New forensic tool provides intelligence to link serial killer victims by analyzing facial similarities
Why it matters: This new tool could help investigators link unsolved sexually motivated serial homicides, potentially solving cold cases that lack traditional evidence.
- Murdoch University researchers, led by Associate Professor Brendan Chapman, developed a forensic intelligence tool called Face Similarity Linkage (FSL) to identify potential connections between victims of sexually motivated crime.
- The FSL tool analyzes 21 key facial landmarks, such as eye corners and nose tips, and converts distances between them into ratios, allowing for reliable comparisons despite variations in photo angle or scale.
- Previous studies suggest that serial killers often choose victims based on features like age, sex, class, and physical appearance, sometimes even seeking resemblances to a parent or family member associated with childhood trauma.
- The research team identified 55 stable facial measurements that remained consistent across different angles, enabling more reliable comparisons of facial structure, even with imperfect images.
- Untrained testers using the tool achieved error rates of around 5%, which researchers believe can be further reduced with proper training, and the tool has the potential for automation with artificial intelligence.
Researchers at Murdoch University have developed a new forensic intelligence tool, Face Similarity Linkage (FSL), that can help police link victims of serial offenders by analyzing subtle facial similarities, even from imperfect photos. This tool measures 21 key facial landmarks and converts them into ratios to overcome distortions from photo angles, potentially solving cold cases where traditional evidence like DNA or fingerprints is absent.
