An echo from the telecom boom
Tunable optical filter helps IR imaging.
Tunable optical filter helps IR imaging.
The telecommunications boom of the late 1990s produced many venture-capital-funded startups that were intended to commercialize university research. One of them, Aegis Semiconductor (Woburn, MA), used research from Princeton University to create low-cost tunable-optical-filter components and modules for telecommunications. In 2005, the founder of Aegis, Matthias Wagner, led a spin-off company, RedShift Systems (Waltham, MA), to take further advantage of the technology in a nontelecommunications arena: thermal imaging. Potentially large volumes of sales to the security and thermography markets attracted a $12 million first round of venture-capital funding to boost the initial infusion from Aegis.
Three types of infrared cameras have dominated the detection of long-wavelength IR (LWIR) radiation, which lies within the 8 to 14 µm spectrum. Low-cost and less-sensitive microbolometer-based units use vanadium oxide or amorphous silicon, and sensitive-yet more expensive-cameras are based on mercury cadmium telluride (HgCdTe or MCT) and gallium arsenide (GaAs) quantum-well IR photodetector (QWIP) detectors.
To lower the cost of these LWIR cameras, RedShift has developed a low-cost optical thermal camera that uses an optical wavelength converter, the Thermal Light Valve (TLV), based on amorphous-silicon and silicon nitride thin films originally developed by Aegis. RedShift is now releasing TLV-based camera modules for use by OEMs. The first evaluation modules are scheduled for shipment to qualified OEMs later this year.
The Aegis tunable telecommunications filters are composed of a transparent substrate, integrated resistive heating layer, and dielectric mirrors sandwiched around a thermo-optic tunable semiconductor layer. The wavelength of light transmitted through the filter is a function of the temperature of the semiconductor layer. To change the transmitted wavelength, the current in the resistive heating layer is changed, resulting in a tunable filter. RedShift has used this tunable semiconductor technology to build a 160 x 120 thermal imager in which the wavelength of the tunable layer is shifted not by resistive heating but by incident LWIR radiation.
Because the filter’s resonant wavelength changes are due to this incident LWIR radiation, probing the device with a fixed-wavelength laser and measuring the reflected laser light provides a method of measurement of the LWIR frequency. RedShift has developed a small thermal-imaging camera that uses the technology to image reflected laser light onto a standard CMOS imager. The TLV is mounted into a vacuum-packaged housing with a LWIR lens. Incident LWIR radiation incident on the TLV then changes the filter resonant wavelength of each pixel in the array. By flooding the array with an 850 nm vertical-cavity surface-emitting laser, reflected laser light is then focused onto a CMOS imager.
Because the near-IR probe signal reflected from the TLV depends on the incident LWIR radiation, the intensity of light received by the CMOS imager is effectively modulated by the LWIR signature of the observed scene. Images from the CMOS imager are then processed and encoded into NTSC format, which allows the camera to display images on low-cost broadcast-compatible systems.
The camera incorporates a digital-signal processor to perform image filtering and automatic gain correction of the captured image. The spectral sensitivity of the TLV-based camera in the 8 to12 µm LWIR region is 150 mK, a figure comparable, for example, with that of uncooled microbolometers but still far less than the 25 mK for MCT-based devices.
RedShift estimates that OEM cameras based on this technology can be built for about $1000, making them more cost-effective than microbolometer-based devices. At the recent SPIE Defense & Security Symposium (April 2006; Orlando, FL), RedShift demonstrated a prototype of the camera and an OEM imaging module the company plans to offer to camera manufacturers targeting LWIR firefighting, security, and thermography applications.
These markets have long been ripe for a low-cost technology that can make this part of the infrared spectrum accessible at an affordable price. Perhaps some of the over-enthusiasm of the telecom boom era will fulfill a bit of destiny here.
CONARD HOLTON is editor in chief of Vision Systems Design; e-mail: firstname.lastname@example.org.