Explore the datasets and CityGML pre-processing code developed for the 3D-PV-Locator paper. When building on this work, please cite our work as indicated below.
This repository provides you with three demo notebooks.
This notebook allows you to visualize the locations of all the images in our classification dataset on an interactive map.
For example, the locations of non-PV images in our training dataset are illustrated below:
This notebook allows you to create your own dataset by downloading images from the openNRW server directly.
This notebook demonstrates the code to extract 3D rooftop information from the CityGML files provided by the state of North Rhine-Westphalia (here)
For example, after processing the exemplary .gml files in data/GML/, we can load the extracted rooftop polygons with their respective attributes in QGIS:
The repository requires the packages GeoPandas, Scipy, Geopy, Pandas, Numpy, Lxml, and related packages used.
Please cite our work as
@article{MAYER2022,
title = {3D-PV-Locator: Large-scale detection of rooftop-mounted photovoltaic systems in 3D},
journal = {Applied Energy},
volume = {310},
pages = {118469},
year = {2022},
issn = {0306-2619},
doi = {https://doi.org/10.1016/j.apenergy.2021.118469},
url = {https://www.sciencedirect.com/science/article/pii/S0306261921016937},
author = {Kevin Mayer and Benjamin Rausch and Marie-Louise Arlt and Gunther Gust and Zhecheng Wang and Dirk Neumann and Ram Rajagopal},
keywords = {Solar panels, Renewable energy, Image recognition, Deep learning, Computer vision, 3D building data, Remote sensing, Aerial imagery},
}