Security veins: Advanced biometric authentication through AI and infrared

Hyperspectral imaging is a technology that detects slight differences in color to pinpoint the characteristics and conditions of an object. While a normal camera creates images using red, green, and blue, a hyperspectral camera can obtain over 100 images in the visible to near-infrared light range in a single shot. As a result, hyperspectral imaging can obtain information that the human eye cannot see.

Specially Appointed Associate Professor Takashi Suzuki at the Osaka Metropolitan University Center for Health Science Innovation captured images of the palms of human hands using a hyperspectral camera and an AI-based region of interest.

Hemoglobin contained in red blood cells absorbs light, so it is possible to observe the state of the blood vessels in the palm. Since the distribution pattern of blood vessels differs from person to person, it is also possible to differentiate between individuals.

Further, vein patterns are not visible on the surface of the skin like the face or fingerprints, so this bioinformation is considered highly secure as it cannot be easily read.

To test this, Dr. Suzuki developed a method for identifying biometric information regardless of position or orientation using AI-based image recognition. The findings were published in the Journal of Biomedical Optics.

By superimposing the images in order of wavelength and cutting them based on the coordinates on the palm obtained through AI, the researcher was able to obtain images with more accurate positionings, smaller sizes, and greater information content than conventional methods.

"It was possible to distinguish between the subjects. Furthermore, accuracy of the developed method was verified and a high discrimination accuracy was confirmed," stated Dr. Suzuki.

"Biometric authentication using hyperspectral images provides remarkably high security through the palm of a hand, thus it could even be used as keys to a house. If the capability to read the state of health from the hyperspectral imaging of the palm becomes possible, a daily health management system could be developed with health data obtained through biometric unlocking."