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 16, 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

496

Recognition of inscribed cursive Pashtu numeral through optimized deep learning.

Sibtain Syed1, Khalil Khan2, Maqbool Khan1,3

  • 1Department of IT & CS, Pak-Austria Fachhochschule Institute of Applied Sciences and Technology, Haripur, KP, Pakistan.

Peerj. Computer Science
|August 15, 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

Congenital heart disease diagnosis using machine learning: a systematic literature review.

Frontiers in medicine·2026
Same author

EEG-based harmful brain activity classification using deep learning and feature fusion.

Scientific reports·2026
Same author

Federated Gastrointestinal Lesion Classification with Clinical-Entropy Guided Quantum-Inspired Token Pruning in Vision Transformers.

Diagnostics (Basel, Switzerland)·2026
Same author

Hybrid vision transformer framework for congenital heart disease diagnosis.

Scientific reports·2026
Same author

Artificial intelligence enabled fouling prediction and effect of adsorbent sources in submerged fluidized bed ceramic membrane reactor for food industry wastewater treatment.

Environmental research·2026
Same author

Intelligent delignification: leveraging explainable AI for ozone transport modeling and optimization.

Scientific reports·2025

This study introduces optimized machine learning models for Pashtu numeric recognition. The Long Short-Term Memory (LSTM) model slightly outperformed the Convolutional Neural Network (CNN) model in accurately identifying Pashtu numerals.

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Natural Language Processing

Background:

  • Pashtu, a widely spoken language in Southeast Asia, presents unique challenges for numeric recognition due to its cursive script.
  • Traditional methods struggle with Pashtu's script, necessitating advanced computational approaches.

Purpose of the Study:

  • To develop and optimize machine learning models for accurate optical character recognition (OCR) of Pashtu numerics (0-9).
  • To compare the performance of Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN) models for this task.

Main Methods:

  • A dataset of Pashtu numerics was organized, preprocessed (resized to 32x32, normalized), and split (80:20 ratio).
  • Optimized hyperparameters for LSTM and CNN models were selected using a trial-and-error approach.
Keywords:
Convolution neural networksLong short term memoryOptical character recognitionPashtu scriptPattern recognition

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 16, 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

496
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
  • Model performance was evaluated using accuracy, loss graphs, classification reports, and confusion matrices.

    • Main Results:

    • Both LSTM and CNN models achieved high accuracy, nearing 98%, in recognizing Pashtu numerics.
    • The LSTM model demonstrated a marginal performance advantage over the CNN model.
    • LSTM achieved a macro-average precision of 0.9877, recall of 0.9876, and F1 score of 0.9876.

    Conclusions:

    • Optimized machine learning models, particularly LSTM, are effective for Pashtu numeric recognition.
    • The proposed models offer a robust solution for overcoming the challenges posed by Pashtu's cursive script.
    • Further research can explore more sophisticated architectures for enhanced Pashtu OCR accuracy.