Phoradendron Velutinum UAV Multispectral and RGB Dataset with Segmentation Masks

This dataset contains co-registered multispectral and RGB aerial images of Phoradendron velutinum (true mistletoe), acquired using a DJI Phantom 4 Multispectral UAV over the San Bartolo Ameyalco Conservation Area in Mexico City. Each image sample includes a five-band multispectral image (.TIF), an RGB image (.JPG), and an expert-annotated binary segmentation mask (.PNG) indicating mistletoe presence. 250 curated image tiles (300 x 300 px) are distributed across five folders. The dataset was first described and used in the scientific article: Mejía-Zuluaga et al. (2022), Remote Sensing, 14(3), 801 (https://doi.org/10.3390/rs14030801). It supports research in remote sensing, vegetation monitoring, and developing machine learning models for plant-parasite detection. Image preprocessing includes radiometric and geometric corrections, and band co-registration using SURF feature matching. Segmentation masks were created from GPS-referenced field samples and manual photointerpretation. This is the first public release of the dataset.