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

The 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. Our 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, fixed-tilt) and type (rooftop, ground, carport). Specifically, a Faster-RCNN Resnet-50 feature pyramid network (FPN) model was trained and validated on over 770 manually labeled satellite images. This model was used to perform object detection on satellite imagery, locating and classifying individual solar installations' mounting configuration and type. Preliminary model results showed a combined mean average precision score (mAP) score across classes of 49.87% using an Intersection over Union (IoU) threshold of 0.5. We intend to release the complete image data set with labels on the NREL DuraMAT DataHub, to encourage further research in this area. Additionally, a pipeline for automated metadata extraction, including detection of mounting configuration and type as well as azimuth estimation, will be released via the NREL Panel-Segmentation package for public use.