• Home
  • Search
  • Simultaneous Optimisation of Image Quality Improvement and Text Content Extraction from Scanned Documents
  • Cite Icon6
  • https://doi.org/10.1109/icdar.2019.00189Copy DOI Icon

Simultaneous Optimisation of Image Quality Improvement and Text Content Extraction from Scanned Documents

  • Sep 1, 2019
  • Shashank Mujumdar +3 more
Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Convolutional neural networks are shown to achieve breakthrough performance for the task of single image super resolution (SISR) for natural images. These state-of-the-art (SOA) networks have been adapted to the task of single text image super resolution and have been shown to boost the optical character recognition (OCR) performance. However, these approaches depend on variations of the standard mean squared error (MSE) loss in order to train the SR network for improving the text image quality which does not guarantee optimal OCR performance. In this paper, we propose to combine the OCR performance into the loss function during network training. This results in the generation of high resolution text images that achieve high OCR performance that is comparable to the ground truth high-resolution text images and surpassing those of the SOA baseline results. We define novel intuitive metrics to capture the improvement in the OCR performance and provide extensive experiments to qualitatively and quantitatively assess improvement in the results of our proposed approach against the SOA baselines on the standard UNLV dataset.

Similar Papers
  • Research Article
  • Citations19

Ultralight-Weight Three-Prior Convolutional Neural Network for Single Image Super Resolution

  • Dec 01, 2023
  • IEEE Transactions on Artificial Intelligence
  • Alireza Esmaeilzehi +2
  • Conference Article
  • Citations26

CNN-based text image super-resolution tailored for OCR

  • Dec 01, 2017
  • Haochen Zhang +2
  • Research Article
  • Citations11

DHBSR: A deep hybrid representation-based network for blind image super resolution

  • May 28, 2024
  • Computer Vision and Image Understanding
  • Alireza Esmaeilzehi +3
  • Research Article
  • Citations30

SRNSSI: A Deep Light-Weight Network for Single Image Super Resolution Using Spatial and Spectral Information

  • Jan 01, 2021
  • IEEE Transactions on Computational Imaging
  • Alireza Esmaeilzehi +2
  • Conference Article
  • Citations2

Super-Resolution of Color Images Using CNN

  • Jun 01, 2018
  • Ashvini A Tandale +1
  • Conference Article
  • Citations11

Autoencoder Image Denoising to Increase Optical Character Recognition Performance in Text Conversion

  • Nov 23, 2022
  • Nur Alamsyah +3
  • Conference Article

A Multi-Scale Layer-Channel Attention Network for Image Super-Resolution

  • Oct 22, 2021
  • Jian Wang +5
  • PDF
  • Research Article
  • Citations40

Fully 1 × 1 Convolutional Network for Lightweight Image Super-resolution

  • Aug 20, 2024
  • Machine Intelligence Research
  • Gang Wu +3
  • Book Chapter
  • Citations2

Single Image Super Resolution Using Local and Non-local Priors

  • Jan 01, 2018
  • Tianyi Li +4
  • Book Chapter
  • Citations6

Hyperspectral Image Super-Resolution Using Multi-scale Feature Pyramid Network

  • Jan 01, 2020
  • He Sun +4
  • Conference Article
  • Citations54

A hierarchical, HMM-based automatic evaluation of OCR accuracy for a digital library of books

  • Jun 11, 2006
  • Shaolei Feng +1
  • Research Article
  • Citations10

An End-to-End Tag Recognition Architecture for Industrial Meter

  • Jan 01, 2024
  • IEEE Transactions on Industrial Informatics
  • Xiaoyuan Deng +4
  • Book Chapter

Deep CNN for Single Image Super Resolution Using Skip Connections

  • Jan 01, 2019
  • Sathisha Basavaraju +3
  • Research Article
  • Citations15

Diffusion models for image super-resolution: State-of-the-art and future directions

  • Nov 21, 2024
  • Neurocomputing
  • Garas Gendy +2
  • Conference Article

CL-SRGAN: generative adversary network equipped with curriculum learning for image super-resolution

  • Apr 29, 2023
  • Meiying Chen +3
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.