Panel-Segmentation
- National Renewable Energy Lab. (NREL), Golden, CO (United States)
The Panel-Segmentation package is a computer vision (CV) framework for automated metadata extraction for solar PV arrays using satellite imagery. With Panel-Segmentation, a user inputs a set of latitude-longitude coordinates to automatically generate a satellite image using the Google Maps API, and runs the image through a deep learning framework to get a pixel-by-pixel representation of the solar panel(s) in an image. If a solar array is located, the user can then perform spectral clustering to cluster individual solar arrays in an image (if applicable), and run each cluster through an azimuth estimation algorithm. This package framework is versatile, allowing for future CV techniques to be added and/or modified.
- Site Accession Number:
- SWR-21-18
- Software Type:
- Scientific
- License(s):
- MIT License
- Programming Language(s):
- Python
- Research Organization:
- National Renewable Energy Laboratory (NREL), Golden, CO (United States)
- Sponsoring Organization:
- USDOE Office of Energy Efficiency and Renewable Energy (EERE), Renewable Power Office. Solar Energy Technologies OfficePrimary Award/Contract Number:AC36-08GO28308
- DOE Contract Number:
- AC36-08GO28308
- Code ID:
- 49010
- OSTI ID:
- code-49010
- Country of Origin:
- United States
Similar Records
Panel-Segmentation: A Python Package for Automated Solar Array Metadata Extraction Using Satellite Imagery
Quantifying Error in Photovoltaic Installation Metadata: Preprint
Conference
·
Fri Aug 26 00:00:00 EDT 2022
·
OSTI ID:1885131
Quantifying Error in Photovoltaic Installation Metadata: Preprint
Conference
·
Mon Jul 22 00:00:00 EDT 2024
·
OSTI ID:2406864