Detection API

Aerial wildfire detection in a single API call. POST an image — get smoke and fire detections back in milliseconds. RGB today; thermal confirmation next.

The image API takes one still frame per request; live video runs on our edge device (see Live feeds & cameras). Overlay values are illustrative.

Quickstart

Your first detection in three steps.

1. Get a key. Request access — we issue your X-API-Key.
2. POST an image to /v1/detect:

curl -X POST "https://api.aerialiq.app/v1/detect?mode=wildfire" \
  -H "X-API-Key: your_api_key" \
  -F "file=@frame.jpg;type=image/jpeg"

3. Read the JSON. Each detection carries a class_name, a confidence, and a pixel bbox — draw the boxes and you're done.

{ "detections": [ { "class_name": "smoke", "confidence": 0.94,
                   "bbox": { "x1": 812.4, "y1": 233.0, "x2": 1106.7, "y2": 498.2 } } ],
  "model_version": "wildfire-v1", "inference_ms": 38.6 }

Base URL

https://api.aerialiq.app/v1

All endpoints are HTTPS.

Authentication

Access is provisioned per partner organisation. You'll receive an API key — send it on every request as the X-API-Key header. Self-serve keys and usage dashboards are on the roadmap; contact us to request access.

X-API-Key: your_api_key

Detect

Run wildfire detection on a single image. Synchronous — the response carries the detections directly.

Parameters

Image input

Request

curl -X POST "https://api.aerialiq.app/v1/detect?mode=wildfire" \
  -H "X-API-Key: your_api_key" \
  -F "file=@frame.jpg;type=image/jpeg"

import requests

with open("frame.jpg", "rb") as f:
    resp = requests.post(
        "https://api.aerialiq.app/v1/detect",
        params={"mode": "wildfire"},
        headers={"X-API-Key": "your_api_key"},
        files={"file": ("frame.jpg", f, "image/jpeg")},
        timeout=30,
    )
resp.raise_for_status()
data = resp.json()
for d in data["detections"]:
    print(d["class_name"], round(d["confidence"], 2), d["bbox"])

Response 200 OK

{
  "mode": "wildfire",
  "model_version": "wildfire-v1",
  "image_width": 1920,
  "image_height": 1080,
  "detections": [
    {
      "class_id": 0,
      "class_name": "smoke",
      "confidence": 0.94,
      "bbox": { "x1": 812.4, "y1": 233.0, "x2": 1106.7, "y2": 498.2 }
    }
  ],
  "detection_count": 1,
  "inference_ms": 38.6,
  "spray_savings": null
}

Fields

Detection object

Modes

Errors

Standard HTTP status codes. Error bodies are {"detail": "…"}.

Data handling

Images are processed in-memory and not retained. No uploaded image is written to disk or stored. Only detection metadata (counts, model version, timing) may appear in operational logs. It's a property of how the endpoint works, not a policy promise.

Limits & versioning

Live feeds & cameras

The image API is for snapshots — one frame per request. To run detection on a live drone or camera feed, you have two paths that detect every frame:

Got drones or cameras? Talk to us about edge and stream ingestion — the public image API detects one frame at a time.

Changelog

© AerialIQ · Detection API · info@edgegenix.com