Deep Learning for Sugarcane Disease Detection: A Field-Validated EfficientNet-B4 Approach
Keywords:
Deep learning, Sugarcane Diseases, CNN, ResNet, Image ClassificationAbstract
Abstract
Sugarcane (Saccharum officinarum) serves as a critical economic mainstay in Pakistan, where it ranks among the top agricultural commodities. Devastating fungal and bacterial infections – particularly red rot (Colletotrichum falcatum), smut (Sporisorium scitamineum), and leaf scald (Xanthomonas albilineans) – inflict annual yield losses exceeding 20-50%, translating to crippling financial blows surpassing $350 million. Conventional disease identification techniques remain impractical for most farmers: they demand specialized expertise, involve time-consuming laboratory processes (typically 3-5 days), and prove cost-prohibitive across rural regions where sugarcane cultivation predominates.
This research pioneers an intelligent vision-based diagnostic system leveraging deep convolutional neural networks (CNNs) to automate sugarcane disease recognition directly from field imagery. To overcome dataset limitations endemic to agricultural AI applications, we compiled a comprehensive repository of 15,000 high-resolution field images capturing healthy and diseased specimens across diverse growth stages and environmental conditions. Strategic data augmentation through geometric transformations (multi-axis rotation, flipping, scaling) and photometric adjustments (variable brightness/contrast) expanded this corpus to 26,500 training samples. Three state-of-the-art architectures were rigorously evaluated: custom 8-layer CNN, ResNet50, and EfficientNet-B4, with the latter fine-tuned using transfer learning principles initialized on ImageNet weights.
The optimized EfficientNet-B4 model demonstrated exceptional proficiency, achieving 94% classification accuracy, 92% recall, and a 91% F1-score during real-world field validation – significantly outperforming traditional methods. Deployment occurs through an intuitive cross-platform application (mobile/web) enabling farmers to capture leaf/stem images and receive instant diagnoses (<2 seconds) with management recommendations, even offline. This work delivers three pivotal contributions: 1) A novel disease detection framework validated under operational farm conditions, 2) Public release of the largest curated sugarcane pathology image dataset to date, and 3) A farmer-centric tool advancing UN Sustainable Development Goals (SDGs) – specifically SDG 2 (Zero Hunger) through yield protection and SDG 9 (Industry, Innovation) by democratizing cutting-edge AI for precision agriculture.
Keywords: Deep Learning, Sugarcane Diseases, CNN, ResNet, EfficientNet, Image Classification, Precision Agriculture, SDGs.
References
Pakistan Bureau of Statistics, Agricultural Statistics 2023. Islamabad: GOP Press, 2023.
FAO, Sugarcane Production Systems. Rome: Food and Agriculture Organization, 2022.
R. Viswanathan, "Pathological diagnosis limitations in sugarcane," Plant Pathol. J., vol. 38, no. 2, pp. 145–158, 2021.
S. Ramesh et al., "Cost-effectiveness analysis of serological diagnostics," J. Plant Dis., vol. 129, pp. 112–125, 2022.
A. Khan et al., "Molecular diagnostic barriers in developing economies," Crop Prot., vol. 157, p. 105982, 2022.
M. F. A. Rahman et al., "Color feature extraction for plant disease identification," Comput. Electron. Agric., vol. 168, p. 105108, 2020.
S. Phadikar et al., "Texture-based rice disease classification," Int. J. Agric. Technol., vol. 14, no. 7, pp. 1159–1172, 2018.
J. G. A. Barbedo, "Plant disease segmentation in digital images," Multimedia Tools Appl., vol. 80, pp. 12447–12467, 2021.
V. Prakash et al., "SVM-based sugarcane rust detection," Sugar Tech, vol. 22, pp. 132–140, 2020.
K. P. Ferentinos, "Pre-deep learning accuracy ceilings," Biosyst. Eng., vol. 210, pp. 22–35, 2021.
L. G. Almeida et al., "Satellite monitoring of sugarcane smut," Remote Sens. Appl., vol. 28, p. 100825, 2022.
Y. Zhang et al., "Hyperspectral detection of sugarcane diseases," Plant Phenomics, vol. 2021, p. 9878549, 2021.
R. A. Johnson et al., "Economic barriers in precision agriculture," Agron. J., vol. 114, no. 2, pp. 1021–1033, 2022.
S. P. Mohanty et al., "Using deep learning for image-based plant disease detection," Front. Plant Sci., vol. 7, p. 1419, 2016.
D. P. Hughes et al., "The PlantVillage dataset," arXiv:1511.08060, 2015.
K. P. Ferentinos, "Deep learning models for plant disease detection and diagnosis," Comput. Electron. Agric., vol. 145, pp. 311–318, 2018.
