What Drone Photos Say About You: The Metadata DJI Writes Into Every File

Most privacy advice about photos assumes a phone or a camera. Drones are worse. A photo taken from a consumer drone carries not just where the photo was taken, but where the drone took off from, how fast it was flying, which direction it pointed, how high it climbed, and a serial number tied to a product registration. Anyone who receives the original file can reconstruct the entire flight segment around the shutter click, often down to the street you launched from. If you share drone footage of your house, your land, or your business, you are sharing more than the view.

The two metadata layers in a drone JPEG

DJI and the other consumer manufacturers write two separate blocks into every image. The first is ordinary EXIF, the same structure your phone produces: capture time, exposure settings, and GPS coordinates in the standard GPS tag set. The second is an XMP block, and this is where drones diverge from phones. XMP is free-form XML, so manufacturers stuff it with flight telemetry. Typical fields in a DJI still image include FlightYawDegree (which way the aircraft nose pointed), GimbalYawDegree (which way the camera looked), FlightRollDegree and FlightPitchDegree, absolute RelativeAltitude, and SpeedX, SpeedY, SpeedZ in metres per second.

Read together, those fields tell a story a plain GPS pin does not. A hovering drone at 40 metres points straight down at one building. A moving drone at 15 m/s sweeping across a boundary tells a different one, and that difference matters if the photo ever surfaces in a dispute about where exactly the aircraft was. The XMP also usually carries AircraftName, firmware versions, and a serial-style identifier for the camera and the airframe.

The fields that follow you home

The GPS coordinates in a drone photo are precise, but the launch location is the field people forget. Many DJI models log the home point recorded at takeoff alongside per-shot coordinates, and even where they do not, a sequence of shots from one flight traces a path that ends where it started. Fifteen photos of a sunset over your property can contain enough positions to mark the field you launched from, the road you drove in on, and roughly where the car was parked.

Serial numbers raise the same tracking problem we covered with phone and camera body serials, with one addition: drones are registered. In many jurisdictions a consumer drone above a weight threshold must be tied to an operator ID, and the airframe serial in a leaked photo can be cross-referenced against marketplace listings, warranty registrations and registration databases. A hobby photo posted with its metadata intact can connect an anonymous account to a named purchase.

Timestamps deserve a note too. Drone clocks drift like any camera, and many owners never sync them, but the GPS subsystem usually stamps UTC accurately. That means one file can carry two clocks that disagree, which occasionally matters in the other direction: a drone photo can be useful evidence precisely because its telemetry is hard to fake casually.

What survives sharing

Everything we described in what messaging apps do with photo metadata applies to drone files, with a caveat in the wrong direction. The XMP telemetry block is larger and more exotic than phone EXIF, and a lot of casual stripping tools only handle standard EXIF tags. Tools that advertise GPS removal sometimes leave the DJI XMP fields untouched, including coordinates duplicated inside the XMP block itself. Screenshots and platform re-encoding usually destroy the whole block, but a file sent directly, by email attachment, USB stick, or chat apps that preserve files, arrives with the entire flight record inside.

How to check and clean a drone image

• Drag the file onto the reader on our homepage before you share anything you would not want traced. It runs locally and shows both the EXIF and the XMP blocks, so the flight fields are visible in plain text, not hidden behind a summary.
• After stripping, re-check. Strip, then re-read the file, because a tool that reports success while leaving XMP telemetry behind is common enough to be the rule rather than the exception.
• Treat a batch as one story. Removing GPS from each image while leaving altitude, speeds and timestamps intact still lets a reader reconstruct the flight path. Strip the whole telemetry block or re-encode before publishing.
• If you publish drone content publicly, decide once: either every file you post is cleaned, or you accept that your launch sites are public. Doing it per photo guarantees the one you forget is the one that matters.
• Remember the pixels too. A wide aerial shot identifies a property even with the metadata gone, and one of the many lessons from our piece on what home photos reveal applies from the air at greater range.

The short version: a drone photo is a flight log wearing the costume of a picture. Fly however you like, but open the file before you share it.