Panel Segmentation: A Python Package for Automated Solar Array Metadata Extraction Using Satellite Imagery

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 installations in images [1]. Our continued research focuses on detection classification installation mounting configuration (tracking or fixed-tilt; rooftop, ground, carport). Specifically, faster-region-based convolutional neural network Resnet-50 feature pyramid model was trained validated 862 manually labeled images. This used perform object imagery, locating classifying installations' type. Model results showed mean average precision score 77.79%, with strongest at detecting fixed-tilt ground mount carport installations. its outputs have been incorporated into package's extraction pipeline, which returns arrays imagery [2]. complete image dataset labels has released U.S. Department (DOE) DuraMAT DataHub, encourage further this area [3].