With Infrared, Military Owns More Than the Night

Hank Hogan, Contributing Editor, hank.hogan@photonics.com

Short-, long- and mid-wave IR imaging helps defense agencies find targets and even determine intent.

At the US Army’s Night Vision and Electronic Sensors Directorate, the goal is to see – day or night – through smoke, fog, dust or any other airborne obscurant. The directorate conducts sensor research and development so that soldiers can see effectively in a variety of conditions.

This capability is important so that soldiers can tell whether someone is holding a hoe or a rifle. And seeing more clearly can better reveal intent, an important piece of information: If the object is indeed a weapon, it’s also important to know if it’s being carried in a relaxed pose or in an aggressive stance.

“This is especially true for today’s conflicts where combatants and noncombatants are mixed together,” said Thomas Bowman, director of the directorate’s ground combat systems division.

The US Army’s Night Vision and Electronic Sensors Directorate runs a microfactory laboratory at Fort Belvoir, Va., as part of its IR sensor research and development. Courtesy of Night Vision and Electronic Sensors Directorate Public Affairs.

Part of the Army’s research and development is in infrared imaging, which enables the spotting of targets, intruders and hidden bombs, thereby protecting troops and making the application of force more discriminating.

The Army constantly seeks to improve the resolution and capability of sensors. At the same time, the goal is also to make systems less expensive, bulky, heavy and power-hungry. These targets are particularly important in technology carried by troops.

Fortunately, there are innovations that promise to help correct some of these problems. They include improvements in long- and mid-wave IR wavelengths, which run 8 to 15 and 3 to 8 µm, respectively. Enhancements also are afoot in short-wave IR, which spans 1.4 to 3 µm. Together, with the fusion of information from other sensors, these should improve IR imaging in military applications.

Enhanced night-vision goggles like these, developed for the US Army’s Night Vision and Electronic Sensors Directorate, fuse IR-based thermal imaging with visible image intensification. Courtesy of Night Vision and Electronic Sensors Directorate Public Affairs.

“There is much more opportunity to develop night-vision sensors, giving our soldiers greater situational understanding at increased ranges while reducing size, weight, power and cost,” Bowman said.

The long and medium view

A supplier of long- and mid-wave IR systems to the military and commercial markets, Wilsonville, Ore.-based Flir makes its sensors out of materials that are expensive to manufacture, such as the compound semiconductor cadmium telluride. That affects system price.

So, too, does volume. A maker of visible-wavelength camcorders might make and sell millions of units a year; Flir makes and sells orders of magnitude less, said John Lester Miller, chief technology officer of Flir’s surveillance division.

The situation might be changing, thanks to the increasing use of infrared in the automotive industry. For instance, significant volumes may be on the horizon due to the advent of driverless vehicles.

Thanks to compressive sensing imaging, a single element can replace a focal plane array, potentially lowering system costs. Courtesy of InView Technology.

In military applications, the trend is to go big, Miller said. “There’s lots of development of large focal plane arrays and HD format arrays.”

Flir has a 1280 x 720-pixel IR focal plane array in production today, with indium antimonide the sensor material and a pixel pitch as low as 12 µm. That pitch is now approaching the wavelength of the light being captured, at least for long-wave IR. If it drops even more, it will have implications for sensor sensitivity.

For the future, Miller points to such promising materials as strained superlattice gallium arsenide. Besides being highly sensitive and low cost, this material allows a warmer operating temperature and thereby cuts power consumption.

Because of the power savings and accompanying maintenance cost reductions, the military would like to use uncooled IR systems. However, with current technology, there is some loss in performance.

The New York skyline imaged by an IR sensor during a demonstration. Courtesy of HGH Infrared.

For instance, with a cooled long-wave IR sensor, it’s possible to spot people at distances of up to 3 km with a panoramic detector. An uncooled sensor, on the other hand, can only do the same at 2 km or less, said Josh Howlett, sales manager for North America at Cambridge, Mass.-based HGH Infrared Systems.

The company currently makes cooled IR systems. Its IR Revolution 360 product scans the perimeter of an area once a second, acting much like an infrared version of radar.

HGH Infrared’s system creates a panoramic image by scanning a linear detector across a scene. Future versions of the product will use an array sensor, significantly improving performance. The company is also working on an uncooled mid-wave sensor, which will appear by the end of 2013. The move down in wavelength will result in a move up in the ability to see in certain environments, Howlett said.

