citations

BM3D image denoising using heterogeneous computing platforms

Sarjanoja

Boutellier

Hannuksela

2015

2015 Conference on Design and Architectures for Signal and Image Processing (DASIP)

View full text Add to dashboard Cite

Noise reduction is one of the most fundamental digital image processing problems, and is often designed to be solved at an early stage of the image processing path. Noise appears on the images in many different ways, and it is inevitable. In general, various image processing algorithms perform better if their input is as error-free as possible. In order to keep the processing delays small in different computing platforms, it is important that the noise reduction is performed swiftly.The recent progress in the entertainment industry has led to major improvements in the computing capabilities of graphics cards. Today, graphics circuits consist of several hundreds or even thousands of computing units. Using these computing units for general-purpose computation is possible with OpenCL and CUDA programming interfaces. In applications where the processed data is relatively independent, using parallel computing units may increase the performance significantly. Graphics chips enabled with general-purpose computation capabilities are becoming more common also in mobile devices. In addition, photography has never been as popular as it is nowadays by using mobile devices.This thesis aims to implement the calculation of the state-of-the-art technology used in noise reduction, block-matching and three-dimensional filtering (BM3D), to be executed in heterogeneous computing environments. This study evaluates the performance of the presented implementations by making comparisons with existing implementations. The presented implementations achieve significant benefits from the use of parallel computing devices. At the same time the comparisons illustrate general problems in the utilization of using massively parallel processing for the calculation of complex imaging algorithms. Keywords TIIVISTELMÄKohinanpoisto on yksi keskeisimmistä digitaaliseen kuvankäsittelyyn liittyvistä ongelmista, joka useimmiten pyritään ratkaisemaan jo signaalinkäsittelyvuon varhaisessa vaiheessa. Kohinaa ilmestyy kuviin monella eri tavalla ja sen esiintyminen on väistämätöntä. Useat kuvankäsittelyalgoritmit toimivat paremmin, jos niiden syöte on valmiiksi mahdollisimman virheetöntä käsiteltäväksi. Jotta kuvankäsittelyviiveet pysyisivät pieninä eri laskentaalustoilla, on tärkeää että myös kohinanpoisto suoritetaan nopeasti.Viihdeteollisuuden kehityksen myötä näytönohjaimien laskentateho on moninkertaistunut. Nykyisin näytönohjainpiirit koostuvat useista sadoista tai jopa tuhansista laskentayksiköistä. Näiden laskentayksiköiden käyttäminen yleiskäyttöiseen laskentaan on mahdollista OpenCL-ja CUDAohjelmointirajapinnoilla.Rinnakkaislaskenta usealla laskentayksiköllä mahdollistaa suuria suorituskyvyn parannuksia käyttökohteissa, joissa käsiteltävä tieto on toisistaan riippumatonta tai löyhästi riippuvaista. Näytönohjainpiirien käyttö yleisessä laskennassa on yleistymässä myös mobiililaitteissa. Lisäksi valokuvaaminen on nykypäivänä suosituinta juuri mobiililaitteilla.Tämä diplomityö pyrkii selvittämään viimeisimmän kohinanpoistoon käytettävän tekni...

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Parallelizationmentioning

confidence: 98%

Section: Reference Implementationsmentioning

confidence: 99%

See 2 more Smart Citations

BM3D image denoising using heterogeneous computing platforms

Sarjanoja

Boutellier

Hannuksela

2015

2015 Conference on Design and Architectures for Signal and Image Processing (DASIP)

View full text Add to dashboard Cite

show abstract

“…Internal denoising algorithms use larger patches and search-windows at higher noise levels [13][14][15][16], improving reconstructed quality. Extensions to the NLM algorithm proposed adaptive spatial support for superior results, by classifying patches as textured or smooth by edge-detection and morphological operations [17], or by clustering the SVD (singular-value decomposition) of the blocks' gradient fields [18], allowing spatial adaptation for each block.…”

Section: Related Workmentioning

confidence: 99%

A largest matching area approach to image denoising

Gaston

Crookes

2016

2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Deep learning based automated detection of intraretinal cystoid fluid

Ahmed

Panhwar

Baqai

et al. 2021

Int J Imaging Syst Tech

View full text Add to dashboard Cite

The diversified ocular disorders in which cystoid macular edema (CME) occurs, are strongly associated with the vision loss. Optical coherence tomography (OCT) scans that allow screening of the retina, contain artifacts including blur‐edges, speckle noise, and so forth, which create difficulty in identifying retinal fluid. In this work, major image preprocessing techniques such as minimum filtering, block‐matching and 3D filtering, and Richardson–Lucy deconvolution method are applied to minimize noise and other degradation effects from OCT scans, ensured by maintaining image quality assessment scores, that is, blind‐less image spatial quality evaluator and sharpness estimation score. Furthermore, this work develops an automated method based on deep learning (DL) to detect the presence and progression of retinal fluid, that is, CME and its quantification. Specifically, U‐net model with efficient tuning of hyperparameters is proposed to obtain optimal results. The algorithm's performance is evaluated for accurate identification of fluid localization in the case of diabetic macular edema (DME), having CME regions, and compared against the manual segmentation of fluid by experts, which are highly correlated. This enhanced DL method achieves better performance than the algorithms previously reported. The proposed algorithm is evaluated on several performance metrics during training and reports the best score of 99.81, 0.50, 83.34, 86.56, 80.89, and 99.48% for model accuracy, binary cross‐entropy loss, Dice coefficient (DC), precision, recall, and area under the ROC‐curve, respectively. Additionally, fourfold cross‐validation technique is also applied and an average DC score of 84.00% is obtained. This approach can be helpful to prevent vision loss.

show abstract

An Analysis and Implementation of the BM3D Image Denoising Method

Cited by 310 publications

References 13 publications

BM3D image denoising using heterogeneous computing platforms

BM3D image denoising using heterogeneous computing platforms

A largest matching area approach to image denoising

Deep learning based automated detection of intraretinal cystoid fluid

Contact Info

Product

Resources

About