Deep context-aware descreening and rescreening of halftone images

  title={Deep context-aware descreening and rescreening of halftone images},
  author={Tae-Hoon Kim and Sang Il Park},
  journal={ACM Transactions on Graphics (TOG)},
  pages={1 - 12}
A fully automatic method for descreening halftone images is presented based on convolutional neural networks with end-to-end learning. [...] Key Method The method consists of two main stages. In the first stage, intrinsic features of the scene are extracted, the low-frequency reconstruction of the image is estimated, and halftone patterns are removed. For the intrinsic features, the edges and object-categories are estimated and fed to the next stage as strong visual and contextual cues.Expand
Deep Halftoning with Reversible Binary Pattern
  • Menghan Xia, Wenbo Hu, Xueting Liu, Tien-Tsin Wong
Existing halftoning algorithms usually drop colors and fine details when dithering color images with binary dot patterns, which makes it extremely difficult to recover the original information. ToExpand
An Efficient Convolutional Neural Network Model Combined with Attention Mechanism for Inverse Halftoning
  • Linhao Shao, E. Zhang, Mei Li
  • Computer Science
  • Electronics
  • 2021
A simple yet effective deep learning model combined with the attention mechanism is proposed, which can better guide the network to remove noise dot-patterns and restore image details, and improve the network adaptation ability, and outperforms the state-of-the-art methods. Expand
Inverse Halftoning Methods Based on Deep Learning and Their Evaluation Metrics: A Review
This paper proposed a classification method for inverse halftoning methods on the basis of the source of halftone images, and studied existing image quality evaluation including subjective and objective evaluation by experiments to demonstrate that methods based on multiple subnetworks and methodsbased on multi-stage strategies are superior to other methods. Expand
Deep Restoration of Vintage Photographs From Scanned Halftone Prints
This research adopts a novel strategy of two-stage deep learning, in which the restoration task is divided into two stages: the removal of printing artifacts and the inverse of halftoning, which significantly outperforms the existing ones in visual quality. Expand
FHDe2Net: Full High Definition Demoireing Network
The Full High Definition Demoiréing Network (FHDeNet) is proposed, consisting of a global to local cascaded removal branch to eradicate multi-scale moiré patterns and a frequency based highresolution content separation branch to retain fine details. Expand
Deep Learning-Based Forgery Attack on Document Images
The forged-andrecaptured samples created by the proposed text editing attack and recapturing operation have successfully fooled some existing document authentication systems. Expand
Wavelet-Based Dual-Branch Network for Image Demoireing
Although designed for image demoireing, WDNet has been applied to two other low-levelvision tasks, outperforming state-of-the-art image deraining and derain-drop methods on the Rain100h and Raindrop800 data sets, respectively. Expand
ARMAS: Active Reconstruction of Missing Audio Segments
The combination of steganography, halftoning (dithering), and state-of-theart shallow (RFRandom Forest and SVRSupport Vector Regression) and deep learning (LSTMLong Short-Term Memory) methods are proposed and show that the proposed solution is effective and can enhance the reconstruction of audio signals performed by the side information Steganography provides. Expand
Combination of Convolutional and Generative Adversarial Networks for Defect Image Demoiréing of Thin-Film Transistor Liquid-Crystal Display Image
This study investigated the problem and proposed an approach to eliminate moiré patterns from defect images based on a generative adversarial network architecture by using the U-net network as a generator and adding a discriminator, and quantitatively and qualitatively outperforms other methods. Expand
Semantic photo manipulation with a generative image prior
This paper adapts the image prior learned by GANs to image statistics of an individual image and can accurately reconstruct the input image and synthesize new content, consistent with the appearance of theinput image. Expand


Image Companding and Inverse Halftoning using Deep Convolutional Neural Networks
This work introduces deep learning technology to tackle two traditional low-level image processing problems, companding and inverse halftoning and develops an effective deep learning algorithm based on insights into the properties of visual quality of images and the internal representation properties of a deep convolutional neural network (CNN). Expand
Context Encoders: Feature Learning by Inpainting
It is found that a context encoder learns a representation that captures not just appearance but also the semantics of visual structures, and can be used for semantic inpainting tasks, either stand-alone or as initialization for non-parametric methods. Expand
Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network
  • C. Ledig, Lucas Theis, +6 authors W. Shi
  • Computer Science, Mathematics
  • 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2017
SRGAN, a generative adversarial network (GAN) for image super-resolution (SR), is presented, to its knowledge, the first framework capable of inferring photo-realistic natural images for 4x upscaling factors and a perceptual loss function which consists of an adversarial loss and a content loss. Expand
Training-Based Descreening
The proposed scheme not only suppresses the Moireacute artifacts, but can be trained with intrinsic sharpening for deblurring scanned documents, and once optimized for a periodic clustered-dot halftoning method, the same algorithm can be used to inverse halftones scanned images containing stochastic error diffusion halftone noise. Expand
EnhanceNet: Single Image Super-Resolution Through Automated Texture Synthesis
This work proposes a novel application of automated texture synthesis in combination with a perceptual loss focusing on creating realistic textures rather than optimizing for a pixelaccurate reproduction of ground truth images during training to achieve a significant boost in image quality at high magnification ratios. Expand
Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification
This work proposes a Parametric Rectified Linear Unit (PReLU) that generalizes the traditional rectified unit and derives a robust initialization method that particularly considers the rectifier nonlinearities. Expand
Holistically-Nested Edge Detection
HED turns pixel-wise edge classification into image-to-image prediction by means of a deep learning model that leverages fully convolutional neural networks and deeply-supervised nets to approach the human ability to resolve the challenging ambiguity in edge and object boundary detection. Expand
Enhanced Deep Residual Networks for Single Image Super-Resolution
This paper develops an enhanced deep super-resolution network (EDSR) with performance exceeding those of current state-of-the-art SR methods, and proposes a new multi-scale deepsuper-resolution system (MDSR) and training method, which can reconstruct high-resolution images of different upscaling factors in a single model. Expand
Let there be color!
A novel technique to automatically colorize grayscale images that combines both global priors and local image features and can process images of any resolution, unlike most existing approaches based on CNN. Expand
DeepContour: A deep convolutional feature learned by positive-sharing loss for contour detection
This work shows that contour detection accuracy can be improved by instead making the use of the deep features learned from convolutional neural networks (CNNs), while rather than using the networks as a blackbox feature extractor, it customize the training strategy by partitioning contour (positive) data into subclasses and fitting each subclass by different model parameters. Expand