Exploring Deep Learning and Explainable AI for Precise Rice Leaf Disease Detection and Classification
Objective
The objective of this project is to develop an AI-driven system for the precise detection and classification of rice leaf diseases using deep learning models, including DenseNet121, EfficientNetV2B0, and a hybrid model combining both. The system aims to achieve high accuracy in disease prediction, with DenseNet121 excelling in feature extraction and EfficientNetV2B0 contributing to robustness. The integration of Explainable AI (XAI) techniques, such as Grad-CAM++, ensures transparency and interpretability of the model's decisions. Additionally, a user-friendly Flask-based application is designed to facilitate seamless interaction and provide a comprehensive solution for rice disease management.
Abstract
This project explores the application of deep learning and explainable AI (XAI) for precise detection and classification of rice leaf diseases. A hybrid model combining DenseNet121 and EfficientNetV2B0 is employed to classify rice leaf disease images from the Kaggle dataset "Rice Leaf Diseases." The DenseNet121 model excels in feature extraction, achieving a high validation accuracy of 92.33%, while EfficientNetV2B0, though less effective with a validation accuracy of 33.33%, contributes to enhancing the model's robustness. The hybrid approach further improves the model’s accuracy to 92.33%, demonstrating significant improvement over individual models. The integration of Grad-CAM++ enables interpretability, providing visual explanations for model predictions. A Flask-based application is developed to facilitate easy interaction with the model, featuring modules for Home, Register, Login, Prediction, and Logout, ensuring a complete system workflow for rice leaf disease detection. This approach effectively combines the strengths of DenseNet121 and EfficientNetV2B0, optimizing disease classification performance while providing transparency through explainable AI techniques.
Keywords: Deep Learning, Rice Leaf Disease, DenseNet121, EfficientNetV2, Hybrid Model, Explainable AI, Grad-CAM++, Flask, Classification, Kaggle Dataset
NOTE: Without the concern of our team, please don't submit to the college. This Abstract varies based on student requirements.
Block Diagram
Specifications
Hardware Requirements
The hardware requirements specify the physical resources necessary to run the rice leaf disease detection system effectively. For optimal performance, the following are the recommended hardware specifications:
· Processor: Intel Core i5 or better (Quad-core or higher) for efficient processing and faster model inference.
· Hard Disk: 256GB SSD or higher for faster data storage and access, particularly for storing images and models.
· RAM: 16GB or more to handle large datasets, perform real-time image processing, and run deep learning models efficiently.
· Graphics Card (GPU): NVIDIA GTX 1060 or better with at least 4GB of VRAM for faster training of deep learning models using GPU acceleration.
· Monitor: 1920x1080 resolution or higher for better visualization of images and model outputs.
· Keyboard: Standard Windows Keyboard for easy interaction with the system.
· Mouse: Two or Three Button Mouse for efficient navigation and image upload tasks.
· Network: Stable internet connection for accessing the web interface, uploading images, and downloading model updates or datasets.
Software Requirements
The software requirements specify the environment and tools necessary to develop, run, and deploy the rice leaf disease detection system. The required software components for this project are as follows:
· Operating System: Windows 7/8/10, macOS, or Linux for flexible development and deployment environments.
· Programming Language: Python for model development, training, and deployment, due to its extensive support for deep learning frameworks and libraries.
· Libraries:
o Pandas: For data manipulation and analysis, particularly for handling datasets and preprocessing.
o Numpy: For numerical operations and handling multidimensional arrays.
o Matplotlib/Seaborn: For data visualization to understand dataset distribution and model performance.
o scikit-learn: For machine learning algorithms and evaluation metrics.
o TensorFlow/PyTorch: For deep learning model development, training, and inference.
o Keras: For building deep learning models in a user-friendly manner, running on top of TensorFlow.
o OpenCV: For image processing tasks such as resizing and augmentation.
o Grad-CAM++: For visual explainability of the model’s predictions.
· IDE/Workbench:
o Visual Studio Code: A lightweight and versatile code editor with Python support.
o PyCharm: An IDE optimized for Python development with features like code completion, debugging, and project management.
o Jupyter Notebooks: For experimenting, running code in cells, and visualizing data and model results interactively.
Paper Publishing
Request Call Back
Would you like to receive a free callback now?
Choose the best time for callback:
Leave your message and we will contact you as soon as possible
6-2-85/B, Old Maternity Hospital Road, Thyagaraja Nagar, Tirupati, Andhra Pradesh – 517501
+91 9030333433
+91 9393939065
0877-2261612
Disclaimer - Takeoff Edu Group Projects are not associated or affiliated with IEEE in any way. The IEEE Projects mentioned here are mentioned in the context of student projects, whose ideas are derived from IEEE publications, not projects of or by IEEE.
