Solar panels

Other objects On request

Detected rooftop solar panels

Example result: rooftop and ground-mounted panels.

Cost

Priced per project

Request access and we connect the model to your account.

A detection and segmentation model for photovoltaic panels, trained on aerial imagery. It finds panels on rooftops and in solar farms.

What you get

Solar panel polygons.

Works best with

Aerial imagery at zoom 19–21.

Usage examples

  • Inventory rooftop solar in a municipality

  • Monitor solar farm build-out

  • Estimate installed panel area for grid planning

Geo domain

Global

Method

Detector + segmentation

GSD / zoom

z19–21

Test score

Pixel-wise F1 > 0.77

How it works

The model combines detection and segmentation to find photovoltaic panels on rooftops and in solar farms and outline them as polygons. It is trained on aerial imagery and needs high-resolution aerial imagery at zoom 19–21 for accurate detection. The test score is a pixel-wise F1 above 0.77.

Use it to monitor solar energy installations and for urban planning.

Need a different object?

We train custom models on your imagery and classes, and connect them to your Mapflow account.