InsightNet: A Deep Learning Framework for Enhanced Plant Disease Detection and Explainable Insights

Area of Science:

• Agricultural Science
• Computer Science
• Biotechnology

Background:

• Plant diseases significantly threaten global food security and agricultural sustainability.
• Current disease inspection methods are manual, time-consuming, and lack scalability for modern agriculture.
• Developing automated, accurate plant disease detection is crucial for efficient crop management.

Purpose of the Study:

• To develop and evaluate advanced deep learning models for detecting and classifying plant leaf diseases.
• To enhance the accuracy and efficiency of plant disease diagnosis in key crop species.
• To contribute to the advancement of precision farming and sustainable agricultural practices.

Main Methods:

• A deep learning model utilizing the MobileNet architecture was designed with deeper convolutional layers and dropout regularization.
• The model was trained and validated on datasets for tomato, bean, and chili plant diseases.
• Gradient-weighted Class Activation Mapping (Grad-CAM) was employed for model interpretability.

Main Results:

• The proposed deep learning model achieved high classification accuracies: 97.90% for tomato, 98.12% for bean, and 97.95% for chili plants.
• The model demonstrated effective disease detection and classification capabilities across multiple plant species.
• Grad-CAM provided insights into the model's diagnostic decision-making process.

Conclusions:

• Deep learning models, particularly MobileNet-based architectures, offer a powerful solution for accurate and scalable plant disease diagnosis.
• This research supports the integration of AI in agriculture for improved crop health monitoring and sustainable food production.
• The developed models contribute to precision agriculture by enabling timely and precise identification of plant diseases.