K. He et al., "Deep residual learning for image recognition," in Proc. IEEE CVPR, 2016, pp. 770–778.
R. Singh et al., "ResNet for rice disease detection," IEEE Sensors J., vol. 21, no. 16, pp. 17595–17602, 2021.
M. Tan and Q. V. Le, "EfficientNet: Rethinking model scaling for convolutional neural networks," in Proc. ICML, 2019, pp. 6105–6114.
T. DeVries et al., "Geometric augmentation impact," Pattern Recognit. Lett., vol. 133, pp. 1–7, 2020.
C. Shorten et al., "Photometric augmentation survey," J. Mach. Learn. Res., vol. 22, no. 1, pp. 1–45, 2021.
A. Antoniou et al., "GANs for data augmentation," Neural Comput. Surveys, vol. 20, no. 1, pp. 1–35, 2019.
Y. Wei et al., "CycleGAN for agricultural robustness," Comput. Electron. Agric., vol. 187, p. 106246, 2021.
M. A. Islam et al., "Aerial detection of sugarcane smut using Mask R-CNN," Remote Sens., vol. 13, no. 15, p. 2987, 2021.
S. Khan et al., "CNN-based recognition of sugarcane diseases in controlled environments," Comput. Electron. Agric., vol. 198, p. 107036, 2022.
J. A. Silva et al., "Multi-disease classification in sugarcane using deep learning," Plant Methods, vol. 19, no. 1, p. 54, 2023.
F. Li et al., "The field-lab discrepancy in agricultural AI," Nat. Mach. Intell., vol. 4, pp. 104–114, 2022.
P. T. Karule et al., "Field performance degradation in plant disease models," Precis. Agric., vol. 23, pp. 189–205, 2022.
A. Howard et al., "Searching for MobileNetV3," in Proc. IEEE ICCV, 2019, pp. 1314–1324.
R. Ghosal et al., "Challenges in multi-disease detection for staple crops," Phytopathology, vol. 112, no. 7, pp. 1423–1435, 2022.
S. Kamilaris et al., "Deep learning in agriculture: A survey," Comput. Electron. Agric., vol. 147, pp. 70–90, 2018.
F. Chollet, Deep Learning with Python. Shelter Island, NY: Manning Publications, 2018.
T. DeVries and G. W. Taylor, "Improved regularization of convolutional neural networks with physics-based augmentation," in Proc. IEEE CVPR Workshops, 2021, pp. 81–89.
K. Simonyan and A. Zisserman, "Very deep convolutional networks for large-scale image recognition," arXiv:1409.1556, 2014.
J. Yosinski et al., "How transferable are features in deep neural networks?" in Proc. NIPS, 2014, pp. 3320–3328.
I. Loshchilov and F. Hutter, "Decoupled weight decay regularization," arXiv:1711.05101, 2017.
K. P. Ferentinos, "Recall-oriented evaluation in plant disease diagnostics," Comput. Electron. Agric., vol. 203, p. 107521, 2022.
R. Viswanathan and P. Padmanaban, "Morphological diagnostics of sugarcane smut," Plant Dis., vol. 106, no. 8, pp. 2019–2026, 2022.
M. Lopez et al., "PCR-based diagnostic protocols for agricultural pathogens," J. Plant Pathol., vol. 104, no. 3, pp. 345–361, 2022.
G. P. Rao et al., "Differential diagnosis of red rot and leaf scald in sugarcane," Plant Dis., vol. 107, no. 5, pp. 1321–1333, 2023.
Y. Chauhan, N. Negi and S. Sharma, "Artificial Intelligence Based Sugarcane Leaf Disease Prediction System for Smart Farming," 2024 7th International Conference on Circuit Power and Computing Technologies (ICCPCT), Kollam, India, 2024, pp. 106-111, doi: 10.1109/ICCPCT61902.2024.10673170
Srinivasan, S., Prabin, S. M., Mathivanan, S. K., Rajadurai, H., Kulandaivelu, S., & Shah, M. A. (2025). Sugarcane leaf disease classification using deep neural network approach. BMC Plant Biology, 25(1), 282.
Downloads
Published
Versions
- 2025-12-26 (2)
- 2025-12-21 (1)
How to Cite
Issue
Section
License
Copyright (c) 2025 University of Sindh Journal of Information and Communication Technology
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
University of Sindh Journal of Information and Communication Technology (USJICT) follows an Open Access Policy under Attribution-NonCommercial CC-BY-NC license. Researchers can copy and redistribute the material in any medium or format, for any purpose. Authors can self-archive publisher's version of the accepted article in digital repositories and archives.
Upon acceptance, the author must transfer the copyright of this manuscript to the Journal for publication on paper, on data storage media and online with distribution rights to USJICT, University of sindh, Jamshoro, Pakistan. Kindly download the copyright for below and attach as a supplimentry file during article submission
