Lightweight Deep Learning System for Multi-Crop Leaf Disease Detection and Classification in Realtime Environments

Sridevi Sakhamuri, Tadi Chandrasekhar, Y. Mohamed Badcha, M. Silpa Raj, Marrapu Aswini Kumar, Abinaya T.

2025

Abstract

A lightweight deep learning model for online detection and classification of multi-crop plant leaf diseases is proposed in this work. Adapted for optimization with the construction of the convolutional neural network, and combined with a scalable data augmentation pipeline in the streaming encoder-decoder, the system guarantees high recognition accuracy and low computational cost, which is suitable for edge devices (e.g., smartphones and drones). The model is trained on geographically varied dataset and is equipped with explainability module to provide visual cues on disease localization. Experiments show that our approach achieves better performance than the traditional models, especially in different environmental light and background conditions, and thus has practical value for the farmers and the agronomists.

Paper Citation

in Harvard Style

Sakhamuri S., Chandrasekhar T., Badcha Y., Raj M., Kumar M. and T. A. (2025). Lightweight Deep Learning System for Multi-Crop Leaf Disease Detection and Classification in Realtime Environments. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - Volume 1: ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 257-263. DOI: 10.5220/0013862300004919

in Bibtex Style

@conference{icrdicct`2525,
author={Sridevi Sakhamuri and Tadi Chandrasekhar and Y. Badcha and M. Raj and Marrapu Kumar and Abinaya T.},
title={Lightweight Deep Learning System for Multi-Crop Leaf Disease Detection and Classification in Realtime Environments},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - Volume 1: ICRDICCT`25},
year={2025},
pages={257-263},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013862300004919},
isbn={978-989-758-777-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - Volume 1: ICRDICCT`25
TI - Lightweight Deep Learning System for Multi-Crop Leaf Disease Detection and Classification in Realtime Environments
SN - 978-989-758-777-1
AU - Sakhamuri S.
AU - Chandrasekhar T.
AU - Badcha Y.
AU - Raj M.
AU - Kumar M.
AU - T. A.
PY - 2025
SP - 257
EP - 263
DO - 10.5220/0013862300004919
PB - SciTePress