Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Dec 18, 2025

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

Improved Handwritten Digit Recognition Using Convolutional Neural Networks (CNN).

Savita Ahlawat¹, Amit Choudhary², Anand Nayyar³

¹Department of Computer Science and Engineering, Maharaja Surajmal Institute of Technology, New Delhi 110058, India.

Sensors (Basel, Switzerland)

|June 18, 2020

Summary

This summary is machine-generated.

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

Anti-CRISPR-mediated continuous directed evolution of CRISPR-Cas9 in human cells.

Proceedings of the National Academy of Sciences of the United States of America·2026

Same author

Fuzzy Vaults in Biometric Cryptosystems: A Survey of Techniques, Performance, and Applications.

Sensors (Basel, Switzerland)·2026

Same author

Proximity-Induced Rewiring of Oncogenic Kinase Triggers Apoptosis.

ACS central science·2026

Same author

The Interplay of PD-L1, Kynurenine Pathway, and Vitamin D Shapes an Immunosuppressive Microenvironment in Acute Myeloid Leukemia.

Asian Pacific journal of cancer prevention : APJCP·2026

Same author

Survey on Reconnaissance Autonomous Robotic Systems for Disaster Management.

Sensors (Basel, Switzerland)·2026

Same author

Immune-metabolic plasticity in AML: Prognostic roles of aCTLs (activated Cytotoxic T Cells) and Kynurenine.

Current problems in cancer·2026

Same journal

RETRACTED: Zhang et al. A Novel Framework for Reconstruction and Imaging of Target Scattering Centers via Wide-Angle Incidence in Radar Networks. <i>Sensors</i> 2025, <i>25</i>, 6802.

Sensors (Basel, Switzerland)·2026

Same journal

Enhancing Unsupervised Multi-Source Domain Adaptation for Person Re-Identification via Mixture of Experts and Graph-Based Relation.

Sensors (Basel, Switzerland)·2026

Same journal

Development of an Instrumented Glove for Palmar Pressure Assessment in Kayakers.

Sensors (Basel, Switzerland)·2026

Same journal

Development and Experimental Validation of an Autonomous IoT-Based Monitoring System for Real-Time Water Quality Assessment in the Amazon River.

Sensors (Basel, Switzerland)·2026

Same journal

Semi-Supervised Adversarial Learning Framework for Controller Area Network Bus Intrusion Detection.

Sensors (Basel, Switzerland)·2026

Same journal

Smart Optimization Method for Safety Signs in Innovative Manufacturing Environments Integrating Industrial Field IoT Sensors and Knowledge Graphs.

Sensors (Basel, Switzerland)·2026

See all related articles

This study introduces a novel Convolutional Neural Network (CNN) architecture for handwritten digit recognition. The proposed pure CNN design achieves a record 99.87% accuracy on the MNIST dataset, outperforming complex ensemble methods with reduced computational cost.

Area of Science:

Computer Science
Artificial Intelligence
Machine Learning

Background:

Traditional handwriting recognition relies on handcrafted features, posing challenges for Optical Character Recognition (OCR) system training.
Deep learning, particularly Convolutional Neural Networks (CNNs), has revolutionized handwriting recognition, yet demands for higher accuracy persist due to increasing data and computational power.

Purpose of the Study:

To explore CNN design options (layers, stride, kernel size, etc.) for handwritten digit recognition.
To evaluate Stochastic Gradient Descent (SGD) optimization algorithms for performance enhancement.
To develop a pure CNN architecture that rivals or surpasses ensemble methods in accuracy while reducing computational complexity.

Main Methods:

Investigated various CNN architectural parameters including layer depth, stride size, receptive field, kernel size, padding, and dilation.

Keywords:

OCR convolutional neural networks handwritten digit recognition pre-processing

More Related Videos

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

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

Published on: December 15, 2023

Related Experiment Videos

Last Updated: Dec 18, 2025

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

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

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

Published on: December 15, 2023

Evaluated different SGD optimization algorithms to improve handwritten digit recognition performance.

Designed and tested a pure CNN architecture, avoiding ensemble methods to minimize computational cost and testing complexity.

Main Results:

Achieved a new absolute record accuracy of 99.87% for classifying handwritten digits on the MNIST dataset.
Demonstrated that a carefully designed pure CNN architecture can outperform ensemble architectures in recognition accuracy.
Identified an optimal combination of learning parameters for CNN design, leading to superior performance.

Conclusions:

The proposed pure CNN architecture offers a computationally efficient and highly accurate solution for handwritten digit recognition.
Optimizing CNN design parameters and learning strategies is crucial for achieving state-of-the-art performance.
This work sets a new benchmark for handwritten digit classification accuracy on the MNIST dataset.