arXiv cs.AI· Guibin Chen, Emma Fan, Yuzhang Lin·· 4 小时前AI 评分22
Direct Head:从航拍图像分割引导直接估计屋顶光伏装机容量
Segmentation-Guided Direct Capacity Estimation of Rooftop Photovoltaics from Aerial Imagery
AI 导读
研究提出 Direct Head,一种分割引导架构,保留 U-Net 分割与软掩码池化,但直接从光伏区域特征和面积描述符回归非负容量,不再依赖掩码面积乘功率密度。在 7,590 个法国屋顶光伏装置上,相对参数匹配的面积密度神经模型,平均绝对误差降低 8.7% 和 12.5%,对称平均绝对百分比误差降低 14.0% 和 20.7%。
正文 · 原文
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.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.22076 [cs.CV] |
| (or arXiv:2611.22076v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2611.22076 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Guibin Chen Mr [view email]
[v1]
Tue, 17 Nov 2026 01:56:30 UTC (3,773 KB)
来源:arXiv cs.AI · arxiv.org