Python library for YOLO 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.

Image SuperResolution using Random Forest Regressor(RFSR), Support Vector Regressor(SVR) Machine learning Models.

Repository for semantic segmentation of aerial imagery using U-Net, featuring training scripts, data preprocessing, and model evaluation.