Fast blockwise SURE shrinkage for image denoising

Wu, Yue; Tracey, Brian H.; Natarajan, Prem; Noonan, Joseph P.

doi:10.1016/j.sigpro.2014.01.007

Cited by 9 publications

(4 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Authors proposed a denoising algorithm parameterized as a linear expansion of thresholds [2]. Optimization is carried out using Stein's Unbiased Risk Estimator (SURE) [33,34,35]. The thresholding function is point wise and wavelet based.…”

Section: Related Workmentioning

confidence: 99%

Multispectral image denoising with optimized vector non-local mean filter

Said

Hadjidj

Melkemi

et al. 2016

Digital Signal Processing

View full text Add to dashboard Cite

Nowadays, many applications rely on images of high quality to ensure good performance in conducting their tasks. However, noise goes against this objective as it is an unavoidable issue in most applications. Therefore, it is essential to develop techniques to attenuate the impact of noise, while maintaining the integrity of relevant information in images. We propose in this work to extend the application of the Non-Local Means filter (NLM) to the vector case and apply it for denoising multispectral images. The objective is to benefit from the additional information brought by multispectral imaging systems. The NLM filter exploits the redundancy of information in an image to remove noise. A restored pixel is a weighted average of all pixels in the image. In our contribution, we propose an optimization framework where we dynamically fine tune the NLM filter parameters and attenuate its computational complexity by considering only pixels which are most similar to each other in computing a restored pixel. Filter parameters are optimized using Stein's Unbiased Risk Estimator (SURE) rather than using ad hoc means. Experiments have been conducted on multispectral images corrupted with additive white Gaussian noise and PSNR and similarity comparison with other approaches are provided to illustrate the efficiency of our approach in terms of both denoising performance and computation complexity.

show abstract

Section: Related Workmentioning

confidence: 99%

Multispectral image denoising with optimized vector non-local mean filter

Said

Hadjidj

Melkemi

et al. 2016

Digital Signal Processing

View full text Add to dashboard Cite

show abstract

“…An image denoising algorithm based on non-local means (NLM) is proposed in [16], where the NLM parameters are optimized using SURE. Notable denoising algorithms that aim to optimize the SURE objective include wavelet-domain multivariate shrinkage [17], local affine transform for image denoising [18], SURE-optimized blockwise shrinkage for image denoising [19], SURE-optimized Savitzky-Golay filter [20], etc. The SURE approach has also found applications in image deconvolution [21] and compressive sensing [22].…”

Section: Introductionmentioning

confidence: 99%

Signal denoising using the minimum-probability-of-error criterion

Sadasivan

Mukherjee

Seelamantula

2020

SIP

View full text Add to dashboard Cite

We address the problem of signal denoising via transform-domain shrinkage based on a novel risk criterion called the minimum probability of error (MPE), which measures the probability that the estimated parameter lies outside an -neighborhood of the actual value. However, the MPE, similar to the meansquared error (MSE), depends on the ground-truth parameter, and has to be estimated from the noisy observations. We consider linear shrinkage-based denoising functions, wherein the optimum shrinkage parameter is obtained by minimizing an estimate of the MPE. When the probability of error is integrated over , it leads to the expected 1 distortion. The proposed MPE and 1 distortion formulations are applicable to various noise distributions by invoking a Gaussian mixture model approximation. Within the realm of MPE, we also develop an extension of the transform-domain shrinkage by grouping transform coefficients, resulting in subband shrinkage. The denoising performance obtained within the proposed framework is shown to be better than that obtained using the minimum MSEbased approaches formulated within Stein's unbiased risk estimation (SURE) framework, especially in the low measurement signal-tonoise ratio (SNR) regime. Performance comparison with three stateof-the-art denoising algorithms, carried out on electrocardiogram signals and two test signals taken from the Wavelab toolbox, exhibits that the MPE framework results in consistent SNR gains for input SNRs below 5 dB.Index Terms-Minimum probability of error, shrinkage estimator, risk estimation, transform-domain shrinkage, subband shrinkage, expected 1 distortion, Gaussian mixture model.

show abstract

“…The extraction of noise performs a crucial part in an image capture and handling scheme, which can be categorized as Gaussian sound, balance noise, and impulse noise 1 . Random black spots, combined with white dots resulting as pulse noise, make a picture that not only corrupts real picture data, but also severely impacts the visual impact of an image 2 . The decrease of impulse noise is, therefore, significant for software analysis and image processing.…”

Section: Introductionmentioning

confidence: 99%

A new image restoration approach by combining empirical wavelet transform and total variation using chaotic squirrel search optimization

Kumar

Narasimhulu²,

Prasad

2020

Int J Numerical Modelling

View full text Add to dashboard Cite

Image noise is the random variation of brightness or color information in the images. Noise can enter the picture during capturing or transmitting. Different linear and nonlinear methods have been implemented to remove the noise from an image, but transformation of wavelets is becoming increasingly important. We propose a fundamental concept for disintegrating the input bubble, sound, and movement signal of the object/camera into appropriate coefficients by transforming the empirical wavelet and making full changes. A new messy screen search strategy is used to assess the variability at each pixel of the decomposed picture. Empirical wavelet is a kind of wavelet tailored to the data being handled. This is an adaptive technique for decomposing the signal or picture to a set of sections recognized as phases. This wavelet transform is relevant for de‐noising, decompression, etc. Chaotic squirrel search optimizers imitate the vibrant feeding behavior and efficiency of northern flying squirrels, called gliding. Gliding is a powerful system used for lengthy ranges by tiny mammals. Compared to other wavelet transformations, the median peak signal for noise ratio (PSNR) enhancement obtained by EWT is 73%. Numerical tests indicate that our proposed method could be able to substantially enhance restored pictures' efficiency and achieve greater SNR and structural similarity index scores likened to the present state of the art methodologies.

show abstract

Fast blockwise SURE shrinkage for image denoising

Cited by 9 publications

References 42 publications

Multispectral image denoising with optimized vector non-local mean filter

Multispectral image denoising with optimized vector non-local mean filter

Signal denoising using the minimum-probability-of-error criterion

A new image restoration approach by combining empirical wavelet transform and total variation using chaotic squirrel search optimization

Contact Info

Product

Resources

About