Python library for YOLOv8 and YOLOv9 small object detection and instance segmentation

Project to investigate and develop spatial data-driven Geo-AI models (Convolutional Neural Network) to identify urban greenspace from Satellite images by integrating multiple data sources (e.g. vector data of urban parks)

Efficiently handles large images by segmenting them into patches, performing instance/semantic segmentation, and then reconstructing the patches into a complete, segmented mask.