• Home
  • Search
  • Ensemble Model of Attention Mechanism-Based DCGAN and Autoencoder for Noised OCR Classification
  • Cite Icon6
  • https://doi.org/10.30564/jeis.v4i1.6725Copy DOI Icon

Ensemble Model of Attention Mechanism-Based DCGAN and Autoencoder for Noised OCR Classification

Show More
  • Abstract
  • Literature Map
  • Citations
  • Similar Papers
Abstract

Optical Character Recognition (OCR) is a technology that converts images of text into machine-readable formats, essential for digitizing printed texts and enabling digital searches. Traditional OCR methods often struggle with variations in font styles and noise. This paper proposes an innovative approach to enhance OCR classification under challenging conditions by leveraging an ensemble model that combines an Attention Mechanism-Based Generative Adversarial Network (GAN) and an Autoencoder. The GAN generates synthetic data to mitigate the limitations of small datasets, while the autoencoder extracts robust features from noisy images. The model undergoes a two-phase training process, initially learning from the augmented dataset and then fine-tuning on a smaller, labeled dataset. Grad-CAM is used to demonstrate interpretability, highlighting the attention regions during predictions. Experimental results show significant improvements in OCR accuracy and robustness, validating the effectiveness of the proposed method in handling noise and limited training data.

Similar Papers
  • Book Chapter
  • Citations18

Building Super-Resolution Image Generator for OCR Accuracy Improvement

  • Jan 01, 2020
  • Xujun Peng +1
  • Research Article
  • Citations3

Improving Tesseract OCR Accuracy Using SymSpell Algorithm on Passport Data

  • Jan 27, 2025
  • Sinkron
  • Iqbaluddin Syam Had +2
  • Conference Article
  • Citations11

Removal of Obstacles in Devanagari Script for Efficient Optical Character Recognition

  • Dec 01, 2015
  • Vivek Kumar Verma +1
  • Research Article
  • Citations300

Optical Character Recognition by Open source OCR Tool Tesseract: A Case Study

  • Oct 20, 2012
  • International Journal of Computer Applications
  • Chirag Patel +2
  • Research Article
  • Citations6

Reading in the mist: high-quality optical character recognition based on freely available early modern digitized books

  • Apr 06, 2022
  • Digital Scholarship in the Humanities
  • Andrea Sangiacomo +4
  • PDF
  • Research Article
  • Citations2

Pixel-Level Degradation for Text Image Super-Resolution and Recognition

  • Nov 05, 2023
  • Electronics
  • Xiaohong Qian +4
  • Research Article

Deep-learning-based Character Recognition of Aerospace Alloy Components for Automated Quality Inspection

  • Dec 26, 2025
  • Sensors and Materials
  • Ming-Chan Lee +4
  • Dissertation

Improving OCR Accuracy of Damaged Pictures with Generative Adversarial Networks

  • Dec 11, 2019
  • Pu Du
  • Conference Article
  • Citations4

Feature Extraction Method of Machine Translation Equivalent Pairs in Chinese-English Comparable Corpus based OCR Recognition

  • Jun 03, 2021
  • Bo Wang
  • Conference Article
  • Citations38

Beyond OCR + VQA

  • Oct 17, 2021
  • Gangyan Zeng +3
  • Research Article
  • Citations6

Deep Learning-Based Multifunctional End-to-End Model for Optical Character Classification and Denoising

  • Nov 15, 2023
  • Journal of Computational Methods in Engineering Applications
  • Shuguang Xiong +2
  • Conference Article
  • Citations1

Issue Based OCR Error Prediction in Video Streams

  • Sep 23, 2020
  • Dirk Siegmund +2
  • Conference Article
  • Citations63

A Fast Alignment Scheme for Automatic OCR Evaluation of Books

  • Jan 01, 2011
  • Ismet Zeki Yalniz +1
  • Conference Article
  • Citations1

Towards a Higher Accuracy of Optical Character Recognition of Chinese Rare Books in Making Use of Text Model

  • May 08, 2019
  • Hsiang-An Wang +1
  • Research Article
  • Citations9

Oriya Character Recognition using Neural Networks

  • Jan 01, 2012
  • International Journal of Computer and Communication Technology
  • Soumya Ranjan Mishra +2
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.