Scientists are planning to train computers to spot anatomical features in footages and images shared online of child sexual abuse. The new study aims to identify perpetrators from images of their hands, the BBC reported.

According to researchers, often the back of the hands are the only visible features in footages and images. This study will reveal how truly unique hands are by looking at anatomical differences between them. Scientists will design algorithms that can help police identify the culprits just by the image of their hands.

The study will be conducted in the universities of Lancaster and Dundee. The researchers from these universities are accepting applications from “citizen scientists” to contribute to their study. Currently, the study is in its preliminary stage where scientists are collecting enough data to go further with the research, according to the BBC report.

Commenting on the study, Forensic anthropologist Professor Dame Sue told BBC that human hands demonstrate many differences due to evolution, the influence of genetics, aging, habits, environment, or even accidents.

She further added that the features including vein patterns, skin creases, freckles, moles, and scars are different even between our left and right hand. They are different between identical twins as well. Currently, the researchers are factoring in all the aspects that make a hand unique. If they succeed in their study, it will be a “step-change” in science.

A web-based app, h-unique.lancaster.ac.uk, is collecting images of people aged 18 and beyond for the project. The images are not shared with any external agencies and will be destroyed at the end of the five-year research project, funded through a 2.5m euros (£2.1m) grant from the European Research Council, as per BBC report.

According to Dr Bryan Williams, cited in the BBC report, lecturer in biometrics and human identification at Lancaster University, the tools that the scientists are planning to develop will be reliable to use in criminal courts. They could also be used to assist law enforcement agencies to rapidly and autonomously analyze hours of footage and thousands of offensive images.

comment COMMENT NOW