Quick answer
Candidate roof outlines or segments identified from georeferenced imagery and supported elevation data, requiring source and geometry review before design use.
Answer: Map-based roof detection identifies candidate roof boundaries or segments from georeferenced imagery and, where supported, additional elevation data. It helps start a remote layout, but an outline is not proof of roof pitch, current conditions or usable installation area. Review the intended building, source date, spatial alignment and detected features before treating the result as a design input.
Detection is different from measurement and approval
A detector may identify a building outline, individual roof faces or visible objects. Those outputs depend on the imagery, model and available supporting data. A two-dimensional overhead image alone does not uniquely establish three-dimensional pitch, height or concealed geometry. A building footprint is also not necessarily the roof edge: overhangs and image displacement can produce differences.
Google’s Solar API Building Insights reference illustrates a provider-specific data product with roof-segment information and separate area/ground-footprint fields. Such documented fields do not mean that every map image contains those measurements or that SurgePV uses this API. Verify the actual data product and software functionality.
A LiDAR roof model derives geometry from laser-survey data. Map-based detection may instead use image segmentation or another reconstruction method. Preserve the distinction when evaluating the output and its quality evidence.
A practical detection review
| Step | Question to resolve |
|---|---|
| Select the property | Is this the intended building, not a neighbor or another structure on the parcel? |
| Inspect the source | What was captured, when, under what licensing and processing conditions? |
| Check alignment | Are projection, coordinate references and units consistent? |
| Review outlines | Are eaves, attached structures and obscured edges correctly represented? |
| Review segments | What evidence supports face boundaries, pitch and orientation? |
| Review obstructions | What visible or site-confirmed features are missing or uncertain? |
| Release the input | What has been checked and what still requires confirmation? |
Trees, shadows, reflections, roof complexity and viewing angle can obscure features. Another permitted image, elevation source or site evidence may resolve a gap; no source can be assumed to reveal everything beneath a canopy. Higher pixel resolution does not by itself guarantee better positional accuracy or complete obstruction detection.
Google’s Building Insights request guide documents imagery-quality availability and possible NOT_FOUND errors. It also notes that building results may differ from geocoding results. These are specific service behaviors, not a promise of global coverage or an integration claim.
Calculate usable area without double counting
Detected gross area and usable area must use the same surface/footprint basis. Exclusions can overlap, so subtract their geometric union rather than add every exclusion label independently.
For a hypothetical planar face, gross area is 100 m². Access/setback exclusions cover 20 m² and obstruction exclusions cover 10 m², with 4 m² shared between them. The excluded union is 20 + 10 − 4 = 26 m²; the remaining area is 100 − 26 = 74 m².
Subtracting both areas without their overlap would incorrectly give 70 m². This arithmetic is not a compliant layout: actual access/setback requirements follow the applicable project rules, and panel dimensions, spacing and geometry further constrain placement. Shade is an energy-model consideration and may inform selected keepouts; it is not automatically a binary area subtraction everywhere.
Keep source dates and uncertainty visible
Check whether roof replacements, new skylights, equipment or vegetation changes occurred after capture. A fixed imagery-age threshold cannot prove current conditions. Record source acquisition dates separately from the date someone reviewed the model.
Use the actual spatial and quality documentation when combining imagery with LiDAR data integration. Do not merge data with inconsistent references through an undocumented visual adjustment, or describe an inferred face as directly surveyed.
Compare remote geometry with appropriate current site or plan evidence for the intended decision. Roof condition, framing, electrical suitability and jurisdictional approval require separate evidence and qualified review. Detection does not establish compliant clearances or eliminate necessary site verification.
Team handoff
Designers should inspect and correct candidate boundaries with documented reasons and sources. Site teams should report missed features or changed conditions before affected procurement or construction decisions. Sales teams should label preliminary layouts and explain what may change, without promising a percentage accuracy, proposal time or conversion gain.
Confirm software functions, supported providers and data-use rights before offering automation. Keep the accepted detection/model version and unresolved-item list so subsequent reviewers can trace what was actually checked.
Frequently asked questions
Can an overhead image alone establish roof pitch?
Not uniquely. Pitch needs appropriate additional measurement, elevation or reconstruction evidence.
Does better pixel resolution guarantee accurate dimensions?
No. Positioning, viewing geometry, source quality and processing also affect dimensions and boundary reliability.
Are detected roof boundaries automatically compliant setbacks?
No. Apply the applicable project requirements and verify the layout separately.
Can map detection replace all site verification?
No. Determine which geometry and non-geometric conditions still need appropriate current evidence and qualified review.
Related Glossary Terms
About the Contributors
CEO & Co-Founder · SurgePV
Keyur Rakholiya is identified by SurgePV as its CEO and a company co-founder. His SurgePV author page lists only role information that can be tied to the public profile below; credentials, project totals, testing claims, media appearances, and speaking engagements are not asserted without retained evidence.
Editorial contributor · SurgePV
Rainer Neumann is credited as an editorial contributor on SurgePV content. This profile does not assert engineering credentials, project totals, software-testing experience, education, speaking engagements, or media citations because independent verification evidence is not retained in the publication record.
