Siyathembatshabalala7/Illegal-Dumping-Site-Detection-System-using-deep-learning

Project aims to help fight illegal dumping in South Africa. This project aims to help fight illegal dumping in South Africa by leveraging modern deep learning and computer vision techniques. Using the YOLO framework and a publicly available dataset from Kaggle, we trained a model capable of identifying and mapping illegal dumping sites from satellite imagery. The model’s efficiency was evaluated within the South African context using randomly selected satellite images from known dumping hotspots. The system successfully detected several illegal dumping sites. To make the system more accessible, we also developed a user-friendly interface that allows users to: 1 Upload a satellite image of an area of interest. 2 Run the detection model to automatically identify illegal dumping sites. 3 Visualize and interpret the results directly within the application.