Here’s a revised version with shorter, clearer sentences to address Yoast’s sentence-length warning. I’ve kept the meaning and key details intact.

Researchers at Murdoch University have developed a forensic intelligence tool. The tool could help police link victims of serial offenders by analyzing facial appearance.

Previous studies show that age, sex, class, and physical appearance can influence an offender’s choice of victim. Researchers also suspect that some serial killers target victims who resemble an opposite-sex parent. Others may resemble a family member linked to childhood trauma.

“Serial homicide and sexual offenders are sometimes linked to many more offences than they are formally charged with,” said Associate Professor Brendan Chapman. He serves as Chair in the School of Medical, Molecular and Forensic Sciences at Murdoch University.

“Cases without powerful evidence often remain unsolved or unlinked and go cold. A new method could help investigators link victims through subtle facial geometry similarities. It may provide intelligence that helps investigators focus their efforts.”

Chapman led the research team as they explored Face Similarity Linkage (FSL). The team examined FSL as a potential forensic tool. They wanted to identify possible connections between victims of sexually motivated crimes.

The researchers analyzed photos of three people from a public research image database. They photographed each person from nine different angles. The approach replicated informal images that police often receive. These images can come from social media or personal collections.

The researchers then identified 21 key facial landmarks. These included the corners of the eyes, edges of the lips, chin, and tip of the nose. They measured the distances between these landmarks. They then converted the measurements into ratios.

This method helped reduce distortion caused by photo angles. It also reduced problems caused by differences in image scale.

The researchers conducted several tests to reduce variations between readers and camera angles. They then identified the facial measurements that best distinguished the three unrelated individuals.

The team published the study in The Police Journal: Theory, Practice and Principles. The researchers identified 55 facial measurements that remained stable across different angles.

These measurements allowed researchers to compare faces more reliably. They could identify similarities and differences in facial structures. The method worked even when investigators used imperfect photos.

“With only simple instructions, our untrained testers were still able to keep error rates to around 5%, and we know we can drive that even lower with proper training,” Chapman said.

The tool could also support automated analysis through artificial intelligence.

“We were quite deliberate in developing the technique to be easily adopted by investigators without the need for expensive or proprietary programs,” Chapman said. “However, we’ve opened the door for AI to take it even further.”

An automated system could screen large numbers of victim images quickly. It could also help investigators identify potential links more reliably.

Such a system could provide valuable leads when evidence remains scarce. Chapman said the technique would not replace DNA evidence.

Instead, it could flag potential links between victims when DNA evidence does not exist or has degraded. The technology could therefore give investigators a new starting point in difficult cases.