“It helps us in environments that are moisture-rich, such as maritime. Mid-wave has a little bit higher performance,” he said.

At the short end

Because of its wavelength, short-wave infrared scatters less off smog, haze, smoke and dust than does visible light, said Tara Martin, director of business development at UTC Aerospace Systems in Princeton, N.J. The company makes indium gallium arsenide sensors that detect light out to about 1700 nm.

Short-wave IR sensors can offer better imaging through smoke (top), fog (middle) and other atmospheric obscurants (bottom) than visible sensors can. Courtesy of UTC Aerospace Systems.

Less scattering leads to clearer imaging, Martin said. “We’ve been compared to many silicon imagers with far more pixels than what we have, but we have achieved much better identification capability.”

The current state of the art for the company’s technology is a pixel pitch of 12.5 µm and an array of 1280 x 1024 pixels. Those numbers are worse than what’s found in visible sensors, where multiple megapixels are routine.

Short-wave IR has been used in night-vision applications. Because of the desire for low-light performance, pixel size is important, as more area means more photons collected. As with other sensors, power consumption also is important.

Indium gallium arsenide imagers are more expensive than those made of silicon, but that’s understandable given the differences in production volume of the two materials. Once quantities go up, the cost for indium gallium arsenide sensors should go down.

One company hoping to speed this cost cutting along – and thereby to expand the use of short-wave IR sensors – is Austin, Texas-based startup InView Technology. The company uses a compressive sensing imaging approach originally developed at Rice University in Houston.

InView’s products employ a digital micromirror device to modulate the light onto a detector, tilting the individual micromirrors so as to take a carefully chosen set of successive measurements. The result is a significant compression of up to 90 percent. That is, a megapixel image can be produced from 100,000 measurements.

What this means is that a 4 x 8 pixel array can replace a megapixel focal plane array, even at video frame rates. As a result, a much smaller amount of expensive indium gallium arsenide is needed. That reduces the cost of an imaging system.

“We can really dramatically impact that price by a factor of five,” said CEO Bob Bridge.

The company shipped its first commercial product this year. The approach can be extended to mid-wave IR and is also suitable for multispectral imaging, as it’s possible to put a silicon detector on top of an indium gallium arsenide one. The infrared will penetrate the upper chip, while the visible light is captured by it.

The future and fusion

Further out, in terms of potential product availability, are very inexpensive mid- and long-wave IR sensors, if research under way at Northeastern University in Boston pans out. There, physics professors Srinivas Sridhar and Swastik Kar head a group studying how to turn graphene into a suitable material for IR sensors.

By itself, graphene, a 1-D array of carbon atoms, doesn’t respond in a useful way to infrared photons. When combined with boron nitride in the right configuration, however, the result is a material system that exhibits a temperature-dependent change in resistance, one that Sridhar reported was competitive with what’s used in current uncooled thermal IR detectors. A key advantage of the new material is how it’s manufactured.

“The process by which we are making this can be easily scaled up, and that’s very important if you want to have very cheap thermal imagers,” Sridhar said.

Sridhar thinks the material will be inexpensive enough that IR imaging could become a feature found on cellphones. The Northeastern group’s project is being done in collaboration with the military, which has expressed interest in the sensor technology.

Fusing visible/near-IR intensified imaging with thermal imaging reveals someone standing in a darkened doorway, allowing detection and showing details that would be difficult using one spectral band alone. Courtesy of Naval Sea Systems Command.

Finally, it should be noted that infrared imaging by itself may not be enough. For instance, systems that work well at sea and near the ocean are of interest to Keith Lannan, the technical warrant holder for electro-optic and infrared sensing systems at the Naval Sea Systems Command. Lannan’s responsibilities are for surface vessels, not submarines.

The sea is prone to fogs, spray and other imaging challenges, and the Navy would like imaging technology that works well in all kinds of weather conditions. In addition, imaging at greater distances is an important capability, Lannan said.

The two requirements might best be tackled by using improved imagers operating in different spectral bands, as each slice of spectrum offers different strengths and weaknesses. Systems combining, or fusing, information collected at different wavelengths could better see through the obscurants in the air, a capability that would bring military benefits.

“You may be able to detect there is a target there, whereas without fused imagery, you might be impacted by the environment,” Lannan said.