Orthomosaics you can measure on
Hundreds of overlapping photos become one scale-correct image of your site, plus the elevation model and 3D mesh underneath it. Here is how it works and what to ask for.
From overlapping photos to a measurable map
Photogrammetry finds the same feature in many photos taken from different positions and works out where it sits in 3D. Do that for millions of features and you get a dense point cloud. Build a surface from the points and you get a digital surface model. Project every photo onto that surface, correcting for tilt and relief, and you get an orthomosaic: an image where one pixel is the same size everywhere and distances can be measured directly.
That last point is the difference between an orthomosaic and an ordinary aerial photo. An oblique photo, or even a straight-down photo of a sloped site, distorts scale across the frame. A proper orthomosaic does not, which is why it can go under a site plan in AutoCAD or into ArcGIS or QGIS as a base layer. How close it sits to true position is a separate question, covered under accuracy.
The same flight also yields contours, a bare-earth terrain model after filtering, and a textured 3D mesh. Adding angled (oblique) passes around buildings improves walls, overhangs and facades for the mesh. Where the ground is hidden under trees, LiDAR is the better tool; for a site you will fly again and again, see construction progress.
GSD, overlap and the other flight choices
| Setting | What it means | Why it matters |
|---|---|---|
| Ground sample distance (GSD) | The ground size of one image pixel, for example 1.5 cm. It equals sensor width times flight height, divided by focal length times image width in pixels. | Sets the smallest thing you can see and caps achievable accuracy. Halving GSD roughly quadruples the photos and processing. |
| Front overlap | How much each photo shares with the next along the flight line. Mapping practice is commonly around 75 to 80 per cent. | Too little and the software cannot match photos; gaps and warping follow. |
| Side overlap | How much neighbouring flight lines share, commonly around 60 to 70 per cent, more for trees and uniform surfaces. | Low side overlap shows up as holes and noisy elevations between lines. |
| Flight height | Height above the ground, capped at 122 m (400 ft) under standard Canadian and US rules. | Drives GSD and coverage. Terrain-following keeps GSD even on slopes. |
| Shutter and light | Mechanical shutter, fast enough speeds, consistent sun. | Blur, rolling-shutter distortion and moving shadows all degrade matching. |
| Oblique passes | Tilted-camera orbits or grids around structures. | Needed for good walls and facades in a mesh; not needed for a flat ortho. |
Common orthomosaic problems
Wavy lines and melted buildings
Straight edges ripple and roof lines lean where the surface model is poor, usually from low overlap, trees or reflective roofs.
Holes over water, glass and snow
Photogrammetry needs texture. Calm water, glass curtain walls, fresh snow and uniform membranes give it nothing to match.
Ground elevations under trees
A camera sees the canopy, not the ground. The surface model follows the tree tops.
A good-looking map in the wrong place
Without control, the map can be shifted or scaled slightly. It looks perfect until it is overlaid on a legal plan.
Moving objects
Excavators, trucks and cranes move between photos and leave ghosts or smears.
What to prepare
Define the boundary
A KMZ, a PDF site plan or a pin and a description. Include any neighbouring land you need for context, such as street frontage or adjacent buildings for shadow studies.
State the purpose
Marketing overview, design base, volume work and progress record each call for different resolution and control.
Name the coordinate system
Ask your surveyor or engineer which projection, horizontal datum and vertical datum the project uses. We match them.
Share hazards and contacts
Cranes, power lines, blasting, crews, gates and the name of the person on site.
Pick a window
Dry, light wind, even light. Midday limits long shadows; bright overcast is often best for texture.
Who uses the orthomosaic, and what they need from it
Architect
- They need from you
- An accurate base with neighbouring structures and trees for siting and views.
- You need from them
- The CAD units, base drawing and any existing survey to align to.
- On the DroneLink record
- Ortho, contours and mesh delivered together by link, each labelled with its datum and date.
Realtor or marketing team
- They need from you
- A clean overhead for lot lines and context graphics.
- You need from them
- Approval of which frames and annotations to use.
- On the DroneLink record
- Review and approval on the platform, with comments pinned to the image. See real estate photography.
Municipality
- They need from you
- Current site conditions for a referral or permit discussion.
- You need from them
- Their accepted formats and whether a surveyor's plan is required instead.
- On the DroneLink record
- A dated PDF plus the GeoTIFF, shared from the one record.
What comes back
- Orthomosaic as GeoTIFF, with a lighter JPG or KMZ for Google Earth and email.
- Digital surface model and, where filtered, a bare-earth terrain model as GeoTIFF.
- Contours at an agreed interval as DXF, DWG or SHP.
- Photogrammetric point cloud as LAS or LAZ.
- Textured 3D mesh as OBJ for visualization and massing.
- A short processing report: date, GSD, control used, coordinate system and checkpoint results where checkpoints were surveyed.
Questions
What is an orthomosaic?
An orthomosaic is a map-accurate aerial image stitched from many overlapping photos and corrected for camera tilt and terrain. Scale is consistent across the image, so you can measure distances and areas on it and overlay it on CAD or GIS drawings.
What is the difference between an orthomosaic and an aerial photo?
An aerial photo has perspective distortion: scale changes across the frame and tall objects lean. An orthomosaic is corrected so each pixel represents the same ground size, which makes it usable as a measured base map.
What resolution do I need for a drone map?
For general site context, a few centimetres per pixel is usually plenty. Design work, volume calculations and fine detail call for finer ground sample distance and more careful control. Tell us the purpose and we recommend a GSD.
Can drone photogrammetry see through trees?
No. A camera maps what it sees, so under canopy the surface follows the tree tops. For ground under vegetation, LiDAR with multiple returns is usually the better tool.
What file format is an orthomosaic delivered in?
Usually GeoTIFF, which carries its own coordinates and opens in ArcGIS, QGIS, AutoCAD Civil 3D and many design tools. We can add a KMZ for Google Earth or a JPG with a world file.
How long does it take to process a drone map?
Processing time depends on the number of photos, resolution and products. Small sites are quick; large high-resolution jobs take longer, and quality checking adds time. We give an expected delivery date with the estimate, not a guarantee.
Can I use a drone orthomosaic for a building permit?
Sometimes as supporting context, but many municipalities require a plan by a licensed land surveyor for boundaries and grades. Check with your municipality, and involve your surveyor early.
Tell us about your Miami site.
An address and what you need is enough to start. We reply with what it will take, who will fly it and when.