Impulsive Noise Removal with an Adaptive Weighted Arithmetic Mean Operator for Any Noise Density

González-Hidalgo, Manuel; Massanet, Sebastià; Mir, Arnau; Ruiz-Aguilera, Daniel

doi:10.3390/app11020560

Cited by 8 publications

(7 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The method of completing the square, such as the derivation in Appendix C, can be used to transform Equation (8) to…”

Section: Q 1 Sub-problemmentioning

confidence: 99%

“…Common methods for removing salt and pepper noise include median-filter-based methods [7][8][9][10][11][12][13][14][15][16], total variation (TV)-based methods [17][18][19][20][21], Euler's elastica variational method [22], deep-learning-based methods [23][24][25], and wavelet transform (WT)-based methods [26]. Median-filter-based methods replace the noise pixel with the median value of adjacent pixels.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Salt and pepper noise removal method based on stationary Framelet transform with non‐convex sparsity regularization

Chen

Huang

Wang

et al. 2022

IET Image Processing

View full text Add to dashboard Cite

Salt and pepper noise occurs randomly and causes image degradation. Numerous denoising methods have been proposed to suppress this noise. However, existing methods have two main limitations. First, noise characteristics, such as noise location information and sparsity, are often described inaccurately or even ignored. Second, many existing methods separate the contaminated image into a recovered image and a noise part, leading to the recovery of an image with unsatisfactory smooth and detailed parts. In this study, the authors introduce a noise detection strategy to determine the position of the noise and employ a non-convex sparsity regularization depicted by l p quasi-norm to describe the sparsity of the noise, thereby addressing the first limitation. We adopt the morphological component analysis framework with stationary Framelet transform to decompose the processed image into the cartoon, texture, and noise parts to resolve the second limitation. Then, the proposed model is applied by using the alternating direction method of multipliers (ADMM). Finally, experiments are conducted to verify the proposed method and compare it with some current state-of-the-art denoising methods. The experimental results show that the proposed method can remove salt and pepper noise while preserving the details of the processed image and outperforming some state-of-the-art methods.

show abstract

“…The method of completing the square, such as the derivation in Appendix C, can be used to transform Equation (8) to…”

Section: Q 1 Sub-problemmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Salt and pepper noise removal method based on stationary Framelet transform with non‐convex sparsity regularization

Chen

Huang

Wang

et al. 2022

IET Image Processing

View full text Add to dashboard Cite

show abstract

“…For RVIN, the recommended KMDCIFF have a similarity of over 80% at a noise density of 40% and a similarity of over 66% at an extreme noise density of 80%. [46] 0.9305 0.8349 0.8176 EMDABF [19] 0.9445 0.8525 0.8245 VEERA [17] 0.9441 0.8698 0.8269 ASWM [20] 0.9874 0.9702 0.9212 AFMF [11] 0.9749 0.9624 0.9215 DAMRmF [12] 0.9854 0.9701 0.9198 DBNATMTF [13] 0.9548 0.9343 0.8909 KMDCIFF 0.9862 0.9748 0.9288 RVIN ANN [21] 0.8747 0.7987 0.7125 SAFE [45] 0.9199 0.8461 0.7573 EMDABF [19] 0.9199 0.8194 0.7539 JIN [18] 0.9268 0.8295 0.7749 CHEN [22] 0.8916 0.7931 0.7321 DBMF [14] 0.8342 0.7724 0.7002 ZHU [15] 0.8648 0.7995 0.7254 AWAM [16] 0.9146 0.8254 0.7712 KMDCIFF 0.9178 0.8621 0.8045 Barbara FVIN SVMF [47] 0.8614 0.7865 0.7123 AFSWMF [46] 0.9047 0.8258 0.6903 EMDABF [19] 0.9099 0.8318 0.6927 VEERA [17] 0.9089 0.8343 0.7113 ASWM [20] 0.9798 0.9479 0.8254 AFMF [11] 0.9645 0.9358 0.8142 DAMRmF [12] 0.9541 0.9124 0.7548 SVMF [47] 0.9245 0.8752 0.7124 KMDCIFF 0.9771 0.9512 0.8355 RVIN ANN [21] 0.7458 0.7125 0.6125 SAFE [45] 0.8668 0.7362 0.6425 EMDABF [19] 0.7698 0.7249 0.6358 JIN [18] 0.7795 0.7459 0.6597 CHEN [22] 0.7549 0.7198 0.6216 DBMF [14] 0.7124 0.6584 0.5987 ZHU [15] 0.7654 0.7259 0.6458 AWAM [16] 0.7785 0.7452 0.6874 KMDCIFF 0.8459 0.7698 0.6987 Baboon FVIN SVMF [47] 0.8543 0.7737 0.6772 AFSWMF [46] 0.8816 0.8149 0.6826 EMDABF [19] 0.8963 0.8198 0.6785 VEERA [17] 0.8951 0.8214 0.6958 ASWM [20] 0.9788 0.9451 0.8245 AFMF…”

