Clustering-based natural image denoising using dictionary learning approach in wavelet domain

Khmag, Asem; Ramli, Abd Rahman; Kamarudin, Noraziahtulhidayu

doi:10.1007/s00500-018-3438-9

Cited by 38 publications

(24 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the future, we will extend the proposed method that can remove other types of noise such as random-valued noise (RVIN) and mixed noise [39][40]. Also, a combination of the proposed method with the traditional learning [44] or the deep learning technique [41][42] will be also studied.…”

Section: Discussionmentioning

confidence: 99%

Image Noise Removal Method Based on Thresholding and Regularization Techniques

et al. 2022

View full text Add to dashboard Cite

In this article, a salt and pepper noise (SPN) removal method is proposed based on thresholding and regularization techniques. The proposed method utilizes the ability to remove noise from an image denoising model based on Total Variation (TV) regularization and characteristics of SPN. First, a technique based on the characteristic of SPN is proposed to detect noisy pixels. Second, a modified TV regularization-based method is applied to restore the above noisy pixels. In addition, numerical implementation of the model based on the Nesterov optimal method is also provided. Five test cases with various noise levels for a large natural image dataset were studied in the experiments. The peak signal-to-noise ratio and structural similarity metrics were employed to assess the image quality after denoising. The experimental results indicated that the proposed method removes SPN remarkably and outperforms state-of-the-art image denoising methods for SPN.

show abstract

Section: Discussionmentioning

confidence: 99%

Image Noise Removal Method Based on Thresholding and Regularization Techniques

et al. 2022

View full text Add to dashboard Cite

show abstract

“…These noise signals generally exhibit flat broadband characteristics, so they can be considered as additional Gaussian white noise [24]. In this way, the image received by the terminal is the sum of the original signals and the additional noises [25].…”

Section: Wavelet Threshold Denoising Methodsmentioning

confidence: 99%

An Adaptive Threshold Method for WMSN Image Denoising

Zhi¹,

Wang²

2020

RCES

View full text Add to dashboard Cite

Though the wavelet threshold algorithm has been demonstrated to be a very effective tool to denoise the images with low levels of noise, it usually losses the power to well preserve meaningful details in images. To this end, this paper proposed an adaptive threshold denoising method for images that are usually interfered by noise on the Wireless Multimedia Sensor Network (WMSN). First, a scale parameter equation was defined according to different sub-band characteristics after the images were subject to wavelet decomposition, so as to determine the adaptive optimal threshold suitable for each scale level; Second, a new derivable threshold function was designed in this paper; Third, after comparison, proper wavelet basis function was selected for image denoising accordingly. Moreover, the test results on several test images proved the superiority of the proposed method over some classical methods in terms of PSNR and MSE.

show abstract

“…The image obtained by wavelet decomposition is a wavelet series, which is the result of a linear filtering. If the wavelet filter has a linear phase or generalized linear phase, it is possible to fully reconstruct the original image [13][14][15]. Thus, no orthogonal wavelet except for Haar wavelet boasts the capability of full reconstruction.…”

Section: Influence Of Orthogonality and Bi-orthogonality On Denoisingmentioning

confidence: 99%

Wavelet-Based Self-adaptive Hierarchical Thresholding Algorithm and Its Application in Image Denoising

Zhang¹,

Zhu²,

Lin³

2019

View full text Add to dashboard Cite

This paper attempts to construct a suitable wavelet for image denoising based on wavelet thresholding algorithm. First, the author discussed how image thresholding is affected by the wavelet orthogonality and bi-orthogonality, the features of vanishing moments and the odd or even symmetry of the decomposition end filter. The discussion shows that the most desirable wavelet for image denoising is the biorthogonal wavelet, in which the decomposition end filter has zero point even symmetry, the low-pass decomposition enjoys a wide support interval, and the high-pass decomposition filter has a short support and attenuates fast. On this basis, three zero point even symmetric biorthogonal wavelets with different vanishing moment features were developed through the parametric construction of fixed-length tightly-supported (FLTS) biorthogonal wavelet, and a self-adaptive hierarchical thresholding algorithm was designed. The simulation results show that the developed wavelets have excellent denoising ability and enhance the images with rich details. Coupled with the self-adaptive hierarchical thresholding algorithm, these wavelets can effectively improve the image quality.

show abstract

Clustering-based natural image denoising using dictionary learning approach in wavelet domain

Cited by 38 publications

References 33 publications

Image Noise Removal Method Based on Thresholding and Regularization Techniques

Image Noise Removal Method Based on Thresholding and Regularization Techniques

An Adaptive Threshold Method for WMSN Image Denoising

Wavelet-Based Self-adaptive Hierarchical Thresholding Algorithm and Its Application in Image Denoising

Contact Info

Product

Resources

About