Problem

Covert and robust facial recognition of individuals is a goal for military check points, law enforcement, and border control. Facial recognition performance in the visible spectrum has greatly improved over the years. There are situations, however, in which visible imagery cannot be used for facial recognition and thermal imaging is required, particularly in dark settings.


Thermal imaging relies on temperature differences to provide contrast in scenes. When scenes don’t have significant temperature differences, recognition or identification cannot be accomplished. Thermal images of human faces rarely provide sufficient facial feature contrast necessary for recognition algorithms.

Solution

Polarization enhanced thermal imagery highlights rich geometric and textural features that are hidden in standard thermal imagery. Polaris has shown that facial recognition using our thermal polarimetric imagery yields significantly higher thermal-to-visible match rates than when using standard thermal imagery. This approach enables passive, i.e., no external illumination required, day and night capability.

Polarization addresses the capability gaps in low-light and nighttime scenarios by acquiring naturally emitted radiation in the thermal infrared spectrum from facial skin tissue. The polarimetric images capture key facial details and geometry not available in conventional thermal imagery.

Because facial recognition software uses novel algorithms capable of matching polarimetric thermal facial imagery to visible facial imagery, this approach provides interoperability with existing biometric databases containing visible-only face imagery.

Features

By adding the polarization sensing capability to existing thermal sensors, contrast can be enhanced over that of conventional thermal imaging due to differential polarization emission. The figure below shows an example of how a thermal image (center) loses the detail present in the visible image (left). The thermal polarimetric image (right) restores the detail of the visible image due to polarimetric, and not thermal, differences.

Polaris has shown that the level of detail in the polarimetric image is sufficient for thermal-to-visible facial recognition.

Source: 2016 IEEE 8th International Conference on Biometrics Theory, Applications and Systems (BTAS), “Estimation of visible spectrum faces from polarimetric thermal faces”

Optics & Photonics Special Edition

Front and Profile Face in Polarimetric Visible and Infrared
Source: Optics & Photonics News, “Changing the Paradigm in Human Identification”

Our Products & Capabilities

Polarimetric thermal cameras for passive thermal polarimetric imagery (no external illumination needed) for day/night applications

Infrared cameras with covert illumination for day/night applications

Standalone sensors to accompany existing facial recognition software

Complete facial recognition systems including all necessary hardware and software

Sensors and sensor systems to meet your SWaP and situational requirements