Point a camera at a building from two different spots on the street. You now have everything you need to calculate that building's exact height, without ever touching a measuring tape to it. That's the entire premise of photogrammetry, and it sounds almost like a magic trick until you see the geometry behind it.

Photogrammetry is the science of extracting real-world measurements, size, shape, distance, coordinates, purely from photographs. No contact with the object required. It's why drone survey teams can map an entire quarry or a kilometer of coastline without a single person walking the terrain point by point.

📷 The Trick: Two Views, One Point

Here's the geometry underneath it all. If you photograph the same object from two known positions, you can draw an imaginary ray from each camera position through the object's image on the photo, out into real space. Where those two rays cross is the object's real 3D location. This is called intersection, and it's the same principle your own two eyes use every waking moment to judge depth.

🏔️ Far Range vs Close Range

Photogrammetry splits along two practical lines: how far the camera is from the subject, and where the camera is mounted. Far range work, mostly done from aircraft or drones, is what produces topographic maps and digital terrain models. Close range, or terrestrial, work happens from a fixed ground position and is the tool of choice for architects and civil engineers documenting building facades, monitoring structural deformation, or recording heritage sites.

🤝 Photogrammetry's Cousin: Remote Sensing

The two terms get used almost interchangeably, and for good reason. Remote sensing broadly covers gathering information about the earth's surface without contact, which technically includes photogrammetry as one of its founding techniques. The distinction that still matters practically: photogrammetry is quantitative, focused on precise geometric measurement, while photo interpretation is qualitative, focused on identifying and understanding what's actually in the image.

Every coordinate on a modern map once existed only as the crossing point of two imaginary lines drawn through a pair of photographs.

From the Field

On drone survey projects, this exact intersection principle is what makes it possible to generate an accurate 3D model of a site from dozens or hundreds of overlapping photos, without a single ground measurement being taken at most points. The handful of ground control points that do get physically surveyed exist specifically to anchor that photogrammetric model to real-world coordinates, tying the geometry together so the final map isn't just internally consistent, it's actually correct against the ground.

The next time a drone flies overhead capturing a site, remember what's actually happening: not a single measurement is being taken by touch. Every dimension in the final map is being calculated purely from where light rays crossed between two or more photographs, the same principle that's been quietly doing the work of surveying since long before drones existed.