A comparative study of photogrammetric methods using panoramic photography in a forensic context.

Taking measurements of a scene is an integral aspect of the crime scene documentation process, and accepted limits of accuracy for taking measurements at a crime scene vary throughout the world. In the UK, there is no published accepted limit of accuracy, whereas the United States has an accepted limit of accuracy of 0.25 inch. As part of the International organisation for Standardisation 17020 accreditation competency testing is required for all work conducted at the crime scene. As part of this, all measuring devices need to be calibrated within known tolerances in order to meet the required standard, and measurements will be required to have a clearly defined limit of accuracy. This investigation sought to compare measurement capabilities of two different methods for measuring crime scenes; using a tape measure, and a 360° camera with complimentary photogrammetry software application. Participants measured ten fixed and non-fixed items using both methods and these were compared to control measurements taken using a laser distance measure. Statistical analysis using a Wilcoxon Signed Rank test demonstrated statistically significant differences between the tape, software and control measurements. The majority of the differences were negligible, amounting to millimetre differences. The tape measure was found to be more accurate than the software application, which offered greater precision. Measurement errors were attributed to human error in understanding the operation of the software, suggesting that training be given before using the software to take measurements. Transcription errors were present with the tape measure approach. Measurements taken using the photogrammetry software were more reproducible than the tape measure approach, and offered flexibility with regards to the time and location of the documentation process, unlike manual tape measuring.