Peduli Stroberi

This project is a Python-based decision support system designed to assist farmers in detecting strawberry leaf diseases. By leveraging YOLOv8 computer vision, it addresses agricultural economic threats by providing a technological solution to identify infections from uploaded photos, helping minimize crop yield losses through early intervention.

The goal was to build an automated diagnostic tool to identify three specific strawberry diseases: Leaf Scorch, Leaf Blight, and Leaf Spot. The system aims to provide accessible digital expertise with a 93.15% accuracy rate, offering immediate identification and mitigation strategies to ensure better crop management.

I developed a full-stack web application using Flask and Python, implementing the YOLOv8 algorithm for object detection. The model was trained on a dataset of 582 strawberry leaf images. The workflow includes image preprocessing, model inference for disease classification, and a responsive frontend to display diagnostic results.

The final deliverable is a functional Web-Application for Strawberry Leaf Disease Detection. It features an image upload portal, a YOLOv8-powered detection engine with 93.15% accuracy, and an output dashboard that visualizes detected pathogens along with descriptive treatment information to aid agricultural decision-making.

Object Detection: Implementing YOLOv8 to locate and classify specific disease spots on strawberry leaves.

Supervised Learning: Training the model using a curated dataset of 582 labeled strawberry leaf images.

Inference Pipeline: Developing a backend process that takes uploaded images and returns model predictions instantly.

Responsive UI Design: Creating a mobile-friendly interface to ensure accessibility for farmers in the field.