Solar panels (Neural Network)

The Solar panels Neural Network is a Convolution Neural Network that identifies solar panels, mainly situated on roofs of buildings.

An Inference Overlay can be configured with this Neural Network. Its default settings are:

Preferred grid cell size: 0.1m

Inference mode: BBox Detection

Mask threshold: 0.9

Score threshold: 0.2

Stride fraction: 0.50 (50%)

Identifiable features:

1. Solar panel blue (oldest type)
2. Solar panel gray (mid-aged type)
3. Solar panel black (newest type)

• Solar panel inference labels.jpg

  Labeled features on 0.1m satellite image

• Solar panel inference scores.jpg

  Label Scores

• Solar panel inference masks.jpg

  Pixel Masks

• Solar panel inference boxes.jpg

  Bounding Boxes

How-to's

• How to detect solar panels using an Inference Overlay
• How to update Buildings's solar panel attribute based on an Inference Overlay

See also

• Inference Overlay
• Neural Network
• Solar panels (Function Value)
• Solar panels (Indicator)

---

Inference Overlay

Inference Overlay

Tutorials

Overview

Theory

Benchmarks

Group

Model attributes

Inference

Neural Network • Input tensor • Output tensor

Model attributes

Box overlap • Inference mode • Mask threshold • Prioritize labels • Score threshold • Stride fraction

Prequels

A • B

Result types

Boxes • Labels • Masks • Scores • Debug gt instances • Center points

Neural Networks

Foliage • Solar panels