IDloop researcher pitches advantages of structured light for 3D contactless biometrics
Capture methods for fingerprint biometrics are continuing to evolve, and 3D contactless fingerprints images measured with structured light is showing potential for addressing some of the field’s persistent challenges, attendees heard in the latest virtual lunch talk from the European Association for Biometrics (EAB).
IDloop Head of Research Dr. Tom Michalsky delivered the presentation on ‘3D Imaging of Biometric Features of Human Skin.’ IDloop makes contactless biometric devices to enable faster workflows and lower maintenance costs.
Michalsky reviewed the standard method of contact fingerprint biometrics collection through frustrated total internal reflection. This is the usual scanner architecture, based on a light source, a prism, and a lens and camera. TFT sensors are also increasingly familiar, and have the advantage of offering the correct scale, and avoiding prisms, but according to Michalsky, they retain other disadvantages of contact-based scanners, such as leaving latent prints and the possibility of bacterial transmission.
The development of 2D contactless fingerprinting followed, and is showing improved usability over contact scanners. Recent research explores the problem of scale ambiguity depending on the distance of the hand from the camera lens. Despite attempts to compensate for this scale issue, smaller fingers tend to deliver a lower matching rate with contactless images. Contactless 2D fingerprints can also be affected by perspective distortions, Michalsky says, with side-views altering the distance between capillary lines and minutiae in a way that cannot be corrected for without more information.
Michalsky also discussed what he calls ‘2.5D imaging’ for fingerprints, which is based on a rough 3D structure of the finger, which addresses some but not all of the challenges with 2D imaging.
IDloop’s optical contactless technology, by contrast, uses absolute scale with no perspective distortions. The images are created with structured light, projected onto the subject. This approach delivers resolution below 10 micrometers, he says.
Other common 3D measurement techniques like stereo vision and time-of-flight were also described in the presentation. Stereo vision cannot be used for 3D fingerprint imaging, due to its inadequacy with periodic structures, while time-of-flight is restricted to roughly 1cm depth resolution.
Structured light 3D fingerprint and footprint biometrics
The development of structured light traces back to the sixties, but recent advances in the application of the technology to biometrics have dealt with specific challenges. One example involves artifacts caused by the minute motion of a finger shaking slightly as its image is being captured. Reconstruction algorithms are used to perform motion correction in IDloop’s system.
Structured light systems must also be very finely calibrated, and researchers found the standard calibration technique lacking, with unacceptable levels of distortion.
Michalsky dates true contactless 3D fingerprints free of motion artefacts only to this year. Pores, a third-level feature, can be measured, in addition to more easily visible biometric features.
The researcher also shared data on the fingerprint ridge and valley structure of a two year-old subject, which shows the continuing difficulty of biometric verification of young children, but also suggest clearer images may be easier to capture with contactless 3D imaging than other methods.
True 3D contactless fingerprinting enables the use of local curvature as an additional data-point for enhanced matching accuracy, according to Michalsky.
Michalsky suggests some advantages of the method in application. For the identification of small children, footprints taken with the same method could possibly be used even to reunite kidnapped babies with their parents. It could also be used in presentation attack detection, where it would easily identify printed spoof artefacts.