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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

HP1γ promotes the progression of colorectal cancer through interaction with RBPJ of the Notch signaling pathway.

Cell biology and toxicology·2026
Same author

α-Galactosylceramide-expanded virtual memory CD8<sup>+</sup> T cells confer protection against a broad range of pathogens.

Frontiers in immunology·2026
Same author

The Mechanism of NLRP3 Inflammasome Activation and Its Roles in Chronic Rhinosinusitis.

World journal of otorhinolaryngology - head and neck surgery·2026
Same author

The Ideal Number of Examined Lymph Node Stations for Accurate Nodal Staging and Prognosis in Pathological T1-2 Esophageal Squamous Cell Carcinoma: A Large-Scale Multicenter Cohort Study.

Annals of surgical oncology·2026
Same author

Thoracic endovascular aortic repair for high-risk Stanford type B aortic dissection in a 15-year-old patient with genetic mutations: a case report with follow-up of four years.

Cardiology in the young·2026
Same author

A multicenter randomized controlled trial evaluating a new radiofrequency ablation system in the treatment of primary great saphenous vein incompetence: Six-month results of the ACOART RF CLOSURE study.

Phlebology·2026

Related Experiment Video

Updated: Sep 17, 2025

Author Spotlight: Quantitative Characterization of Liquid Photosensitive Bioink Properties for Continuous Digital Light Processing Based Printing
04:32

Author Spotlight: Quantitative Characterization of Liquid Photosensitive Bioink Properties for Continuous Digital Light Processing Based Printing

Published on: April 14, 2023

1.1K

Printed document layout analysis and optical character recognition system based on deep learning.

Dong-Lin Li1, Shih-Kai Lee2, Yin-Ting Liu2

  • 1Department of electrical engineering, National Taiwan Ocean University, Beining Rd., Keelung City, 202301, Taiwan. ericli@email.ntou.edu.tw.

Scientific Reports
|July 3, 2025
PubMed
Summary
This summary is machine-generated.

This study introduces a deep learning system for document analysis and text recognition. It accurately processes printed documents locally, converting them into editable formats like JSON.

Keywords:
CNNDeep learningLayout analysisOCRYOLO

More Related Videos

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

9.1K
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

4.1K

Related Experiment Videos

Last Updated: Sep 17, 2025

Author Spotlight: Quantitative Characterization of Liquid Photosensitive Bioink Properties for Continuous Digital Light Processing Based Printing
04:32

Author Spotlight: Quantitative Characterization of Liquid Photosensitive Bioink Properties for Continuous Digital Light Processing Based Printing

Published on: April 14, 2023

1.1K
Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

9.1K
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

4.1K

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Machine Learning

Background:

  • Printed document processing often requires specialized software.
  • Existing Optical Character Recognition (OCR) methods can be slow or inaccurate.
  • Deep learning offers potential for improved document analysis.

Purpose of the Study:

  • To develop an efficient and accurate deep learning system for document layout analysis and text recognition.
  • To enable local processing of scanned documents and image files.
  • To output recognized text in user-friendly, editable formats.

Main Methods:

  • Utilized YOLOv4 and YOLOv8 deep learning models for document layout analysis (identifying titles, paragraphs, tables, images).
  • Implemented character segmentation for each identified document element.
  • Employed Convolutional Neural Networks (CNNs) for text recognition.
  • Integrated recognized text into editable formats (JSON, Microsoft formats).

Main Results:

  • Achieved accurate identification of document elements like titles, paragraphs, tables, and images.
  • Demonstrated high accuracy in text recognition through CNNs.
  • Enabled fast and convenient OCR processing on a local computer.

Conclusions:

  • The proposed deep learning system provides a robust solution for document layout analysis and OCR.
  • The method offers a significant improvement in speed and accuracy for local document processing.
  • Editable output formats enhance the usability of recognized text.