Preprint Dataset Version 2 Preserved in Portico This version is not peer-reviewed

A Season Independent U-net Model for Robust Mapping of Solar Arrays Using Sentinel-2 Imagery

Version 1 : Received: 20 May 2020 / Approved: 21 May 2020 / Online: 21 May 2020 (11:41:37 CEST)
Version 2 : Received: 21 July 2021 / Approved: 21 July 2021 / Online: 21 July 2021 (14:53:14 CEST)

How to cite: Evans, M.; Minich, T.; Soobitsky, R.; Mainali, K. A Season Independent U-net Model for Robust Mapping of Solar Arrays Using Sentinel-2 Imagery. Preprints 2020, 2020050345 (doi: 10.20944/preprints202005.0345.v2). Evans, M.; Minich, T.; Soobitsky, R.; Mainali, K. A Season Independent U-net Model for Robust Mapping of Solar Arrays Using Sentinel-2 Imagery. Preprints 2020, 2020050345 (doi: 10.20944/preprints202005.0345.v2).

Abstract

We have an unprecedented ability to map the Earth’s surface as deep learning technologies are applied to an abundance of high-frequency Earth observation data. Simple, free, and effective methods are needed to enable a variety of stakeholders to use these tools to improve scientific knowledge and decision making. Here we present a trained U-Net model that can map and delineate ground mounted solar arrays using publicly available Sentinel-2 imagery, and that requires minimal data pre-processing and no feature engineering. By using label overloading and image augmentation during training, the model is robust to temporal and spatial variation in imagery. The trained model achieved a precision and recall of 91.5% each and an intersection over union of 84.3% on independent validation data from two distinct geographies. This generalizability in space and time makes the model useful for repeatedly mapping solar arrays. We use this model to delineate all ground mounted solar arrays in North Carolina and the Chesapeake Bay watershed to illustrate how these methods can be used to quickly and easily produce accurate maps of solar infrastructure.

Supplementary and Associated Material

https://osf.io/dau8w/: Open Science Framework repository containing training data, output data, and model weights files
https://github.com/mjevans26/Satellite_ComputerVision: GitHub repository containing code for sampling earth observation imagery, model training, and prediction

Subject Areas

computer vision; deep learning; Earth Engine; remote sensing; renewable energy; Tensorflow

Comments (1)

Comment 1
Received: 21 July 2021
Commenter: Michael Evans
Commenter's Conflict of Interests: Author
Comment: In this updated version we provide readers with access to the trained U-Net model that delineates ground mounted solar arrays from Sentinel-2 data, as well as outputs from this model across North Carolina and the Chesapeake Bay watershed. In conjunction on this focus around providing the model and it's outputs to practicioners we provide additional details about the model structure, its training, and how readers can implement it themselves, as well as a more thorough analysis of its performance on validation data.
+ Respond to this comment

We encourage comments and feedback from a broad range of readers. See criteria for comments and our diversity statement.

Leave a public comment
Send a private comment to the author(s)
Views 0
Downloads 0
Comments 1
Metrics 0


×
Alerts
Notify me about updates to this article or when a peer-reviewed version is published.
We use cookies on our website to ensure you get the best experience.
Read more about our cookies here.