The week's pick
Random Articles
Reseach Article
Efficient Convolutional Neural Network for Real-Time Crop Disease Detection in Precision Agriculture
| International Journal of Computer Applications |
| Foundation of Computer Science (FCS), NY, USA |
| Volume 187 - Number 98 |
| Year of Publication: 2026 |
| Authors: Aarthika Anil Birajdar, Vaishnavi Sanjay Dhuttarge, Siddheshwari Kailas Degaonkar, Anuja Macchindranath Gurav, Sharvari Ravikiran Garad, Ojausvi Ajit Bhave, Aakash Shivdas Chatake, Neeta Alange |
10.5120/ijca02c2de6188d7
|
Aarthika Anil Birajdar, Vaishnavi Sanjay Dhuttarge, Siddheshwari Kailas Degaonkar, Anuja Macchindranath Gurav, Sharvari Ravikiran Garad, Ojausvi Ajit Bhave, Aakash Shivdas Chatake, Neeta Alange . Efficient Convolutional Neural Network for Real-Time Crop Disease Detection in Precision Agriculture. International Journal of Computer Applications. 187, 98 ( Apr 2026), 47-51. DOI=10.5120/ijca02c2de6188d7
Abstract
Early detection of crop diseases is essential for minimizing agricultural losses and optimizing pesticide application in precision agriculture. Traditional crop monitoring methods rely heavily on manual inspection by agricultural experts, which can be time‑consuming, inconsistent and often inaccessible to small‑scale farmers. This study proposes a lightweight convolutional neural network (CNN) architecture for real‑time crop disease detection using leaf images. The model is trained on a curated subset of the PlantVillage dataset containing balanced samples of healthy and diseased leaves from multiple crop categories. Systematic data augmentation and rigorous hyperparameter tuning are employed to ensure generalization across varying environmental conditions. The proposed model achieves competitive accuracy while maintaining a compact footprint (<4 MB) and fast inference (<100 ms), making it suitable for deployment on resource‑constrained edge devices. Comprehensive experiments demonstrate the effectiveness of the approach, and comparative analysis shows favorable trade‑offs between accuracy, model size and latency relative to established architectures.
References
- FAO, “The State of Food and Agriculture 2019,” United Nations Food and Agriculture Organization, Rome, 2019.
- A. Krizhevsky, I. Sutskever and G. Hinton, “ImageNet classification with deep convolutional neural networks,” in Proc. NIPS, pp. 1095–1105, 2012.
- J. Yosinski, H. Lipson, J. Clune and Y. Bengio, “How transferable are features in deep neural networks?” in Proc. NIPS, pp. 3320–3328, 2014.
- M. Long, M. Jordan, J. Wang and Z. Cao, “Learning transferable features with deep adaptation networks,” in Proc. ICML, pp. 97–105, 2013.
- J. Dean, G. Hinton and O. Vinyals, “Distilling the knowledge in a neural network,” arXiv:1503.02531, 2015.
- S. Han, W. Dally and H. Mao, “Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding,” arXiv:1510.00149, 2015.
- M. Tan and Q. Le, “EfficientNet: Rethinking model scaling for convolutional neural networks,” in Proc. ICML, pp. 6105–6114, 2019.
- M. Zhu, A. Chen, A. Howard et al., “MobileNets: Efficient convolutional neural networks for mobile vision applications,” arXiv:1704.04861, 2017.
- A. Ramcharan, P. McCloskey and E. Baranowski, “Image-based cassava disease detection using deep learning,” Frontiers in Plant Science, vol. 10, Art. 1852, 2019.
- D. Hughes, S. Mohanty and M. Salathé, “An open access repository of images on plant health to enable the development of mobile disease diagnostics,” Frontiers in Plant Science, vol. 7, Art. 1419, 2016
Index Terms
Keywords