Segmentation-Guided Direct Capacity Estimation of Rooftop Photovoltaics from Aerial Imagery
Organizations: Department of Electrical and Computer Engineering, New York University, New York, USA
Abstract
Distribution planners need installation-level inventories of rooftop photovoltaic (PV) capacity, but interconnection records may be incomplete or spatially aggregated. Conventional image-based estimators segment PV arrays, convert the mask to physical area, and multiply by an estimated power density, coupling capacity prediction to mask-derived area. We propose Direct Head, a segmentation-guided architecture that retains U-Net segmentation and soft-mask pooling while directly regressing nonnegative capacity from PV-region features and area descriptors. We evaluate three independent runs on 7,590 French rooftop installations under the original split and a department-based holdout. Relative to a parameter-matched area-density neural model, Direct Head reduces mean absolute error by 8.7% and 12.5%, and symmetric mean absolute percentage error by 14.0% and 20.7%, respectively. Comparisons with conventional estimators, controlled architecture variants, auxiliary-loss ablations, and representative cases show that segmentation provides useful spatial evidence without requiring capacity to be determined by an area-density product.
Figures & tables
| Protocol | Method | MAE ( ) | sMAPE (%) | IoU |
|---|---|---|---|---|
| Original | Area constant density | |||
| Area linear regression | ||||
| Area + color random forest | ||||
| Area-density neural model | ||||
| Direct Head (proposed) | ||||
| Administrative | Area constant density |
| Original | Administrative | |||||
|---|---|---|---|---|---|---|
| Variant | MAE | sMAPE | IoU | MAE | sMAPE | IoU |
| Area-density neural model | ||||||
| Global-pooling (GP) | ||||||
| No-capacity-feedback (NCF) | ||||||
| Direct Head | ||||||
| Configuration | MAE | sMAPE | IoU |
|---|---|---|---|
| Full objective | 16.854 | ||
| Without area loss | 1.615 | 0.728 | |
| Without segmentation loss | 1.736 | 18.594 | 0.085 |