Detecting Cotton Plant Diseases Using Deep Learning
Author Information
Author(s): Johri Prashant, Kim SeongKi, Dixit Kumud, Sharma Prakhar, Kakkar Barkha, Kumar Yogesh, Shafi Jana, Ijaz Muhammad Fazal
Primary Institution: Galgotias University
Hypothesis
Can advanced deep transfer learning techniques improve the detection of cotton plant diseases?
Conclusion
The EfficientNetB3 model achieved the highest accuracy of 99.96% in detecting various cotton plant diseases.
Supporting Evidence
- The EfficientNetB3 model outperformed others in accuracy, loss, and root mean square error.
- Other models also showed high precision, recall, and F1 scores close to 0.98 or 1.00.
- The study emphasizes the potential of deep transfer learning for sustainable agriculture.
Takeaway
This study shows that computers can help farmers find diseases in cotton plants quickly and accurately using pictures.
Methodology
The study used deep transfer learning models like EfficientNet, Xception, and ResNet to classify images of healthy and diseased cotton plants.
Potential Biases
Models may overfit due to small or homogeneous datasets.
Limitations
The study faced challenges with accurately identifying regions of interest in images and the computational complexity of the models.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website