Wide Area Iris Capture

Wide-area Rapid Iris Image Capture with Pan-tilt-zoom Cameras

Ruigang Yang, Laurence Hassebrook, Stacy Wilson, Brent Seales, Xinyu Huang, Charles Carpenter

Center for Visualization and Virtual Environment
University of Kentucky

Abstract

In response to the DHS interest in fast biometric measurement, this project will develop a system to rapidly capture iris images of moving human subjects at long range. Working in concert with a currently available commercial iris identification software package, our system will provide fast, accurate, and automated biometric identification first for homeland security and also for other fields requiring identification or authentication.

From a technical standpoint, the goal is to develop a complete software/hardware package that is expected to advance the state of the art in iris scan technology by (a) increasing the scan range by five to ten fold (i.e., distances greater than 10 feet) and (b) allowing continuous iris scanning of moving subjects.

We propose the development of computer vision techniques to automatically detect, track, and extract high-resolution iris images using steerable pan-tilt-zoom (PTZ) cameras. The use of our solution, shown in the figure below, is meant to be simple and reliable. Imagine a target area equipped with a mix of inexpensive fixed stereo cameras and high-resolution PTZ cameras. Our fixed stereo cameras detect motion and have the ability to track individuals in 3D as they walk into the field of view. This real-time 3D information is used to steer the PTZ cameras to zoom-in and focus on the subject's face. Servo-controlled active yet invisible illumination will be used to enhance the contrast of the face image. Iris images are then detected and converted to a canonical format that can be sent to existing iris identification software, such as the widely adopted package from Iridian Technologies (http://www.iridiantech.com/).

Our team has a unique combination of expertise in visual tracking, servo control, optics, and hardware design and manufacturing. Together with a partnership with a leading provider in iris identification software, we expect this project to quickly demonstrate a research prototype with strong potential for commercialization.

System Architect