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

Underappreciated Soil NO<sub><i>x</i></sub> Emissions Amplify Background O<sub>3</sub> Pollution During Compound Dry-Hot Extremes in North China.

Environmental science & technology·2026
Same author

Transcutaneous auricular vagus nerve stimulation for postoperative nausea and vomiting in gynecological laparoscopic surgery: a randomized controlled trial protocol.

Annals of medicine·2026
Same author

Efficacy of hyperbaric oxygen therapy for post-stroke depression: Association with inflammatory biomarker reduction in a sham-controlled randomized trial.

Journal of stroke and cerebrovascular diseases : the official journal of National Stroke Association·2026
Same author

The Promotion Effect of Dust on Oxidative Potential of PM<sub>2.5</sub> via Photochemical Aging.

Environmental science & technology·2026
Same author

Crucial Oxidative Stress Risks of Atmospheric Fine Particulate Matter Generated by Sandstorm Process.

Environmental science & technology·2026
Same author

Controllable Protein Design by Prefix-Tuning Protein Language Models.

Journal of chemical information and modeling·2026

Related Experiment Video

Updated: Dec 31, 2025

Digital Handwriting Analysis of Characters in Chinese Patients with Mild Cognitive Impairment
05:58

Digital Handwriting Analysis of Characters in Chinese Patients with Mild Cognitive Impairment

Published on: March 11, 2021

5.0K

Stroke Sequence-Dependent Deep Convolutional Neural Network for Online Handwritten Chinese Character Recognition.

Xin Liu, Baotian Hu, Qingcai Chen

    IEEE Transactions on Neural Networks and Learning Systems
    |January 7, 2020
    PubMed
    Summary

    A new stroke sequence-dependent deep convolutional neural network (SSDCNN) model improves online handwritten Chinese character recognition (OLHCCR) by utilizing stroke order and eight-directional features. This novel approach significantly reduces recognition errors, achieving state-of-the-art accuracy.

    More Related Videos

    Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
    04:48

    Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

    Published on: November 30, 2022

    3.3K
    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
    04:48

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

    Published on: July 5, 2024

    707

    Related Experiment Videos

    Last Updated: Dec 31, 2025

    Digital Handwriting Analysis of Characters in Chinese Patients with Mild Cognitive Impairment
    05:58

    Digital Handwriting Analysis of Characters in Chinese Patients with Mild Cognitive Impairment

    Published on: March 11, 2021

    5.0K
    Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
    04:48

    Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

    Published on: November 30, 2022

    3.3K
    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
    04:48

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

    Published on: July 5, 2024

    707

    Area of Science:

    • Computer Vision
    • Machine Learning
    • Artificial Intelligence
    • Pattern Recognition

    Background:

    • Online handwritten Chinese character recognition (OLHCCR) is a challenging task due to the complexity and variability of human handwriting.
    • Existing methods often rely on static features, potentially overlooking crucial dynamic information like stroke sequence.
    • Eight-directional features are a classical approach but may not capture the full representational power of handwritten characters.

    Purpose of the Study:

    • To introduce a novel deep learning model, the stroke sequence-dependent deep convolutional neural network (SSDCNN), for enhanced OLHCCR.
    • To leverage both stroke sequence information and eight-directional features for more accurate character representation.
    • To achieve state-of-the-art performance in OLHCCR tasks.

    Main Methods:

    • Inputting stroke sequences and transforming them into feature maps that follow the writing order.
    • Employing convolutional, residual, and max-pooling operations to derive stroke sequence-dependent representations.
    • Integrating these representations with eight-directional features using fully connected layers and a softmax classifier for recognition.

    Main Results:

    • SSDCNN achieved a maximum 58.28% reduction in recognition error compared to models using only eight-directional features (2.14% vs. 5.13%).
    • The model demonstrated high accuracy (97.86%), outperforming the winning system of the ICDAR 2013 competition by approximately 18.0% error reduction.
    • An adapted version, SSDCNN+Adapt, reached a new state-of-the-art accuracy of 97.94%.

    Conclusions:

    • The proposed SSDCNN effectively exploits stroke sequence information to learn high-quality representations for OLHCCR.
    • The combination of learned stroke sequence-dependent features and classical eight-directional features offers complementary benefits.
    • SSDCNN represents a significant advancement in OLHCCR, achieving state-of-the-art performance and demonstrating the value of incorporating sequential writing dynamics.