Image denoising via deep network based on edge enhancement

Chen, Xi; Shu, Zhan; Ji, Dong Jin; Xu, Lu; Wu, Congzhong; Li, Xiaohong

doi:10.1007/s12652-018-1036-4

Cited by 13 publications

(6 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Edge information is utilized to enhance images in extensive areas such as super-resolution [7], [8], [21], image inpainting [24] and denoising tasks [25]. Without any doubt, edge information is a critical factor to not only elaborately reconstruct facial details from low-resolution face images but also solve an endemic problem of l 2 distances loss which tends to generate smoothed images in the super-resolution.…”

Section: Proposed Methods a Edge Blockmentioning

confidence: 99%

Edge and Identity Preserving Network for Face Super-Resolution

Kim,

Li,

Yun

et al. 2020

Preprint

View full text Add to dashboard Cite

Face super-resolution has become an indispensable part in security problems such as video surveillance and identification system, but the distortion in facial components is a main obstacle to overcoming the problems. To alleviate it, most stateof-the-arts have utilized facial priors by using deep networks. These methods require extra labels, longer training time, and larger computation memory. Thus, we propose a novel Edge and Identity Preserving Network for Face Super-Resolution Network, named as EIPNet, which minimizes the distortion by utilizing a lightweight edge block and identity information. Specifically, the edge block extracts perceptual edge information and concatenates it to original feature maps in multiple scales. This structure progressively provides edge information in reconstruction procedure to aggregate local and global structural information. Moreover, we define an identity loss function to preserve identification of super-resolved images. The identity loss function compares feature distributions between super-resolved images and target images to solve unlabeled classification problem. In addition, we propose a Luminance-Chrominance Error (LCE) to expand usage of image representation domain. The LCE method not only reduces the dependency of color information by dividing brightness and color components but also facilitates our network to reflect differences between Super-Resolution (SR) and High-Resolution (HR) images in multiple domains (RGB and YUV). The proposed methods facilitate our super-resolution network to elaborately restore facial components and generate enhanced 8× scaled super-resolution images with a lightweight network structure. Furthermore, our network can generate a 128 × 128 superresolved image with 215 fps on a GTX 1080Ti GPU. Extensive experimental results demonstrate that our network qualitatively and quantitatively outperforms state-of-the-art methods on two challenging datasets: CelebA and VGGFace2.

show abstract

Section: Proposed Methods a Edge Blockmentioning

confidence: 99%

Edge and Identity Preserving Network for Face Super-Resolution

Kim,

Li,

Yun

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Jain et al [19] firstly introduce Convolutional neural networks (CNN) which has a small receptive field into image denoising. Chen et al [20] joint Euclidean and perceptual loss functions to find more edge information. According to deep image prior (DIP), present by Ulyanov et al [21], abundant prior knowledge for image denosing already exist in the pre-train convolutional neural network.…”

Section: Single Image Denosingmentioning

confidence: 99%

Unsupervised MRI Images Denoising via Decoupled Expression

Zhang

Xiang

2022

Proceeding of 2021 International Conference on Wireless Communications, Networking and Applications

View full text Add to dashboard Cite

Magnetic Resonance Imaging (MRI) is widely adopted in medical diagnosis. Due to the spatial coding scheme, MRI image is degraded by various noise. Recently, massive methods have been applied to the MRI image denoising. However, they lack the consideration of artifacts in MRI images. In this paper, we propose an unsupervised MRI image denoising method called UEGAN based on decoupled expression. We decouple the content and noise in a noisy image using content encoders and noise encoders. We employ a noising branch to push the noise decoder only extract the noise. The cycle-consistency loss ensures that the content of the denoised results match the original images. To acquire visually realistic generations, we add an adversarial loss on denoised results. Image quality penalty helps to retain rich image details. We perform experiments on unpaired MRI images from Brainweb datesets, and achieve superior performances compared to several popular denoising approaches.

show abstract

“…The second way mainly designed different loss function according to characteristic of nature images to extract more robust features [9]. For example, Chen et al [30] jointed Euclidean and perceptual loss functions to mine more edge information for image denoising. The third way enlarged receptive field size to improve denoising performance via increasing the depth or width of network [185,218,163].…”

Section: Cnn/nn For Awni Denoisingmentioning

confidence: 99%

Deep Learning on Image Denoising: An overview

Tian

Zhang

Zheng³

et al. 2019

Preprint

View full text Add to dashboard Cite

Deep learning techniques have obtained much attention in image denoising. However, deep learning methods of different types deal with the noise have enormous differences. Specifically, discriminative learning based on deep learning can well address the Gaussian noise. Optimization model methods based on deep learning have good effect on estimating of the real noise. So far, there are little related researches to summarize different deep learning techniques for image denoising. In this paper, we make such a comparative study of different deep techniques in image denoising. We first classify the (1) deep convolutional neural networks (CNNs) for additive white noisy images, (2) deep CNNs for real noisy images, (3) deep CNNs for blind denoising and (4) deep CNNs for hybrid noisy images, which is the combination of noisy, blurred and low-resolution images. Then, we analyze the motivations and principles of deep learning methods of different types. Next, we compare and verify the state-of-the-art methods on public denoising datasets in terms of quantitative and qualitative analysis. Finally, we point out some potential challenges and directions of future research.

show abstract

Image denoising via deep network based on edge enhancement

Cited by 13 publications

References 27 publications

Edge and Identity Preserving Network for Face Super-Resolution

Edge and Identity Preserving Network for Face Super-Resolution

Unsupervised MRI Images Denoising via Decoupled Expression

Deep Learning on Image Denoising: An overview

Contact Info

Product

Resources

About