Facial recognition and laser surface scan: a pilot study

Area of Science:

• Forensic Science
• Biometrics
• Computer Vision

Background:

• Facial identification from CCTV footage is crucial in forensic investigations.
• Traditional 2D facial recognition methods face limitations with varying angles and lighting.
• 3D facial surface scanning presents a potential advancement for more robust identification.

Purpose of the Study:

• To evaluate the efficacy of 3D facial surface scanning for forensic identification.
• To assess the accuracy of matching 3D facial scans to 2D photographs in blind trials.
• To determine the reliability of surface scanning for precise cranial measurements.

Main Methods:

• Fifteen male volunteers underwent facial surface scanning using a Polhemus FastSCAN Cobra Handheld Laser Scanner.
• Three 2D photographs (frontal, profile, 45-degree angles) were taken of each volunteer.
• Specialized software (MIMICS, Photoshop) was used to match 3D scans with 2D photos in blind trials, with matches graded as 'good fit,' 'possible fit,' or 'no fit.'

Main Results:

• All surface scans and photos were correctly matched, achieving a discriminatory value of 86.7% when one scan matched two angled photos.
• The surface scanner demonstrated high reliability for point measurements on skulls, with an average variation of 1 mm compared to caliper measurements for five cranial measures.

Conclusions:

• 3D facial surface scanning shows significant promise as a tool for forensic facial identification.
• The technology offers a reliable method for capturing detailed facial morphology, improving accuracy over 2D methods.
• Further research can explore broader applications in forensic science and biometrics.