Not long ago on a Friday evening, it took me nearly three hours to drive during rush hour traffic from downtown Seattle to my home. The problem was not just heavy traffic, but an accident pileup that had closed two out of four southbound traffic lanes.

There were ambulances, firetrucks and police cars scattered everywhere. Part of the effort was incident response and part was one of accident documentation. It took five hours to reopen every lane.

Traffic accidents and lane closures are a fact of life for most of us, but the good news for law enforcement professionals and others who are called upon to document these incidents is the availability of geospatial forensics technology that can expedite documentation and reduce traffic lane closure times by as much as 80 percent. “We’re seeing a significant increase in interest in geospatial forensics technology in many areas of the world,” says Lisa Chen, technical communication manager at Pix4D, a Swiss firm that provides drone mapping software.

By using drones equipped with geospatial technology capabilities, law enforcement agencies and others can fly over an accident site and document it in a fraction of the time that it would take to document the same scene manually, Chen says. “North America is currently the largest drone market, with many law enforcement units using drones equipped with geospatial technology. The U.K. is also aggressively employing it. In other countries, drones with geospatial forensics capabilities are still in a very early stage of consideration, but we are starting to see some countries, like Brazil, begin to make serious investigations of it.”

Drone-Based Business Case

The drone geospatial forensics business case is compelling. Chen explains that before drones and other advanced geospatial technology, law enforcement had to mark key areas of an accident with traffic cones. The distance between markers was measured manually to establish the boundaries of the scene so they could be recorded. “Then, personnel would manually take pictures at the site. In some cases, they would use laser scanners, but this process is relatively slow and takes a great deal of time. Now, with help of flyover drones that are equipped with cameras, law enforcement is able to record evidence and boundaries of sites with the drone, and the accident documentation process is much faster. You no longer have to shut down a lane of traffic for hours at a time.”

The use of drone-based geospatial methods also opens up a case for the continued use of photogrammetry, which originated in the 19th century and is still the most accurate way of taking photos and of making measurements from photographs to establish the 3D positions of surface points.

Geospatial Forensics in a Court of Law

“Laser scanning often times aids forensics investigations by providing a great deal of photogrammetric control at the scene in a very detailed manner,” says Christopher Curtis, president of CBC Geospatial, a geospatial forensic consultancy. “It is an established science that is widely recognized and accepted in the courts.”

Court acceptance of geospatial forensics evidence is paramount in both civil and criminal cases. Curtis has seen it used in cases ranging from wrongful death and personal injury to wildfires and fraud.

One such case in New York State several years ago had the plaintiff suing the city for negligent maintenance of a sidewalk area that contained a crack that caused the plaintiff to slip, fall and injure herself. The plaintiff’s attorney used photogrammetry, which enabled the attorney to show the precise width and grade of the crack.

“In contrast, the law firm for the defendant hired an engineer who used an Internet-based mapping function to map the street and to measure the size of the crack,” says Darryl Murdock, vice president of professional development for the U.S. Geospatial Intelligence Foundation (USGIF). “But the problem with this approach was that the map had very little information that accompanied it. It had the date that the photos were taken, but not the time of day, the point above the grade of the sidewalk that the photo was taken from or the orientation of the camera. Without this information, it is hard to make an exact determination about the nature or the size of the crack.”

Technologies that are Making a Difference

So besides drones and photogrammetry in geospatial forensics, what other technologies are making a difference for law enforcement, insurance investigators and others?

  • Terrestrial laser scanning: “Terrestrial laser scanning is being used on the ground because it can easily capture and record snapshot details of scenes,” Curtis says.
  • Point cloud software: Geospatial forensics makes use of point clouds, which are collections of data points that are defined by a given coordinate system, so they can represent the shapes of real or simulated physical systems. “The high-resolution 3D point cloud enables people to take precise measurements, wherever and whenever they want,” Chen says. “Even for projects without initial geolocation, you can simply assign a known linear distance and the software will recalculate and refine the project scale.”
  • Integrated technologies: Organizations are also learning how to “mix and match” different types of geospatial technologies to solve particular forensics documentation problems. For instance, let’s say that the distance between two objects is too small to use a handheld laser scanner. In this case, an overhead drone or doing the measurement manually might be the best approach.
  • Process automation: Taking and then managing the often voluminous numbers of photos and images of accident and crime scenes can become a challenge. To address this challenge, photogrammetry can retrieve the 3D information from the overlap images. Based on the image content, the software automatically reconstructs the scene in 3D without any user intervention.
    “When images of accident scenes are taken by drones or by laser scanners, both can be easily imported into a point cloud,” Chen explains. “There are areas of overlap that invariably occur when you take many different photos of a scene. To manage this, the point cloud software has built-in algorithms that automatically detect the areas of image overlap and resolve them. This gives the user a continuous and uninterrupted rendering of an accident or crime scene.”
  • Analytics: Analytics is still a nascent field for most geospatial forensics packages, but vendors are working on forming alliances with analytics vendors that will soon make it possible for users to perform more advanced queries of their forensics documentation. “This is a very active area for us,” Chen says. “With strong image processing power, which provides 3D outputs automatically, we are open to implementing more analytics tools, from internal development or external collaborations, for the targeted industry.”

Putting the Technology to Work

As geospatial forensics moves forward, the key for those hoping to utilize it more will be ease of use.

“Ease of use will continue to be a major factor,” Curtis says. “In many cases, firms need to use 3D animation software that is similar to what they use in Hollywood productions for film animation. The difference is, in Hollywood they have contingents of people who might just be working on specific movements of an animated character. The ‘shop’ of most forensic animators is typically much more limited in resources, but must still know how to use the technology.”