Drone Processed Images for Coastal Litter Detection (RGB, Thermal, Multispectral)

This repository contains a processed image dataset acquired during drone-based surveys for coastal litter pollution detection. The data were collected using RGB, thermal, and multispectral camera payloads flown at multiple altitudes (30 m, 60 m, 90 m, and 120 m) over coastal environments with varying shoreline morphology and litter density. The primary objective of this dataset is to support the development, training and evaluation of image processing and machine learning methods for automatic detection, mapping and monitoring of litter along the coastline. By providing multi-sensor and multi-altitude imagery, the dataset enables users to investigate how spatial resolution, spectral information and observation geometry influence the detectability of different types of coastal litter. The processed images included here are derived from dedicated flights carried out under controlled acquisition conditions (consistent flight plans, overlap settings and sensor configurations). Each image is associated with basic metadata such as sensor type and flight altitude; additional metadata can be added by users as needed for their specific workflows. Potential applications of this dataset include (but are not limited to): Training and testing of coastal litter detection algorithms Benchmarking different feature extraction and classification methods Comparative studies of RGB vs thermal vs multispectral sensing for marine litter Methodological research on optimal flight altitude and sensor configuration for coastal monitoring Users are encouraged to cite this dataset when using it in scientific publications, reports or software tools related to coastal litter monitoring, marine pollution assessment or environmental remote sensing.