Multi-Channel and Multi-Model-Based Autoencoding Prior for Grayscale Image Restoration

Li, Sanqian; Qin, Binjie; Xiao, Jing; Liu, Qiegen; Wang, Yuhao; Liang, Dong

doi:10.1109/tip.2019.2931240

Cited by 28 publications

(12 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The dataset of the defective image restoration network is also composed of the training set and test set; a total of 10,000 defective images and 10,000 true images are selected, with a total of 14,000 positive and negative samples [17], including 7,000 positive sample images and 7,000 negative sample images, where 70.00% of the prepared dataset (14,000 images) is used as the training set for the training of the image restoration algorithm. The remaining 30.00% (6000 images) of 3000 negative samples are used as test data to validate the effect of repairing the defective regions in this study [18]. For the 7000 negative samples in the training set of the restoration network 8…”

Section: Analysis Of Resultsmentioning

confidence: 99%

A Partial Differential Equation-Based Image Restoration Method in Environmental Art Design

2021

Advances in Mathematical Physics

View full text Add to dashboard Cite

With the rapid development of networks and the emergence of various devices, images have become the main form of information transmission in real life. Image restoration, as an important branch of image processing, can be applied to real-life situations such as pixel loss in image transmission or network prone to packet loss. However, existing image restoration algorithms have disadvantages such as fuzzy restoration effect and slow speed; to solve such problems, this paper adopts a dual discriminator model based on generative adversarial networks, which effectively improves the restoration accuracy by adding local discriminators to track the information of local missing regions of images. However, the model is not optimistic in generating reasonable semantic information, and for this reason, a partial differential equation-based image restoration model is proposed. A classifier and a feature extraction network are added to the dual discriminator model to provide category, style, and content loss constraints to the generative network, respectively. To address the training instability problem of discriminator design, spectral normalization is introduced to the discriminator design. Extensive experiments are conducted on a data dataset of partial differential equations, and the results show that the partial differential equation-based image restoration model provides significant improvements in image restoration over previous methods and that image restoration techniques are exceptionally important in the application of environmental art design.

show abstract

Section: Analysis Of Resultsmentioning

confidence: 99%

A Partial Differential Equation-Based Image Restoration Method in Environmental Art Design

2021

Advances in Mathematical Physics

View full text Add to dashboard Cite

show abstract

“…With the approximate equivalence between problem (1) and (41) (see [50], [51]), in what follows, we illustrate the efficiency of our algorithm in approximating (41). Some recent methods for image restoration includes; multi-channel and multi-model based auto encoding prior for gray scale image restoration [52]; formatted learning for image restoration [53], image restoration by combined order regularization with optimal spatial adaptation [54], multi-Level encoder-decoder architectures for image restoration [55], riemannian loss for image restoration [56] and modulating image restoration with continual levels via adaptive feature modification layers [57].…”

Section: Application To Image Restoration Problemsmentioning

confidence: 99%

A Family of Derivative-Free Conjugate Gradient Methods for Constrained Nonlinear Equations and Image Restoration

Ibrahim

Kumam

2020

IEEE Access

View full text Add to dashboard Cite

In this paper, a derivative-free conjugate gradient method for solving nonlinear equations with convex constraints is proposed. The proposed method can be viewed as an extension of the three-term modified Polak-Ribiére-Polyak method (TTPRP) and the three-term Hestenes-Stiefel conjugate gradient method (TTHS) using the projection technique of Solodov and Svaiter [Reformulation: Nonsmooth, Piecewise Smooth, Semismooth and Smoothing Methods, 1998, 355-369]. The proposed method adopts the adaptive line search scheme proposed by Ou and Li [Journal of Applied Mathematics and Computing 56.1-2 (2018): 195-216] which reduces the computational cost of the method. Under the assumption that the underlying operator is Lipschitz continuous and satisfies a weaker condition of monotonicity, the global convergence of the proposed method is established. Furthermore, the proposed method is extended to solve image restoration problem arising in compressive sensing. Numerical results are presented to demonstrate the effectiveness of the proposed method.

show abstract

“…It can be noted here that before the above method was proposed, the similar concept of plug and play was proposed in Reference [5], in which the half-quadratic splitting method (HQS) can be used in a variety of subproblems of image restoration. Some researchers have also proposed a denoising autoencoder based on a multichannel model and applied it to the restoration task of single-channel gray-scale infrared image [26]. In addition, there are also methods to apply the half-quadratic splitting method to image super-resolution task and achieve good results [27].…”

Section: Related Workmentioning

confidence: 99%

Convolutional Neural Network Combined with Half-Quadratic Splitting Method for Image Restoration

Luo

Zhang

2020

Journal of Sensors

View full text Add to dashboard Cite

Generally, there are mainly two methods to solve the image restoration task in low-level computer vision, i.e., the model-based optimization method and the discriminative learning method. However, these two methods have clear advantages and disadvantages. For example, it is flexible for the model-based optimization method to handle different problems, but large quantity of computing time is required for better performance. The discriminative learning approach has high computing efficiency, but the application scope is seriously limited by the fixed training model. It would be better to combine the advantages of these two methods. Luckily, with the variable splitting techniques, we insert the trained convolutional neural network (CNN) for denoising as one model to the model-based optimization method to solve other image restoration problems (e.g., deblurring and super-resolution). Final experimental results show that our denoising network is able to provide strong prior information for image restoration tasks. The image restoration effects can reach or approximate the most advanced algorithm in such three tasks as denoising, deblurring, and super-resolution. Moreover, the algorithm proposed in this paper is also the most competitive in terms of computational efficiency.

show abstract

Multi-Channel and Multi-Model-Based Autoencoding Prior for Grayscale Image Restoration

Cited by 28 publications

References 45 publications

A Partial Differential Equation-Based Image Restoration Method in Environmental Art Design

A Partial Differential Equation-Based Image Restoration Method in Environmental Art Design

A Family of Derivative-Free Conjugate Gradient Methods for Constrained Nonlinear Equations and Image Restoration

Convolutional Neural Network Combined with Half-Quadratic Splitting Method for Image Restoration

Contact Info

Product

Resources

About