“Biometrics are our most unique physical (and behavioral) features that can be practically sensed by devices and interpreted by computers so that they may be used as proxies of our physical selves in the digital realm. In this way we can bond digital data to our identity with permanency, consistency, and unambiguity, and retrieve that data using computers in a rapid and automated fashion.”
Devices and Sensors
Devices and sensors are any mechanical or electronic system used to enroll and capture raw biometric samples in a form that can be digitized and converted to a biometric template. For fingerprints, face, iris, voice, and keystroke, these are fingerprint sensors, digital cameras, iris cameras, microphones,and keyboards, respectively. Most fingerprint sensors are based upon either optical or capacitive techniques, but light emitting sensors and multispectral approaches are gaining adoption. Capacitive sensors can be either full-finger or swipe. It is critical to the performance of identification matching for fingerprint images to be captured at sufficient resolution (500 ppi) and contrast for the task at hand, be compressed properly with WSQ, and be free of distortion. An optical sensor uses a prism, light source, and light sensor to capture images of fingerprints. Capacitive sensors are based on a silicon chip that detects electrical currents when the finger ridges make contact. Swipe sensors do not generate image quality sufficient for one-to-many identification. Generally speaking, the quantity and consistency of the biometric samples required is a function of the size of the database that must be searched.
Capture of facial images is performed using mobile phones, consumer-grade digital SLRs, pocket cameras, and webcams. Digital facial images traditionally require an interocular resolution of about 60 pixels for one-to-one matching and 90 pixels for more accurate one-to-many matching. The more critical and challenging factor affecting facial matcher performance is consistency; achieving consistent pose, head angle, and facial expression of the subject, and brightness, contrast, sharpness, and background clutter of the full image.
Iris biometrics have also benefited from dramatic improvements in sensors. Iris matching differs from face in that it requires an infrared image of the iris to optimize the image contrast so as to facilitate machine based analysis. The degree to which a pure infrared image can be captured (with minimal “pollution” from visible light), the better matching performance is achievable. This is why off-the-shelf cameras aren’t yet used for iris image capture, and a special camera is required; a system must illuminate the iris with infrared light and then filter out other wavelengths.
Their audio capabilities and ubiquity make smartphones a particularly viable means to deploy large-scale voice biometrics for one-to-one verification. Voice biometrics are impeded by the same challenges as facial biometrics in that the capture environment can be unpredictable and inconsistent; as with facial images, background noise can interfere with the capture and matching process.