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 Videos

Attention-based handwritten Chinese recognition for power grid maintenance documents.

Dajun Xiao1, Xialing Xu1, Lianfei Shan2,3

  • 1Central China Power Dispatching and Control Center of State Grid, Wuhan, China.

Science Progress
|March 28, 2025
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

[A study on the relationship between glutathione S-transferases gene polymorphism and susceptibility response to hypoxia].

Zhongguo ying yong sheng li xue za zhi = Zhongguo yingyong shenglixue zazhi = Chinese journal of applied physiology·2010
Same author

Synthesis and biological evaluation of new 4β-anilino-4'-O-demethyl-4-desoxypodophyllotoxin derivatives as potential antitumor agents.

European journal of medicinal chemistry·2010
Same author

Roles of reactive oxygen species and mitochondria in cadmium-induced injury of liver cells.

Toxicology and industrial health·2010
Same author

[Neuroprotective effects of emodin on primary rat cortical neurons apoptosis induced by hydrogen peroxide].

Zhong yao cai = Zhongyaocai = Journal of Chinese medicinal materials·2010
Same author

[The protective effect of XD in ConA-induced liver injury].

Xi bao yu fen zi mian yi xue za zhi = Chinese journal of cellular and molecular immunology·2010
Same author

The modulation of surface texture for single-crystalline Si solar cells using calibrated silver nanoparticles as a catalyst.

Nanotechnology·2010

This study introduces a new method for recognizing handwritten Chinese documents, achieving high accuracy. The approach effectively captures character features for improved efficiency in power grid enterprises.

Area of Science:

  • Computer Vision
  • Artificial Intelligence
  • Document Analysis

Background:

  • Handwritten Chinese document recognition is vital for power grid enterprise efficiency.
  • Existing methods may struggle with diverse handwriting styles and complex features.

Purpose of the Study:

  • To propose a novel handwritten Chinese document recognition method.
  • To enhance recognition accuracy and efficiency for power grid applications.

Main Methods:

  • Feature extraction using an inception module for multi-scale spatial characteristics.
  • Attention mechanism (space channel parallel attention) to highlight important features.
  • Bidirectional Long Short-Term Memory (BiLSTM) network for character probability prediction.
  • Transcription layer for loss computation and final result generation.
Keywords:
BiLSTMHandwritten Chinese recognitionattentionconnectionist temporal classificationpower grid documents

Related Experiment Videos

Main Results:

  • The proposed method achieved an accurate rate of 96.92%.
  • The correct rate reached 97.66%.
  • Demonstrated effectiveness in diverse handwriting styles.

Conclusions:

  • The novel method significantly improves handwritten Chinese document recognition accuracy.
  • The approach is effective in capturing intricate handwritten character features.
  • Offers a promising solution for enhancing operational efficiency in relevant industries.