This project focuses on developing a deep learning model for classifying remote sensing data using the UC Merced Land Use Dataset. The dataset consists of images from various urban areas, manually extracted and categorized into 21 classes. The goal is to build a robust model capable of accurately identifying different land use types from aerial imagery.
- Data Preprocessing: Standardization and augmentation techniques were applied to ensure consistency and enhance model performance.
- Model Development: Implemented a Convolutional Neural Network (CNN) architecture optimized for remote sensing image classification.
- Evaluation Metrics: Model performance was evaluated using accuracy, precision, recall, and F1-score metrics on a separate validation set.
- Deployment: Plans for deploying the trained model into a real-time classification system to aid urban planning and environmental monitoring efforts.