Onion Leaf Disease Classification with Attention-Integrated EfficientNet

DOI

10.2991/978-94-6463-948-3_48How to use a DOI?

Keywords

Onion crop disease; deep learning; convolutional neural network; soft attention; precision agriculture

Abstract

Onion is a common and widely grown crop and has its own significance. But it is highly vulnerable to many leaf diseases such as purple blotch, downy mildew, and several other disease. Routine field survey is still the most common way to detect these problems, but it is subjective, slow and is not suitable for large areas. Although machine learning and deep learning techniques have shown good potential, many existing models are still trained on limited datasets and are rarely tested in real farming environments.

This study focuses on two main contributions. First, it presents a structured review of 21 key research works related to onion disease detection, including CNN-based models, IoT monitoring systems, and epidemiological surveys. Second, it proposes a new framework that combines an EfficientNet backbone with a soft attention mechanism to improve feature exactraction. The model was trained and tested on a custom image dataset collected from onion fields under different environments and lighting conditions. The proposed system achieved a classification accuracy of 96.86%, demonstrating traditional Machine Learning techniques and standard CNN baselines. Transfer learning technique improve stability in different conditions, and the lightweight architecture makes it suitable for deployment on mobile devices making it useful for the farmers. Overall, this framework provides a practical and scalable approach for modern agriculture.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Vinaya Kulkarni
AU  - Sanjesh Pawale
AU  - Kirti Wanjale
PY  - 2026
DA  - 2026/01/06
TI  - Onion Leaf Disease Classification with Attention-Integrated EfficientNet
BT  - Proceedings of the International Conference on Sustainable Innovation with Artificial Intelligence and Machine Learning 2025 (ICSIAIML 2025)
PB  - Atlantis Press
SP  - 692
EP  - 704
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-948-3_48
DO  - 10.2991/978-94-6463-948-3_48
ID  - Kulkarni2026
ER  -