Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jun 18, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

512

DCNet: Noise-Robust Convolutional Neural Networks for Degradation Classification on Ancient Documents.

Fitri Arnia1,2,3, Khairun Saddami1,3, Khairul Munadi1,2

  • 1Department of Electrical and Computer Engineering, Universitas Syiah Kuala, Banda Aceh 23111, Indonesia.

Journal of Imaging
|July 31, 2024
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Lightweight convolutional neural network (CNN) model for obesity early detection using thermal images.

Digital health·2024
Same author

Knowledge creation elements for enhancing community resilience towards disaster: A Delphi study.

Jamba (Potchefstroom, South Africa)·2021
Same author

Homomorphic Filtering and Phase-Based Matching for Cross-Spectral Cross-Distance Face Recognition.

Sensors (Basel, Switzerland)·2021
See all related articles

This study introduces a novel noise-robust convolutional neural network (CNN), DCNet, for classifying degraded ancient documents. DCNet significantly improves analysis accuracy, especially for heavily noised historical documents.

Area of Science:

  • Digital Humanities
  • Computer Vision
  • Document Image Analysis

Background:

  • Degraded ancient documents present significant analysis challenges due to combined degradation and digitalization noise.
  • Noise from low-specification devices and poor illumination further complicates document analysis.

Purpose of the Study:

  • To propose a new noise-robust convolutional neural network (CNN) architecture, DCNet, for degradation classification of noisy ancient documents.
  • To enhance the performance of ancient document analysis in the presence of various noise types.

Main Methods:

  • Developed a novel degradation classification network (DCNet) based on ResNet101, MobileNetV2, and ShuffleNet architectures.
  • Introduced a self-transition layer to follow the DCNet architecture.
  • Trained and tested DCNet using document images with and without various noise levels (zero mean Gaussian noise and speckle noise).
Keywords:
deep learningdegradation classificationdegraded ancient document imagesdocument image analysis

More Related Videos

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.0K
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.8K

Related Experiment Videos

Last Updated: Jun 18, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

512
Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.0K
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.8K

Main Results:

  • The proposed DCNet architecture demonstrated superior performance compared to standard MobileNet, ShuffleNet, and ResNet101.
  • DCNet achieved better results than conventional machine learning methods like support vector machines and random forests.
  • The architecture was particularly effective in classifying degraded documents with heavy noise.

Conclusions:

  • The developed DCNet is a highly effective and noise-robust solution for classifying degraded ancient documents.
  • This advancement can significantly improve the accuracy and reliability of digital humanities research and historical document preservation.
  • The proposed architecture offers a promising direction for handling noisy and degraded image data in document analysis.