Skip to Content

Quantum Biometrics Exploits the Human Eye’s Ability to Detect Single Photons

Identifying individuals by the way their eyes detect photons could be a hugely accurate form of biometrics, guaranteed by the laws of quantum mechanics.

When it comes to security, the quantum world offers unparalleled riches. Quantum cryptography, for example, promises absolute secrecy guaranteed by the laws of physics.

That’s why governments, military organizations, and others have rushed to develop and embrace this technology. An important question is how much further quantum security can go.

Today we get an answer of sorts thanks to the work of Michail Loulakis at the National Technical University of Athens in Greece and a few pals who have worked out how to exploit quantum mechanics to securely identify individuals.

Quantum biometrics, they say, makes identification more accurate and harder for a malicious user to foil. What’s more, the team uses the laws of physics to quantify exactly how good quantum biometrics can be.

The new technique is based on the well-known ability of the human eye to detect single photons. Our light-detection machinery relies on rhodopsin molecules in retinal rod cells to detect single photons and then on a complex mechanism of phototransduction to send this signal to the brain.

Experiments dating back to the 1940s show that humans can become aware of a flash of light containing just a handful of photons. This photon detection process is a quantum mechanism and is governed by the laws of quantum physics. However, the actual probability of detecting the flash also depends on the environment within the eye.

This environment determines the number of photons that arrive at the retina and the path they take. So important factors are the optical losses associated with the passage of light through the cornea, the anterior chamber, the pupil, the lens, and the vitreous humor.

The probability of detection also depends on how the light is absorbed at a specific spot on the retina, something that varies across the entire retina.

Measuring the probability of detection is straightforward. The experiments involve repeatedly sending a flash of light into the eye and counting how often the subject becomes aware of it.

By lumping together all the environmental factors into a single parameter called alpha, physicists can then calculate the probability of detection.

Quantum biometrics turns this method on its head. This process starts by assuming a certain probability of detecting a flash and then using the same experiments to measure alpha. In particular, Loulakis and co propose measuring how alpha varies across the field of vision.

The way alpha changes—the alpha map—depends on the unique pattern of nerves, blood vessels, and light-sensitive cells in the eye and so should be unique for all individuals. That makes an alpha map a good biometric signature (which must obviously be kept secret).

Once the map has been measured, the task is to use it to identify an individual. This is where the laws of quantum physics come in so handy because they place well-defined limits on how well an eavesdropper can foil the system.

The identification process is then straightforward. Loulakis and co propose beaming a random pattern of flashes into the eye but to vary the intensity of light in each flash. This pattern is carefully designed to exploit the alpha map so that it is detected as a recognizable pattern by a person with a specific alpha map but seems random to anyone else.  

A malicious eavesdropper, Eve, cannot easily foil this system. One way is for Eve to guess the value of alpha and respond accordingly. But the chances of this being successful can be made arbitrarily small by increasing the number of points at which alpha is measured.

Another way is for Eve to attempt to measure alpha in the subject’s eye. But Loulakis and co say this would require measurement techniques that are well beyond the state of the art.  

An important question is how many measurements are necessary to properly identify an individual. That depends on how accurate the identification needs to be, and there are two ways it can go wrong. The first is a false positive—falsely identifying Eve as the subject. The second is a false negative—misidentifying the subject.

“The probabilities for a false positive and a false negative identification of this biometric technique can readily approach [one in 1 billion] and [one in ten thousand], respectively,” say Loulakis and co.

Loulakis and co say that it should be possible to identify an individual with this level of accuracy using only six interrogations. “Practically, six interrogations can be realized in less than one minute of test time,” say the team.

That’s interesting work that maps out a method for quantum biometrics. However, the team glosses over a number of potential problems. One important question is how to accurately measure anybody’s alpha map in the first place. There is no clear answer to that.

Another problem is how alpha varies over time. Everybody’s eyesight deteriorates as they get older, which suggests that an alpha map would have an expiration date of uncertain length.

There is also the possibility that alpha might vary over much shorter time scales. Most people experience changes in vision with factors such as colds and flu, alcohol consumption, and even with floaters passing across the field of vision. If alpha maps are ever to be considered a viable biometric signature, significant work will be needed to characterize their utility.

Nevertheless, the notion of quantum biometrics underlines the increasing interest in understanding the role of quantum processes in biology. There is clearly much to learn.

Ref: arxiv.org/abs/1704.04367: Quantum Biometrics with Retinal Photon Counting

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.