Open AccessForensic Engineering Application of the Matchmoving Process
Author(s) -
Richard M. Ziernicki,
Angelos G. Leiloglo,
Taylor Spiegelberg,
Kurt Twigg
Publication year - 2018
Publication title -
journal of the national academy of forensic engineers
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.102
H-Index - 1
eISSN - 2379-3252
pISSN - 2379-3244
DOI - 10.51501/jotnafe.v35i2.57
Subject(s) - computer science , computer vision , video camera , process (computing) , artificial intelligence , photogrammetry , camera auto calibration , computer graphics (images) , stereo camera , camera resectioning , videography , orientation (vector space) , drone , geometry , mathematics , operating system , biology , advertising , business , genetics
This paper presents a methodology that uses the photogrammetric process of matchmoving for analyzing objects (vehicles, pedestrians, etc.) visible in video captured by moving cameras. Matchmoving is an established scientific process that is used to calibrate a virtual camera to “match” the movement and optic properties of the real-world camera that captured the video. High-definition 3D laser scanning technology makes it possible to accurately perform the matchmoving process and evaluate the results. Once a virtual camera is accurately calibrated, moving objects visible in the video can be tracked or matched to determine their position, orientation, path, speed, and acceleration. Specific applications of the matchmoving methodology are presented and discussed in this paper and include analysis performed on video footage from a metro bus on-board camera, police officer body-worn camera footage, and race track video footage captured by a drone. In all cases, the matchmoving process yielded highly accurate camera calibrations and allowed forensic investigators to accurately determine and evaluate the dynamics of moving objects depicted in the video.