Hybrid Deep Learning Framework for Real-Time Sugarcane Disease Detection

DOI

10.2991/978-94-6463-740-3_7How to use a DOI?

Keywords

Sugarcane Plant Disease (SPD); Deep learning (DL); Machine Learning (ML); Convolutional Neural Networks (CNNs); Gated Recurrent Unit (GRU); Precision Agriculture

Abstract

Sugarcane is a vital crop in global agriculture, yet it is highly susceptible to diseases that can drastically affect crop productivity and disrupt agricultural planning. Traditional methods for identifying and diagnosing diseases in sugarcane leaves rely heavily on manual inspections, which are labor-intensive, slow, and prone to human error. To overcome these limitations, this study introduces a dynamic, real-time disease detection framework that integrates convolutional neural networks (CNNs) with Gated Recurrent Units (GRUs). Using a publicly available dataset from Kaggle containing a diverse range of sugarcane disease images, this hybrid model architecture was trained and evaluated to provide accurate and efficient disease classification. Three hybrid models—DenseNet201 + GRU, Inception v3 + GRU and InceptionResNetv2 + GRU— were implemented and compared. The DenseNet201 + GRU model achieved the highest performance, with an accuracy of 96.49%, precision of 96.53%, recall of 96.49% and F1-score of 96.49%. These results emphasize the advantages of combining CNNs and GRUs: CNNs excel at extracting spatial features from complex disease patterns in leaf images, while GRUs effectively capture sequential dependencies, enhancing the model’s ability to classify diseases with high precision. The proposed hybrid approach provides a scalable and reliable solution for automated sugarcane disease detection. By incorporating this system into disease management workflows, agricultural stakeholders gain access to timely and accurate diagnostic insights, enabling proactive measures to reduce crop loss and enhance overall yield. This framework exemplifies the potential of advanced deep learning architectures in precision agriculture, offering a sustainable and practical tool for improving disease management in sugarcane cultivation.

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  - Satyam Kumar
AU  - Lakshya Sharma
AU  - Deepti Sharma
PY  - 2025
DA  - 2025/06/25
TI  - Hybrid Deep Learning Framework for Real-Time Sugarcane Disease Detection
BT  - Proceedings of the 6th International Conference on Deep Learning, Artificial Intelligence and Robotics (ICDLAIR 2024)
PB  - Atlantis Press
SP  - 63
EP  - 73
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-740-3_7
DO  - 10.2991/978-94-6463-740-3_7
ID  - Kumar2025
ER  -