Software underlying PhD thesis: AI in the Sky - Advancing Wildlife Survey Methods in Africa with Deep Learning and Aerial Imagery.
ZyPro is a deep learning framework designed to support the PhD research project “AI in the Sky: Advancing Wildlife Survey Methods in Africa with Deep Learning and Aerial Imagery.” The research aims to improve wildlife monitoring using semantic segmentation and object detection on remote sensing imagery. ZyPro provides tools for training, testing, and predicting with U-Net-based neural networks, as well as specialized modules for remote sensing image processing, such as large image handling, multi-channel data processing, image clipping, tiling, and data augmentation. This software project includes custom loss functions, flexible training pipelines, and various utilities to facilitate high-quality analysis of remote sensing data.