Rapid technological advancements in small unmanned aircraft systems (sUAS) have provided a convenient means to capture aerial photography.  Our engineers are FAA certified sUAS pilots; using multiple aerial photographs in combination with traditional land surveying, they can capture and create 2D and 3D photo-realistic, scaled drawings of an accident site, worksite or any outdoor land area.  

What is aerial mapping?

Vertical aerial photographs are photographs taken by an unmanned aerial vehicle (UAV or drone), manned aircraft or satellite that capture a straight down image of the land directly beneath the camera. These images are sometimes called a “bird’s eye view.” Aerial mapping is the process of taking these photographs, combining them with reference measurements, then utilizing photogrammetry software to create large, photorealistic, scaled representations of the mapped area. This is the same technology used to create Google Maps satellite views.  

Aerial Photo Intersection.JPG
Traditional Survey.JPG

Aerial surveying for accident reconstruction.

The goal of an accident site inspection is to observe the accident site, locate and document any physical evidence and capture the geometry of the roadway and surrounding area. Traditionally, measuring the geometry of the roadway requires traditional land surveying equipment to measure hundreds of points in the attempt to outline the road and locate reference objects. This process is very time consuming and results in the minimum amount of data required to create a scaled engineering drawing.


The survey measurements are imported into computer aided design (CAD) software and the drawing is created by “connecting the dots.” This leaves the engineer with the road geometry and any other reference measurements that were taken during the inspection. However, if the engineers miss anything during their inspection, it will not be measured.


Since 2006, accident reconstruction engineers have utilized Google satellite imagery to add some data to their engineering drawings, however the scale of the satellite imagery was not always reliable and had to be verified using inspection measurements. Additionally, the satellites, traveling over 200 miles above the earths surface, can not capture high resolution details and the photographs were often outdated, resulting in low quality imagery that at times did not reflect the current configuration of the roadways.

Satellite w Cars.JPG

With aerial mapping, we can now generate extremely high-resolution aerial drawings to use during our analysis. The drawings are accurately scaled using reference measurements taken during the site inspection and capture everything within the cameras field of view. Therefore, even if our engineers could not see something from their ground inspection, say a tire mark for example, the aerial photographs will have captured it. Aerial mapping provides our engineers with significantly more data to analyze compared to traditional methods.  


The process…

Even with aerial mapping, the process still requires the use of traditional surveying equipment. Before the flight to capture photographs, reference points or ground control points (GCPs) are physically laid out around the perimeter of the area to be mapped. The GCPs can be any shape, but they must be visible to the pilot and in the aerial photographs and indicate the precise location of the survey measurement.

After the GCPs are laid out and measured, the aerial photography flight can commence. Depending on the scale of the land area being mapped, the altitude of the photographs can vary. Because most accident sites are a relatively small area, we typically fly our UAS at a low altitude to capture as much detail as possible. During the flight, vertical aerial photographs are taken in a pattern with some overlap with each neighboring photograph. A typical flight will last between 15 and 30 minutes and yield a couple hundred aerial photographs.

To transform the aerial photographs into a scaled drawing of the accident site, post-processing with photogrammetry software is used. The photogrammetry software identifies common pixel groupings between neighboring photographs. This allows the software to identify the approximate location where each photograph was taken in 3D space. To further refine the photograph locations, the GCP measurements are inputted into the photogrammetry software and located on each photograph. The photogrammetry software then projects the photographs onto a 2D image, creating a composite image, or mosaic, of overlapping photographs. The photographs are skewed such that the final result is an orthographic, or top-down, view of the mapped area. This orthographic mosaic image is called the orthomosaic.

In addition to creating a 2D scaled drawing image, the photogrammetry software can use the locations of the photographs and the locations of the pixels to mathematically triangulate the pixels. This results in millions of photograph pixels suspended in 3D space, called a point cloud. The points in a point cloud are densely packed such that the point cloud is essentially a 3D scale model of the site.