How 3D aerial mapping works

Creating survey-grade topography maps from a drone still sounds like science fiction to most people, but if you understand triangles, you’re on the right path to understanding the process.

Here is Warren Genik of GreenAero in his own words as he explains how it works.

“It all starts with photogrammetry, a method of combining multiple 2D images into one large 3D map. By overlapping aerial photographs, one individual object or key point shown in multiple aerial photos can be identified using specialized software.

“When one of our drones flies over a field, it captures 500 to 600 photos per quarter section using a grid pattern. Each photo covers about three acres. This means that one rock in the field is photographed eight to 12 times from an altitude of 300 feet.

“Using the same set of aerial images, a second unique object is identified and marked as a new key point. A relationship has now been established between the two points using triangulation. This relationship becomes a spatial measurement or a point in 3D space.

“The high number of overlapped photos creates a very detailed point cloud, or 3D model of the field. With the drone travelling 60 km-h, it takes an average of 45 minutes to capture the necessary imagery for an entire quarter.

“By repeating the process many more times using high-end computer servers, this eventually becomes a ‘point cloud’ with millions of 3D points, each with its own unique location and elevation measurement.

“The key to making this model accurate enough for use in precision agriculture, engineering or survey activities comes down to three main points: right time, right place and right image.”

About the author

Ron Lyseng's recent articles


Stories from our other publications