Abstract: The project introduces a deep learning-based web application for plant identification and disease detection, addressing the need for precision agriculture and timely crop health assessment. The application uses a convolutional neural network (MobileNetV2) trained on a dataset of plant leaves, allowing users to upload images for real-time analysis. Multiple image preprocessing techniques, such as Gaussian blur, histogram equalization, and segmentation, enhance accuracy. The model outputs the most probable class label and a confidence score. The Flask web interface is user-friendly, ensuring accessibility for both professionals and general users. Implemented using PyTorch and OpenCV, the system is lightweight, scalable, and can be deployed locally or in the cloud. The application aims to assist farmers, gardeners, and researchers in early disease detection and support timely intervention to reduce crop loss.

Keywords: Deep Learning, Plant Disease Detection, Plant Identification, MobileNetV2, Convolutional Neural Network (CNN), Image Classification, Precision Agriculture, PyTorch, OpenCV, Image Preprocessing, Real-Time Analysis, Flask Web Application, Computer Vision, Crop Health Monitoring, Agricultural Technology.

| DOI: 10.17148/IARJSET.2025.125308