Tiny Panoptes technology holds promise for military surveillance and iris recognition
By Jude Garvey
April 19, 2010
Researchers at the Southern Methodist University (SMU) in Dallas are developing new miniature camera technology and an iris recognition application built on a high-resolution, light and compact platform known as Panoptes. The technology is designed to help the military and border patrol to track combatants in dark caves or alleys and airport security personnel to quickly and unobtrusively identify a subject from an iris scan.
Led by Marc Christensen from SMU, the project has received a $1.6 million grant by the Department of Defense.
The name “Panoptes” comes from the Greek mythological character – Argos Panoptes – a giant sentry with a hundred eyes and in case you were curious, is also short for: "Processing Arrays of Nyquist-limited Observations to Produce a Thin Electro-optic Sensor". The new applications have been dubbed AIM-CAMS, for "Active Illumination with Micro-mirror-arrays for Computational Adaptive Multi-resolution Sensing," and Smart-Iris, for "SMU's Multi-resolution Adaptive Roving Task-specific Iris Recognition Imaging System."
"The Panoptes technology is sufficiently mature that it can now leave our lab, and we're finding lots of applications for it," said Christensen, an expert in computational imaging and optical interconnections. "This new money will allow us to explore Panoptes' use for non-cooperative iris recognition systems for Homeland Security and other defense applications. And it will allow us to enhance the camera system to make it capable of active illumination so it can travel into dark places — like caves and urban areas."
Panoptes technology is capable of producing sharp and clear images without the size and weight of a traditional camera system. It uses arrays of agile and finely controlled microelectromechanical system (MEMS) mirrors - which are integrated with low-resolution sub-imagers on a silicon base - allowing sampling of a wide field of view. The MEMS mirrors are steerable and can redirect plexiglas sub-imagers to zoom in on regions of interest. The images are stored in the onboard computer and are converted to high-resolution images by a super-resolution algorithm.
The sub-imagers are tiny, flat, off-axis-shaped paraboloids and are small enough to fit on the surface of a small coin. This makes them ideal to be tiled, undetected, almost anywhere – on the underside of a small drone, along the walls of a hallway or on the outside of a soldier’s helmet. They measure just 8mm x 5.7mm x 4mm (approx. 0.32” x 0.22” x 0.16”) and have an effective focal length of 4mm (approx. 0.16”).
The micro mirror array is honeycomb-shaped and is made up of 61 hexagonal mirrors - each with three actuators that can move and control the mirrors. The usable circular aperture is 3.9 mms (approx. 0.15”) in diameter. The researchers claim that the digitally restored image - whilst not super-resolution - approaches optical limit. The architecture of Panoptes is unique as it can adapt its field of view to focus on a region of interest, therefore capturing only those images value, and preserving computer power.
Defense contractor Northrop Grumman has field-tested the Panoptes imaging in tactical environment stimulations and it is currently being tested independently by Draper Laboratory. According to Christensen, it could be available for fielded demonstrations by late 2011.
The Smart-Iris will use the adaptive resolution of Panoptes and iris recognition technology. "It's very challenging to get the resolution with a wide field-of-view camera, but with a zoom camera, it's hard to find the iris because it's like looking through a soda straw," Christensen said.
Iris recognition is the is the most accurate biometric available because no two irises are alike, said Delores Etter, a former Deputy Under Secretary of Defense who leads SMU's Biometrics Engineering Research Group. However, there are problems with existing technology due to glare, eyelashes, eyelids or dim lighting.
Panoptes will start the iris-recognition process with a wide field-of-view at low resolution, find a face, and finally narrow to the iris. The Smart-Iris will then extend the range of iris acquisition, so it will be possible for people to move through a doorway and for each one to have their iris scanned – without having to pause. No more need for standing motionless with one’s eye pressed to a scanner. And the camera can maintain a high resolution and more than 150 pixels across the iris.
"Our goal is to develop an iris recognition system that is unobtrusive and accurate. We want to ensure that the right people have access, and that potential intruders are identified, all without impacting flow in high-traffic areas," said Etter, who directs the Lyle School's Caruth Institute for Engineering Education. Good news for the airline security personnel who rely on the latest technology to screen millions of people who pass through airlines security each year.
The AIM-CAMS will use Panoptes technology with Pico projectors. Using Pico projectors will make it possible for the low-resolution camera to be used in dark areas, like caves and alleyways by projecting digital pictures taken by portable devices onto any wall for large-format viewing.
SMU is collaborating on the research with Santa Clara University in California, Northrop Grumman and Draper Laboratory. Funding came from the Defense Advanced Research Projects Agency, Office of Naval Research and Army Research Laboratory.
Just enter your friends and your email address into the form below
For multiple addresses, separate each with a comma