President’s Spotlight—Digital & Multimedia Sciences Section

Source: Marcus Rogers, PhD, Digital & Multimedia Sciences Section Board Representative

We have been presenting presidential spotlights from the Academy sections to highlight the latest significant research or improvements within its discipline. Examples include cutting-edge technology, standards, or methods that improve the practice. Alternatively, they may show a historical account/timeline of the discipline’s use of innovative technology or research. These are presented to inform all members of how each discipline is responding to the challenges of a modern forensic science world—our theme for the year. This featured submission is from the Digital & Multimedia Sciences Section.

The Importance of Forensic Video Analysis

One area within the Digital and Multimedia Sciences is Forensic Video Analysis, which is the scientific examination, comparison, and/or evaluation of video in legal matters. Common tasks within Forensic Video Analysis are authentication examinations, video enhancements, comparative analysis, and photogrammetry, which is used to extract dimensional information from images, such as the height of subjects depicted in surveillance video and accident scene reconstruction. As technology changes in the digital and multimedia sciences, so must the approaches to conducting examinations like those used in forensic video analysis.

One novel approach to examining video evidence was presented at the AAFS 2019 Annual Scientific Conference and subsequently published in the Journal of Forensic Sciences.  “Using Reverse Projection Photogrammetry and File Metadata Analysis to Calculate Speed in Recorded Digital Video” took concepts from photogrammetry and coupled them with a new approach to determine elapsed time in the video to accurately determine vehicle speed.  Using reverse projection photogrammetry, where the vehicle involved is placed in the same position on the roadway as found in the recorded video, the location and distance traveled can be measured prior to impact.

To calculate speed, the elapsed time must be determined. Time calculations in recorded CCTV video are challenging as many systems record with a variable frame rate, where the elapsed time between frames may change between any two frames. Using a specific metadata value found in digital video recordings, the time difference between any two frames can be determined with precision, often to 0.0001 seconds.

Just as important as determining speed is the recognition and calculation of uncertainty in both distance and time measurements. Forensic video analysts must also take into account the resolution and other technical attributes of a video as well as the ability to accurately place a vehicle back into a scene using reverse projection when measuring distance. Additionally, the accuracy of the timing metadata is a consideration, even when those values may only have a 0.001-second variance. Here are some photos showing video of the accident scene with vehicles overlaid on the video and speed calculations based on the research.