Section: Resultsmentioning

confidence: 99%

“…The observable noisy pixels are restored using a weighted mean filter during the removal step. A fuzzy based morphologically oriented filter [16] has recently been developed which was very impactful. Fuzzy morphological procedures and the weighted arithmetic average aggregation function were used in the work to propose a novel filter for acoustic signals.…”

Section: Introductionmentioning

confidence: 99%

A Neighborhood Impact Driven K-Medoid Clustering and Fuzzy Logic Blended Approach for High Density Impulse Noise Detection and Removal

Bandyopadhyay¹,

Deb²,

Chakraborty³

et al. 2022

View full text Add to dashboard Cite

In the field of image processing, removing impulse noise has been regarded as one of the most important tasks, primarily because of the noise pattern it presents. Existing filters used the effect of only those non-noisy pixels which were present inside the specified windows ignoring the effect of the non-noisy pixels present in the surrounding windows. So, the least distant non-noisy pixels in the present window as well as in the surrounding windows may have an influence on the present window's noisy pixels. Hence, considering the above factors, in this paper, a two-step technique named KMDCIFF (K-medoid clustering identified fuzzy filter) is proposed for removing impulse noise from digital images. In the proposed KMDCIFF algorithm, the first step is noise detection using K-medoid clustering, followed by a fuzzy logic-acquainted noise reduction strategy that utilizes the least distant local and non-local non-noisy pixels for removal operation. The detection process involves the application of K-medoid clustering on all 5×5 windows produced by centering each pixel of the considered image. In order to remove noise, a 7×7 window is constructed with each detected noisy pixel in the center. Analyzing the impact of the least distant local and non-local pixel on each noisy pixel, the same is replaced by an estimated pixel’s intensity value obtained from the most influential non-noisy pixels. KMDCIFF is evaluated using well-known metrics for diverse types of images. At a high noise density of 80%, KMDCIFF exhibited significant peak-signal-to-noise-ratios (PSNRs) of 26.97 dB and 29.67 dB and structural similarity indexes (SSIMs) of 0.8045 and 0.9288 on random and fixed valued impulse noise impacted Lena image, respectively. Comparing the results of the contemporary study to those of previous studies of a similar kind in this sector, the results are unswervingly astounding.

show abstract

“…U ovom radu razmatraće se so i biber šum, gde su pikseli pogodeni ekstremnim vrednostima koje se pojavljuju kao crne i bele tačkice. 1 Ovo istraživanje je delimično podržao Fond za nauku Republike Srbije, #GRANT br. 7632, projekat "Mathematical Methods in Image Processing under Uncertainty" MaMIPU i Pokrajinski sekretarijat za visoko obrazovanje i naučnoistraživački rad (AP Vojvodina, Republika Srbija) kroz projekat broj 142-451-3188/2023-01: "Primena agregacionih operatora u prepoznavanju lica".…”

Section: Uvodunclassified

Filter Za Uklanjanje Šuma Na Digitalnoj Slici Zasnovan Na Agregacionim Funkcijama

Ralević,

Kapor

2024

Mathematics in Engineering: Theory and Applications

View full text Add to dashboard Cite

U obradi slika neizbežan je problem pojave šuma u većini digitalnih slika. Najčešći uzroci za to su izvor snimanja, prenos slike ili nešto drugo. Rešenje ovog problema može biti implementacija algoritama koji su otporni na šum. Drugo rešenje je u predobradi slike korišćenjem filtera za uklanjanje šuma, dobijajući sliku bez šuma. Ovi filteri moraju tražiti ravnotežu izmedu uklanjanja zašumljenih piksela i npr. očuvanja ivica i tekstura koji postoje na slici. Tako su poslednjih godina predloženi neki filteri za uklanjanje so i biber šuma kao npr. filter baziran na ponderisanoj aritmetičkoj odnosno na ponderisanoj stepeno-korenskoj sredini kao i fazi matematičkoj morfologiji. U ovom istraživanju predložen je jedan filter sličnog tipa. Takode se vrši provera performansi predloženog filtera i on se poredi sa drugim odgovarajućim filtrima za različite nivoe šuma.

show abstract

Impulsive Noise Removal with an Adaptive Weighted Arithmetic Mean Operator for Any Noise Density

Cited by 8 publications

References 31 publications

Salt and pepper noise removal method based on stationary Framelet transform with non‐convex sparsity regularization

Salt and pepper noise removal method based on stationary Framelet transform with non‐convex sparsity regularization

A Neighborhood Impact Driven K-Medoid Clustering and Fuzzy Logic Blended Approach for High Density Impulse Noise Detection and Removal

Filter Za Uklanjanje Šuma Na Digitalnoj Slici Zasnovan Na Agregacionim Funkcijama

Contact Info

Product

Resources

About