Panel Segmentation: A Python Package for Automated Solar Array Metadata Extraction Using Satellite Imagery
Abstract
The National Renewable Energy Laboratory (NREL) Python panel-segmentation package is a toolkit that automates the process of extracting accurate and valuable metadata related to solar array installations, using publicly available Google Maps satellite imagery. Previously published work includes automated azimuth estimation for individual solar installations in satellite images. Our continued research focuses on automated detection and classification of solar installation mounting configuration (tracking or fixed-tilt; rooftop, ground, or carport). Specifically, a faster-region-based convolutional neural network Resnet-50 feature pyramid network model was trained and validated on 862 manually labeled satellite images. Additionally, this model was used to perform object detection on satellite imagery, locating and classifying individual solar installations' mounting configuration and type. Model results showed a mean average precision score of 77.79%, with the model strongest at detecting fixed-tilt ground mount and fixed-tilt carport installations. The object detection model and its outputs have been incorporated into the panel-segmentation package's automated metadata extraction pipeline, which returns the mounting configuration and azimuth for individual solar arrays in satellite imagery. The complete image dataset with labels has been released on the U.S. Department of Energy (DOE) DuraMAT DataHub, to encourage further research in this area.
- Authors:
-
- NREL, Golden, CO, USA
- Publication Date:
- Research Org.:
- National Renewable Energy Laboratory (NREL), Golden, CO (United States)
- Sponsoring Org.:
- USDOE Office of Energy Efficiency and Renewable Energy (EERE), Renewable Power Office. Solar Energy Technologies Office
- OSTI Identifier:
- 1957702
- Alternate Identifier(s):
- OSTI ID: 1915249; OSTI ID: 1957703
- Report Number(s):
- NREL/JA-5K00-82613
Journal ID: ISSN 2156-3381; 10008194
- Grant/Contract Number:
- AC36-08GO28308; 38258
- Resource Type:
- Published Article
- Journal Name:
- IEEE Journal of Photovoltaics
- Additional Journal Information:
- Journal Name: IEEE Journal of Photovoltaics Journal Volume: 13 Journal Issue: 2; Journal ID: ISSN 2156-3381
- Publisher:
- Institute of Electrical and Electronics Engineers
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING; 14 SOLAR ENERGY; deep learning; metadata extraction; satellite imagery; solar; Azimuth; metadata; mount; satellites; pipelines; photovoltaic systems; object detection; training
Citation Formats
Perry, Kirsten, and Campos, Christopher. Panel Segmentation: A Python Package for Automated Solar Array Metadata Extraction Using Satellite Imagery. United States: N. p., 2023.
Web. doi:10.1109/JPHOTOV.2022.3230565.
Perry, Kirsten, & Campos, Christopher. Panel Segmentation: A Python Package for Automated Solar Array Metadata Extraction Using Satellite Imagery. United States. https://doi.org/10.1109/JPHOTOV.2022.3230565
Perry, Kirsten, and Campos, Christopher. Wed .
"Panel Segmentation: A Python Package for Automated Solar Array Metadata Extraction Using Satellite Imagery". United States. https://doi.org/10.1109/JPHOTOV.2022.3230565.
@article{osti_1957702,
title = {Panel Segmentation: A Python Package for Automated Solar Array Metadata Extraction Using Satellite Imagery},
author = {Perry, Kirsten and Campos, Christopher},
abstractNote = {The National Renewable Energy Laboratory (NREL) Python panel-segmentation package is a toolkit that automates the process of extracting accurate and valuable metadata related to solar array installations, using publicly available Google Maps satellite imagery. Previously published work includes automated azimuth estimation for individual solar installations in satellite images. Our continued research focuses on automated detection and classification of solar installation mounting configuration (tracking or fixed-tilt; rooftop, ground, or carport). Specifically, a faster-region-based convolutional neural network Resnet-50 feature pyramid network model was trained and validated on 862 manually labeled satellite images. Additionally, this model was used to perform object detection on satellite imagery, locating and classifying individual solar installations' mounting configuration and type. Model results showed a mean average precision score of 77.79%, with the model strongest at detecting fixed-tilt ground mount and fixed-tilt carport installations. The object detection model and its outputs have been incorporated into the panel-segmentation package's automated metadata extraction pipeline, which returns the mounting configuration and azimuth for individual solar arrays in satellite imagery. The complete image dataset with labels has been released on the U.S. Department of Energy (DOE) DuraMAT DataHub, to encourage further research in this area.},
doi = {10.1109/JPHOTOV.2022.3230565},
journal = {IEEE Journal of Photovoltaics},
number = 2,
volume = 13,
place = {United States},
year = {Wed Mar 01 00:00:00 EST 2023},
month = {Wed Mar 01 00:00:00 EST 2023}
}
https://doi.org/10.1109/JPHOTOV.2022.3230565
Works referenced in this record:
Field Comparison Study of Fixed-Tilted and Single-Axis Tracking PV Structures in the Desert Environment of Dubai, UAE
conference, June 2020
- Safieh, Ahmad; Elnosh, Ammar; Kaiss, El-Cheikh Amer K.
- 2020 47th IEEE Photovoltaic Specialists Conference (PVSC)
PV Degradation – Mounting & Temperature
conference, June 2019
- Jordan, Dirk C.; Deline, Chris; Deceglie, Michael
- 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC)
A Survey on Performance Metrics for Object-Detection Algorithms
conference, July 2020
- Padilla, Rafael; Netto, Sergio L.; da Silva, Eduardo A. B.
- 2020 International Conference on Systems, Signals and Image Processing (IWSSIP)
Assessment of rooftop photovoltaic potentials at the urban level using publicly available geodata and image recognition techniques
journal, October 2017
- Mainzer, Kai; Killinger, Sven; McKenna, Russell
- Solar Energy, Vol. 155
SolarFinder: Automatic Detection of Solar Photovoltaic Arrays
conference, April 2020
- Li, Qi; Feng, Yuzhou; Leng, Yuyang
- 2020 19th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN)
Microsoft COCO: Common Objects in Context
book, January 2014
- Lin, Tsung-Yi; Maire, Michael; Belongie, Serge
- ECCV 2014: Computer Vision – ECCV 2014
Unsupervised azimuth estimation of solar arrays in low-resolution satellite imagery through semantic segmentation and Hough transform
journal, September 2021
- Edun, Ayobami S.; Perry, Kirsten; Harley, Joel B.
- Applied Energy, Vol. 298
AutoAugment: Learning Augmentation Strategies From Data
conference, June 2019
- Cubuk, Ekin D.; Zoph, Barret; Mane, Dandelion
- 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Automatic Detection and Mapping of Solar Photovoltaic Arrays with Deep Convolutional Neural Networks in High Resolution Satellite Images
conference, October 2020
- He, Kaiji; Zhang, Long
- 2020 IEEE 4th Conference on Energy Internet and Energy System Integration (EI2)
Automatic detection of solar photovoltaic arrays in high resolution aerial imagery
journal, December 2016
- Malof, Jordan M.; Bradbury, Kyle; Collins, Leslie M.
- Applied Energy, Vol. 183
3D-PV-Locator: Large-scale detection of rooftop-mounted photovoltaic systems in 3D
journal, March 2022
- Mayer, Kevin; Rausch, Benjamin; Arlt, Marie-Louise
- Applied Energy, Vol. 310
DeepSolar: A Machine Learning Framework to Efficiently Construct a Solar Deployment Database in the United States
journal, December 2018
- Yu, Jiafan; Wang, Zhecheng; Majumdar, Arun
- Joule, Vol. 2, Issue 12