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

Related Concept Videos

Improving Translational Accuracy

Improving Translational Accuracy

Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...

You might also read

Related Articles

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

Sort by

Same author

Anterior Tibial Tendon Transfer for Relapsed and Residual Clubfoot Deformity and Functional Outcome With Timely Intervention: How Have We Done Over the Last Five Years?

Cureus·2026

Same author

Synergistic integration of Fe-MOF and biomass-derived activated carbon for enhanced ciprofloxacin adsorption: experimental insights and DFT-based mechanistic elucidation.

RSC advances·2026

Same author

Retrospective study on antibiotic resistance profiles of uropathogenic <i>Escherichia coli</i> with emphasis on β-lactamase production in a tertiary hospital in Saudi Arabia.

Frontiers in microbiology·2026

Same author

Evaluating the efficacy of rimegepant as a preventive treatment for chronic and episodic migraine: a three-month longitudinal retrospective cohort study.

Frontiers in neurology·2026

Same author

Mediastinal over-staging with PET-CT in tuberculosis-endemic locally advanced NSCLC: SUV threshold recalibration improves PET/EBUS concordance.

European journal of nuclear medicine and molecular imaging·2026

Same author

Infection following foot and ankle surgery : a subanalysis of data captured from the UK Foot and Ankle Thromboembolism (FATE) audit.

The bone & joint journal·2026

Search research articles

Related Experiment Video

Updated: Oct 3, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

Generative adversarial network based adaptive data augmentation for handwritten Arabic text recognition.

Mohamed Eltay^1,2, Abdelmalek Zidouri^1,2, Irfan Ahmad^2,3

¹Electrical Engineering Department, King Fahd University of Petroleum & Minerals, Dhahran, Saudi Arabia.

Peerj. Computer Science

|February 17, 2022

Summary

This study introduces an adaptive Generative Adversarial Network (GAN) technique to balance character frequencies in handwritten text recognition datasets. This method improves text recognition accuracy by addressing data imbalance, benefiting minority characters.

Keywords:

Adaptive data augmentation Arabic handwriting recognition Convolutional neural networks Deep learning neural Networks Generative adversarial networks Handwritten text generation

More Related Videos

Generating Strictly Controlled Stimuli for Figure Recognition Experiments

Generating Strictly Controlled Stimuli for Figure Recognition Experiments

Published on: March 18, 2019

Related Experiment Videos

Last Updated: Oct 3, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

Generating Strictly Controlled Stimuli for Figure Recognition Experiments

Generating Strictly Controlled Stimuli for Figure Recognition Experiments

Published on: March 18, 2019

Area of Science:

Computer Science
Artificial Intelligence
Machine Learning

Background:

Deep learning for handwritten text recognition requires extensive annotated data.
Data augmentation is crucial for increasing training data volume.
Generative Adversarial Networks (GANs) are effective image augmentation tools but face challenges in text recognition due to data imbalance.

Purpose of the Study:

To present an adaptive GAN-based data augmentation technique for handwritten text recognition.
To address the challenge of class imbalance in text data, where certain characters appear less frequently.
To improve the performance of text recognition systems by generating balanced augmented data.

Main Methods:

Developed an adaptive data augmentation technique using Generative Adversarial Networks (GANs).
Focused on mitigating class imbalance issues inherent in text recognition datasets.
Trained GANs on imbalanced datasets to generate augmented data with balanced character frequencies.

Main Results:

Experimental evaluations on Arabic handwritten text recognition datasets demonstrated the technique's effectiveness.
The proposed GANs successfully generated balanced augmented data, improving representation of minority characters.
Text recognition systems trained on this balanced data showed improved accuracy compared to standard methods.

Conclusions:

The adaptive GAN-based data augmentation effectively addresses class imbalance in text recognition.
Balanced augmented data leads to enhanced performance in handwritten text recognition systems.
This approach offers a viable solution for improving deep learning models in data-scarce and imbalanced scenarios.