SPADE OC2 Mediterranean Oat crop dataset (B3) - VITAFields – Visual Intelligence for Targeted Agriculture #2

This dataset release represents Part 2 of the comprehensive young oat crop agricultural analysis project. While Part 1 provided the raw image captures and camera parameters, Part 2 delivers fully processed, georeferenced orthomosaics and machine-learning-ready image tiles. This processed data is ideal for training computer vision models directly on map-based data for field-scale semantic segmentation, weed/crop classification, and precise spatial analysis. Dataset Characteristics Target Crop: Oats Growth Stage: Young / Early Vegetative Image Types: Processed Orthomosaics and 512 x 512 Image Tiles (RGB and 4-Band Multispectral) Technical Specifications 1 RGB Orthomosaic 34816 x 36864pixels 4-Band Multispectral Orthomosaic 34816 x 36864 pixels (Green, Red, Red Edge, and Near-Infrared) 3030 Total Orthomosaic Tiles 512 x 512 pixels (RGB and 4-Band Multispectral) Annotations JSON FIles in LabelMe Format (Polygons)Masks Binary images PNG format