An improved image mixed noise removal algorithm based on super-resolution algorithm and CNN

Ding, Ling; Zhang, Huyin; Xiao, Jinsheng; Li, Bijun; Lu, Shejie; Norouzifard, Mohammad

doi:10.1007/s00521-018-3777-6

Cited by 10 publications

(9 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Second, we also found supportive results in the literature. For instance, in a noise-removing CNN model [45] reported by Ding et al, they started with a published super-resolution CNN model [46] and then replaced the last upsampling deconvolution layer in the supersolution CNN model with a dimension-enlargement layer without upsampling capabilities. Their new noise-removing CNN model was able to remove noise without upsampling.…”

Section: Discussion and Summarymentioning

confidence: 99%

“…Given the availability of an advanced ultrasound elastography simulation platform [38], [39], computer-simulated but realistic data of this kind can be used for training purposes. Now we have a better understanding the relationship between CNN-based super-resolution and noise removal [45], novel CNN models can be designed to improve the generator network accordingly.…”

Section: Discussion and Summarymentioning

confidence: 99%

See 1 more Smart Citation

An Application of Super-Resolution Generative Adversary Networks for Quasi-Static Ultrasound Strain Elastography: A Feasibility Study

He¹,

Peng²,

Yang³

et al. 2020

IEEE Access

View full text Add to dashboard Cite

In this work, a super-resolution approach based on generative adversary network (GAN) was used to interpolate (up-sample) ultrasound radio-frequency (RF) echo data along the lateral (perpendicular to the acoustic beam direction) direction before motion estimation. Our primary objective was to investigate the feasibility of using a GAN-based super-solution approach to improve lateral resolution in the RF data as a means of improving strain image quality in quasi-static ultrasound strain elastography (QUSE). Unlike natural scene photographs, axial (parallel to the acoustic beam direction) resolution is significantly higher than that of lateral resolution in ultrasound RF data. To better handle RF data, we first modified a superresolution generative adversary network (SRGAN) model developed by the computer vision community. We named the modified SRGAN model as super-resolution radio-frequency neural network (SRRFNN) model. Our preliminary experiments showed that, compared with axial strain elastograms obtained using the original ultrasound RF data, axial strain elastograms using ultrasound RF data up-sampled by the proposed SRRFNN model were improved. Based on the Wilcoxon rank-sum tests, such improvements were statistically significant (p < 0.05) for large deformation (3-5%). Also, the proposed SRRFNN model outperformed a commonly-used method (i.e. bi-cubic interpolation used in MATLAB [Mathworks Inc., MA, USA]) in terms of improving axial strain elastograms. We concluded that applying the proposed (SRRFNN) model was feasible and good-quality strain elastography data could be obtained in in vivo tumor-bearing breast ultrasound data. INDEX TERMS Generative adversarial network, motion tracking, super-resolution, quasi-static ultrasound strain elastography.

show abstract

Section: Discussion and Summarymentioning

confidence: 99%

Section: Discussion and Summarymentioning

confidence: 99%

An Application of Super-Resolution Generative Adversary Networks for Quasi-Static Ultrasound Strain Elastography: A Feasibility Study

He¹,

Peng²,

Yang³

et al. 2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…To learn the valid prior from the external datasets, deep learning based methods [22], [23] have been proposed for mixed noise removal. The merge of deep convolutional neural networks and variational model [22] has boosted the performance of mixed noise removal.…”

Section: Related Workmentioning

confidence: 99%

Mixed Noise Removal with Pareto Prior

Zhou¹,

Yu²,

Xia³

2020

Preprint

View full text Add to dashboard Cite

Denoising images contaminated by the mixture of additive white Gaussian noise (AWGN) and impulse noise (IN) is an essential but challenging problem. The presence of impulsive disturbances inevitably affects the distribution of noises and thus largely degrades the performance of traditional AWGN denoisers. Existing methods target to compensate the effects of IN by introducing a weighting matrix, which, however, is lack of proper priori and thus hard to be accurately estimated. To address this problem, we exploit the Pareto distribution as the priori of the weighting matrix, based on which an accurate and robust weight estimator is proposed for mixed noise removal. Particularly, a relatively small portion of pixels are assumed to be contaminated with IN, which should have weights with small values and then be penalized out. This phenomenon can be properly described by the Pareto distribution of type 1. Therefore, armed with the Pareto distribution, we formulate the problem of mixed noise removal in the Bayesian framework, where nonlocal self-similarity priori is further exploited by adopting nonlocal low rank approximation. Compared to existing methods, the proposed method can estimate the weighting matrix adaptively, accurately, and robust for different level of noises, thus can boost the denoising performance. Experimental results on widely used image datasets demonstrate the superiority of our proposed method to the state-of-the-arts.

show abstract

“…There was a drastic growth for generative models in the last few years. Substantial methods have been proposed to address the image super-resolution problems [1] [16]. The main concept of Generator adversarial network (GAN) [15] states an adversarial game between two networks; Ddiscriminator network and G-generator network.…”

Section: Related Workmentioning

confidence: 99%

Perceptual image quality using dual generative adversarial network

Zareapoor

Zhou

Yang

2019

Neural Comput & Applic

View full text Add to dashboard Cite

Generative adversarial networks have received a remarkable success in many computer vision applications for their ability to learn from complex data distribution. In particular, they are capable to generate realistic images from latent space with a simple and intuitive structure. The main focus of existing models has been on improving the performance, however there is a little attention to make a robust model. In this paper, we investigate solutions to the super resolution problems-in particular perceptual quality-by proposing a robust GAN. The proposed model unlike the standard GAN employs two generators and two discriminators in which, a discriminator determines the samples are from real data or generated one, while another discriminator acts as classifier to return the wrong samples to its corresponding generators. Generators learn a mixture of many distribution from prior to the complex distribution. This new methodology is trained with the Feature Matching loss and allows us to return the wrong samples to the corresponding generators, in order to regenerated the real-look samples. Experimental results in various datasets show the superiority of the proposed model compared to the state of the art methods.

show abstract

An improved image mixed noise removal algorithm based on super-resolution algorithm and CNN

Cited by 10 publications

References 26 publications

An Application of Super-Resolution Generative Adversary Networks for Quasi-Static Ultrasound Strain Elastography: A Feasibility Study

An Application of Super-Resolution Generative Adversary Networks for Quasi-Static Ultrasound Strain Elastography: A Feasibility Study

Mixed Noise Removal with Pareto Prior

Perceptual image quality using dual generative adversarial network

Contact Info

Product

Resources

About