Home / Articles
| PLANTGUARD AI |
|
|
Author Name Ms. J. Aiswarya and Sanjay Karthi S Abstract Agriculture remains one of the most critical sectors globally, contributing significantly to economic stability and food security. However, plant diseases continue to pose major threats to crop yield, farmer income, and sustainable agricultural development. Early and accurate detection of plant diseases is essential to minimize losses and optimize treatment strategies. This paper presents PlantGuard AI, a deep learning-based full-stack web application designed for automated plant disease detection using transfer learning techniques. The system employs the MobileNetV2 convolutional neural network architecture trained on a dataset of 87,867 plant leaf images categorized into 38 disease classes. The model utilizes image preprocessing techniques including resizing to 224×224 pixels and data augmentation strategies such as rotation, zooming, and horizontal flipping to improve generalization performance. The training process uses the Adam optimizer with a learning rate of 0.001 and categorical cross-entropy as the loss function. The system integrates a React.js 18 and Vite-based frontend for responsive and interactive user experience, styled using TailwindCSS and enhanced with Framer Motion animations. The backend
is developed using FastAPI (Python 3.11+) and exposes RESTful APIs for prediction, authentication, and data management. MongoDB 8.0 serves as the NoSQL database, accessed through PyMongo for persistent storage of user credentials, detection history, and metadata. Authentication is implemented using JWT tokens with bcrypt-based password hashing to ensure secure access control. The application allows users to upload leaf images, obtain real-time disease predictions with confidence scores, receive treatment recommendations, and generate downloadable PDF reports. The proposed system demonstrates how deep learning, modern web technologies, and secure database systems can be integrated into a scalable agricultural decision-support platform.
Keywords:
Plant Disease Detection, Deep Learning, Transfer Learning, MobileNetV2, Convolutional Neural Networks (CNN), Computer Vision, FastAPI, React.js, MongoDB, Full -Stack Web Application, Image Classification, Agricultural Decision Support System, JWT Authentication, Data Augmentation.
Published On : 2026-03-07 Article Download : 
|
|
