Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
June 9, 2026Scientific ReportsOpen Access

Hybrid metaheuristic optimization of convolutional neural networks for tomato leaf disease classification

View Full Paper
Ask AI
Bookmark
Share

Authors

RSRajsimar SinghLSLaw Kumar SinghASAkhilesh Sharma

Discussion

Loading...

Member takes

Overview

Randomized trial evaluates hybrid metaheuristic strategies for CNN hyperparameter tuning in tomato leaf disease detection, suggesting improved accuracy.

Key Points

  • This study aims to enhance CNN performance for tomato leaf disease classification through advanced hyperparameter tuning methods.
  • Evaluated four hybrid metaheuristic strategies for tuning CNNs: ALO-WOA, ALO-DA, ALO-PSO, PSO-WOA.
  • Applied methods to a dataset of 21,421 training, 4,586 validation, and 4,602 test images across 10 classes.
  • CNN architecture included three convolutional blocks with tunable dropout and learning rate.
  • ALO-DA achieved the highest test accuracy of 97.83%, followed by ALO-WOA (97.67%) and PSO-WOA (97.52%).
  • ALO-PSO resulted in a lower test accuracy of 95.26%.
  • Hybrid metaheuristics showed more effective hyperparameter searching compared to individual optimizers.

Cite This Study

Singh et al. (2026) studied this question.

synapsesocial.com/papers/6a27adf8a963992e16268160https://doi.org/10.1038/s41598-026-54355-w
View Full Paper
Ask AI
Bookmark
Share