Murdoch University Launches Facial Link Tool
Get the Health newsletter
Daily health & science — research, biotech, public health, the studies worth knowing. Free.
- Murdoch University researchers developed a forensic intelligence tool called Face Similarity Linkage (FSL) to connect victims of sexually motivated serial homicide.
- Associate Professor Brendan Chapman explained that the method isolates 21 facial landmarks and identified 55 stable measurements across angles to compare faces despite imperfect photos.
- The study measured distances between landmarks, converting them to ratios that cancel out angle and scale distortion, and found that untrained testers could keep error rates near 5%.
- Untrained testers achieved the 5% error rate using only simple instructions, and the researchers noted that proper training could reduce errors further.
- The researchers suggest that artificial intelligence could automate the technique, screening large image sets quickly to generate investigative leads where DNA evidence is absent.
- The tool is designed to be low‑cost and adoptable without proprietary software, complementing DNA evidence rather than replacing it.
Why it matters: Police investigators gain a cheap, image‑based way to generate leads in cold serial homicide cases lacking DNA or other forensic evidence, expanding their investigative toolkit. Meanwhile, perpetrators remain harder to catch without this additional intelligence, but the method offers a practical supplement to existing evidence.
Ask SkimNews
