Aerial Imagery API - Roof, Pool and Damage Data

Find out what is on a property without visiting it, from aerial imagery flown at seven and a half centimetres.

What it collects

• Roof facts: material, condition grade on the provider's one to five scale, detected solar panels, roof area, footprint area and building height, for the primary structure and for every other building on the parcel.
• Parcel features: swimming pool with its type, area and whether it has a hot tub, diving board or water slide, plus decks, trampolines, sport courts, playgrounds, enclosures, vehicles, boats, tree cover percentage and hardscape percentage.
• Damage assessment after a catastrophe: condition before and after the event, the catastrophe score, the FEMA damage classification, and per-structure condition where more than one building was assessed.
• Imagery coverage: which orthomosaic and oblique imagery exists over a location, which programme flew it, its first and last capture dates and its ground sample distance range.
• Capture dates: when an area was last flown, so you know how old the derived data actually is.
• Elevation: your own geometry returned with a height on every coordinate, from either the surface model, which includes buildings and trees, or the terrain model, which is bare ground.
• Counts: how many parcels match your filters over an area, so you can size a job before committing to it.

FAQ

What property data can aerial imagery give me?

Roof material, roof condition, solar panel presence, roof and footprint areas and building height for every structure on a parcel, plus pools with type and area, decks, trampolines, sport courts, playgrounds, enclosures, vehicles, boats, tree cover and hardscape percentages. All of it is computed from imagery rather than from public records, so it reflects what is physically there.

How do I find properties with a swimming pool or solar panels?

Run property attributes over a polygon and turn on the pool or roof solar filter. Both are detection filters applied by the provider before the rows come back, so you pay for matches rather than for everything in the area. Run the property count service first to see how many parcels match before collecting them.

Can I filter by roof condition?

Yes. Set a minimum or maximum roof condition and the provider returns only properties in that band. The grade is a one to five scale, and this actor keeps it exactly as the provider publishes it, never rescaled or inverted, because the provider does not state which end is good and inventing a direction would be worse than leaving it alone.

How fresh is the imagery behind the data?

Ask directly. The ortho capture dates service says when an area was last flown, and the coverage services return first and last capture dates with the ground sample distance for each collection. You can also restrict any query to imagery captured between two dates.

Does this actor consume my imagery credits?

No. Every route it calls reads metadata or derived data. Nothing here extracts an image, generates a bundle or orders a product, so a scheduled run cannot spend the credits those cost.

Why did my filter seem to do nothing?

It may not have been sent. This provider ignores a query parameter it does not recognise rather than refusing it, so a filter that a given service does not accept looks exactly like a filter that matched nothing. This actor drops those before the request and names each one in the run log, so an empty run is an explained one.

Are the measurements metric or imperial?

Whichever you choose, and the choice is recorded on every row. The provider takes a unit system per request and does not say in its response which one it used, so an area from this API that did not travel with its unit would not be a measurement at all. Elevation z values are metres regardless, as the provider documents, and the rows say so.

How do I give it a location?

One per line, either a "latitude,longitude" pair or a well known text geometry such as POINT(-87.2095 30.4047) or a POLYGON. Pairs are written latitude first, the way people write coordinates, and converted to the longitude-first order this API expects. The projection is always sent explicitly, because EPSG:4326 and EPSG:3857 put the same numbers in entirely different places.

Do I need my own API key?

Yes. This actor is bring-your-own-key and never ships a key of its own. It takes either an API key, sent in a header rather than in the query string so it does not end up in server logs, or your account username and password, which it exchanges for a session token.

What does it cost?

Pay per result. A property or damage row costs the most, elevation less, imagery coverage a fraction of that because finding out whether an area has been flown should be cheap, and a match count is priced near zero. Platform usage is included. Locations with no coverage are never charged.

Example output

{
  "recordType": "property",
  "requested": "30.404678,-87.209520",
  "found": true,
  "unitSystem": "metric",
  "srid": "4326",
  "structureCount": 1,
  "primaryStructureFootprintArea": 146.33,
  "primaryStructureRoofMaterial": "shingle",
  "primaryStructureRoofCondition": 4,
  "primaryStructureRoofSolar": false,
  "poolDetected": true,
  "poolAreaSqUnits": 48.22,
  "inGroundPool": true,
  "trampolineDetected": false,
  "deckDetected": false,
  "vehicleCount": 1,
  "boatCount": null,
  "treeCoverPercent": 15,
  "parcelSource": "real",
  "latitude": 30.404678,
  "longitude": -87.20952
}

Keyword map

aerial imagery API, property intelligence API, roof condition data API, roof material detection, solar panel detection API, swimming pool detection data, property risk data for insurance, catastrophe damage assessment API, FEMA damage classification data, post-event imagery, building footprint API, building height data, DSM DTM elevation API, tree cover overhang data, underwriting property data, oblique oriented imagery metadata.

Notes

Bring your own key. Read only: no route here extracts imagery or orders a product. Rate limiting, retries and paging are handled for you, and missing or refused credentials end the run cleanly with an explanation rather than failing it. This gateway checks the credential before it looks at the path, so a refusal cannot be told apart from an unknown path; the actor says so rather than asserting your key is wrong when it may be fine.