Deep Convolutional Framework for Multi-Crop Disease Identification

DOI

10.2991/978-94-6239-791-0_16How to use a DOI?

Keywords

Crop Disease Detection; Machine Learning; Precision Agriculture; Image Processing; Artificial Intelligence

Abstract

Early and accurate crop disease identification and categorization are crucial problems for increasing agricultural productivity. Instead of seeking assistance with a professional diagnosis, farmers frequently just observe the plants. This article presents deep learning approaches based on convolutional models for the detection and classification of crop diseases. With multiple convolution and pooling layers, the model progressively increases channel width while reducing the spatial dimensions of the feature maps, enabling the extraction of more discriminative features. Such an approach captures detailed changes, such as the textures and color shifts, as well as the alteration of the edges of necrotic or dead tissue. The final classification step is achieved with a series of highly interconnected layers using softmax, where the outputs are optimized with categorical cross-entropy. The system described is highly accurate and does not suffer from overfitting. The proposed parameterization enables a rapid training system, and the splits have good reproducibility. A reduction in the size of the model to retain rapid inferencing and accurate classification of closely related diseases is possible. A robust, simple, explicit framework, and rapid early-stage diagnosis of disease, enhances the agricultural system and minimizes crop loss. In resource-constrained environments, the system reduces wasteful applications of pesticides and provides expert assessments of plant health.

Copyright

© 2026 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  - Santosh Nivrutti Shelke
AU  - Shrikant Salve
AU  - Mahendra Pratap Yadav
AU  - Pavitha Nooji
AU  - Mahendra K. Nivangune
PY  - 2026
DA  - 2026/10/09
TI  - Deep Convolutional Framework for Multi-Crop Disease Identification
BT  - Proceedings of the International Conference on Artificial Intelligence and Machine Learning for Agro-Ecosystems (ICAIML-AE 2026)
PB  - Atlantis Press
SP  - 188
EP  - 197
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-791-0_16
DO  - 10.2991/978-94-6239-791-0_16
ID  - Shelke2026
ER  -