Number-ology
In 1902, Harry Jackson became the first British criminal to be successfully linked to a crime - stealing billiard balls - by his fingerprint. Fingerprints are unique to individuals and have been an invaluable aid to crime-fighters the world over. Francis Galton played a key role in establishing a scientific basis for fingerprinting.
More than 100 years on, fingerprints remain a mainstay of individual identification. But other approaches are gradually being introduced. Iris patterns are also unique and iris-recognition technology is increasingly being used, at UK airports for example.
Our DNA is also a unique identifier, and the National DNA Database provides a huge store of records of convicted criminals and suspects. It has been used to identify the source of biological material found at crime scenes.
There are hopes that a better understanding of the links between genes and phenotype could improve the predictive power of the database - so a DNA sample could reveal key information about a suspect, such as hair colour, eye colour, height or ethnic group. At the moment, this predictive power is very limited. Information stored in the database is derived from non-coding regions of the genome, so provides no biological information.
Meanwhile, automated face recognition is coming closer to reality. Human brains are very good at picking out individuals, even if they have aged, added or removed facial hair, or are seen in different lights, with different expressions and from different angles. While digital cameras can now pick out a face in a scene, even the most powerful computer would struggle to put a name to the same face seen in different scenes.
It may happen soon, though. A team at Glasgow has found that averaging a set of photos of someone significantly improved a computer system's face recognition ability. The system includes a database of celebrity faces. Although it struggled to recognise a celebrity when presented with a single photo, when 20 photos of a celebrity were merged it was able to identify them every time.
The principal drive for biometric identification comes from countries' national security fears. Being able to identify individuals instantly provides a way to control entry of people into (or around) countries. It has other obvious uses in crime-fighting.
On the other hand, not everyone is keen on being so closely monitored by the state. Once authorities know who we are, they potentially have access to a wide range of information about us. This immediately raises issues of privacy and access to information.
These new technologies are all based on the conversion of human appearance into a unique digital representation. Once this is done, the speed and power of computers can be applied to share, analyse and manipulate data. A computer can do little with a photo, say, but once it has been converted into a digital form that maintains a unique link to the photo's subject, a whole new world opens up.
Image: Human eye; Kate Whitly, Wellcome Images