DE-GAN: A Conditional Generative Adversarial Network for Document Enhancement (2020-09-07T00:00:00.000000Z)

TL;DR

This paper proposes an effective end-to-end framework named document enhancement generative adversarial networks (DE-GAN) that uses the conditional GANs (cGANs) to restore severely degraded document images.

Abstract

Documents often exhibit various forms of degradation, which make it hard to be read and substantially deteriorate the performance of an OCR system. In this paper, we propose an effective end-to-end framework named document enhancement generative adversarial networks (DE-GAN) that uses the conditional GANs (cGANs) to restore severely degraded document images. To the best of our knowledge, this practice has not been studied within the context of generative adversarial deep networks. We demonstrate that, in different tasks (document clean up, binarization, deblurring and watermark removal), DE-GAN can produce an enhanced version of the degraded document with a high quality. In addition, our approach provides consistent improvements compared to state-of-the-art methods over the widely used DIBCO 2013, DIBCO 2017, and H-DIBCO 2018 datasets, proving its ability to restore a degraded document image to its ideal condition. The obtained results on a wide variety of degradation reveal the flexibility of the proposed model to be exploited in other document enhancement problems.

Authors

Mohamed Ali Souibgui

2 papers

Yousri Kessentini

2 papers

TL;DR

Abstract

Authors

References66 items

Large-Scale Visible Watermark Detection and Removal with Deep Convolutional Networks

A New Threshold Selection Method Based on Fuzzy Expert Systems for Separating Text from the Background of Document Images

Historical document image binarization using background estimation and energy minimization

ICFHR 2018 Competition on Handwritten Document Image Binarization (H-DIBCO 2018)

Degraded historical document image binarization using local features and support vector machine (SVM)

Document Enhancement Using Visibility Detection

Document Image Binarization Using Recurrent Neural Networks

Binarization of degraded document images based on hierarchical deep supervised network

De-Mark GAN: Removing Dense Watermark with Generative Adversarial Network

Staff line Removal using Generative Adversarial Networks

Word Level Font-to-Font Image Translation using Convolutional Recurrent Generative Adversarial Networks

High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs

ICDAR 2019 Competition on Document Image Binarization (DIBCO 2019)

Deep Networks for Degraded Document Image Binarization through Pyramid Reconstruction

An automatic visible watermark removal technique using image inpainting algorithms

DeepDeSRT: Deep Learning for Detection and Structure Recognition of Tables in Document Images

Progressive Growing of GANs for Improved Quality, Stability, and Variation

Document Image Binarization with Fully Convolutional Neural Networks

On the Effectiveness of Visible Watermarks

A selectional auto-encoder approach for document image binarization

DualGAN: Unsupervised Dual Learning for Image-to-Image Translation

Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks

Language Independent Single Document Image Super-Resolution using CNN for improved recognition

Handwriting Profiling Using Generative Adversarial Networks

Semantic Segmentation using Adversarial Networks

Image-to-Image Translation with Conditional Adversarial Networks

Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network

Image Restoration Using Very Deep Convolutional Encoder-Decoder Networks with Symmetric Skip Connections

LLNet: A deep autoencoder approach to natural low-light image enhancement

SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation

Document Image Binarization using LSTM: A Sequence Learning Approach

An improved approach for automatic denoising and binarization of degraded document images based on region localization

Fast and accurate scene text understanding with image binarization and off-the-shelf OCR

U-Net: Convolutional Networks for Biomedical Image Segmentation

Image Super-Resolution Using Deep Convolutional Networks

Fully convolutional networks for semantic segmentation

Constrained Energy Maximization and Self-Referencing Method for Invisible Ink Detection from Multispectral Historical Document Images

Auto-Encoding Variational Bayes

Document binarization with automatic parameter tuning

ICDAR 2013 Document Image Binarization Contest (DIBCO 2013)

FAIR: A Fast Algorithm for Document Image Restoration

Image Denoising and Inpainting with Deep Neural Networks

ICFHR 2012 Competition on Handwritten Document Image Binarization (H-DIBCO 2012)

An effective document image deblurring algorithm

Adaptive local thresholding for detection of nuclei in diversity stained cytology images

A Variational Approach to Degraded Document Enhancement

A binarization method with learning-built rules for document images produced by cameras

An Overview of the Tesseract OCR Engine

Behaviour-Based Clustering of Neural Networks Applied to Document Enhancement

High Performance Convolutional Neural Networks for Document Processing

Restoring Warped Document Images through 3D Shape Modeling

An Adaptive Binarization Technique for Low Quality Historical Documents

Removing Shadows from Images

Adaptive document image binarization

A recursive thresholding technique for image segmentation

An introduction to digital image processing

Computational Intelligence

GENERATIVE ADVERSARIAL NETS

ICDAR2017 Competition on Document Image Binarization (DIBCO 2017)

Convolutional Neural Networks for Direct Text Deblurring

ICFHR2014 Competition on Handwritten Document Image Binarization (H-DIBCO 2014)

A Robust Document Image Binarization Technique for Degraded Document Images

Removing Shadows from Images

A Threshold Selection Method from Gray-Level Histograms

2009 10th International Conference on Document Analysis and Recognition ICDAR 2009 Document Image Binarization Contest (DIBCO 2009)

learning, document processing and data fusion, and computer vision

Field of Study

Journal Information

Name

Page

Volume

Venue Information

Name

Type

URL

Alternate Names