Industry Insights
« Back To Vision & Imaging Industry Insights
Member Since 1984
AIA - Advancing Vision + Imaging has transformed into the Association for Advancing Automation, the leading global automation trade association of the vision + imaging, robotics, motion control, and industrial AI industries.
Content Filed Under:
Industry:
Chemical Manufacturing
and Electronics/Electrical Components
Chemical Manufacturing
, Electronics/Electrical Components
, Glass
, Military/Defense
, Pharmaceutical
, Security/Surveillance
, and Semiconductor
Application:
Material Removal / Cutting / Deburring / Grinding / Non-Visible Inspection
Material Removal / Cutting / Deburring / Grinding / Non-Visible Inspection
See More
Infrared Vision: More to It Than What Meets the Eye
POSTED 04/06/2015 | By: Winn Hardin, Contributing Editor
A s
Challenges remain, however, for an equipment class that uses a number of different exotic sensor materials and architectures, from the purely electronic to hybrid microelectromechanical systems (MEMS). Choosing the right solution requires a thorough knowledge of the technologies, their benefits, and limitations.
Growing Application Base
Today, the electronics and semiconductor industries are among the largest civilian applications for IR imaging systems, according to Jens Hashagen, product manager at Allied Vision (Stadtroda, Germany). Led by military and followed by electronics and related industries, including photovoltaics industry, infrared cameras can be found in most major industrial markets, including chemical, pharmaceutical, metal and glass production, agriculture, plus scientific, medical, traffic and security industries. “Hyperspectral imaging is also a very strong market for SWIR cameras to identify specific material properties,” Hashagen says.
Picking up where the visible spectrum drops off at around 750 nm, the infrared spectrum is split into five sections: near (750 nm–1.0 µm), short (1.0-2.7 µm), midwave (3-5 µm), longwave (8-14 µm), and far (14 to 1000 µm, or 1 mm). In summary, the infrared spectrum is huge compared to the approximately 400nm-wide visible spectrum. (Hashagen points out that CCD/CMOS sensor-based near-infrared (NIR) cameras are only sensitive up to 1100 nm, as silicon becomes transparent beyond this wavelength.)
“There’s always someone coming up with a new application in the infrared, particularly in the shortwave,” says Chris Johnston, president of Sierra-Olympic Technologies (Hood River, Oregon), which specializes in infrared imaging systems. “Shortwave infrared cameras are in applications all over the place, in pharmaceuticals, agriculture, airborne imaging and forest-fire detection.”
In the midwave, “the trend is confined primarily to military applications because of the exotic semiconductor sensors and the expensive coolers that midwave requires,” Johnston says. “Within the midwave, the military always has a new requirement, most recently high-definition midwave cameras at 1920 x 1080 pixels. One high-definition midwave camera with a suitable optic can cost $100,000 USD, and the military is one of the few buyers with that sort of budget.”
Johnston also cites a trend toward smaller pixels in the midwave, with the standard being 15 µm today and 12-µm, 10- µm, and even smaller pixels either available or in development. “The government is also interested in two-color sensors where each pixel is sensitive to both the mid- and long-wave infrared,” he adds.
But it’s the longwave infrared that represents a big growth area. “I would estimate that uncooled microbolometer-based cameras represent 98% of all IR cameras, maybe even more,” Johnston says. “Dominant applications are thermography, firefighting, enhanced vision, security/surveillance, R&D, and machine vision. These range from the smaller quarter VGA cameras from DRS and others up to 3 megapixels for some military applications with multimillion-dollar budgets.”
“While the volume comes from less-expensive, low-resolution long-wave cameras, there’s also a push to develop the most cost-effective optics and optical materials,” says Johnston. “The industry is pushing to bring down the cost of both the sensor and optics. You see the same trends in IR optics that you see in the visible optics world. To achieve low cost, IR optical vendors are exploring novel molded optics technologies. Low-cost optics in molded chalcogenide glasses and sintered zinc selenide are now available for low-cost, high-volume applications.”
Machine Vision Compatible IR Cameras
Microbolometer-based cameras can be found in many first-responder kits thanks to their ability to identify hot spots and thermal signatures from injured persons, and they are being used for machine vision as well.
Sierra-Olympic Technologies has recently introduced compact uncooled VOx microbolometer thermal IR imagers for easy configuration into systems for machine vision tasks such as process monitoring and other industrial imaging applications. Sierra-Olympic’s Viento cameras provide 14-bit Camera Link digital output, feature analog video (NTSC and PAL), and have standard mountings, which make them easy to integrate and very affordable for industrial applications that require general-purpose thermal imaging, according to Johnston.
The near-infrared (NIR) and SWIR spectral bands are receiving the most attention from the machine vision field because of their utility and higher resolution for commercial imaging applications. Industrial camera manufacturers JAI and Teledyne DALSA have both recently introduced multi-sensor line-scan cameras that include both visible and NIR linear arrays inside the cameras.
“In 2011, we acquired infrared and scientific camera manufacturer VDS Vosskühler,” says Allied Vision’s Hashagen. “Last year, we released a new SWIR camera from that acquisition. The new camera platform is optimized for industrial applications. We added Power over Ethernet; machine vision compliant feature handling mechanisms (GenICam); multiple lens mounting options; and comprehensive I/O control functions for triggering and synchronization. We also added the latest VGA indium gallium arsenide (InGaAs) sensor from Sofradir in France, which allows us to run the camera at more than 300 frames per second at full resolution, or up to 13.75 kHz if you window the sensor to a lower resolution. This makes it the fastest GigE SWIR camera on the market today.”
Allied Vision also enhanced the image-correction handling capabilities required by all SWIR cameras that use InGaAs sensors. “Non-uniformities occur between the InGaAs active area and the indium bump-bonded silicon readout chip,” Hashagen explains. “Unlike other cameras, we’ve implemented non-uniformity, defective pixel, and background correction inside the camera, which gives you very good image quality.”
U.S. Eases ITAR Regulations
According to Sierra-Olympic’s Johnston, “historically, all InGaAs cameras in the U.S. have been classified as defense articles under the International Traffic in Arms Regulation (ITAR) law, and thus have required strict licensing controls for any shipment or transfer of information related to these items to any non-U.S. person. For commercial use and transactions related to these items, most commercial customers find ITAR severely restrictive and will tolerate the licensing controls only under limited circumstances. There recently has been some easing of restrictions of some classes of InGaAs cameras, and under the commodity jurisdiction process, have been reclassified as dual-use and thus regulated by the more favorable Export Administration Regulations (EAR) law under the U.S. Department of Commerce. While still a controlled item and subject to licensing for exports to most countries, there will be instances where a class of InGaAs cameras will be exportable under license exemptions. This is a very big deal for InGaAs. It is still early in this reclassification to EAR, and we consider this a positive development.”
Expanding the customer base for SWIR cameras will lead to larger volumes, pushing the price for InGaAs-based cameras down, which is good news for the machine vision industry.
Challenges remain, however, for an equipment class that uses a number of different exotic sensor materials and architectures, from the purely electronic to hybrid microelectromechanical systems (MEMS). Choosing the right solution requires a thorough knowledge of the technologies, their benefits, and limitations.
Growing Application Base
Today, the electronics and semiconductor industries are among the largest civilian applications for IR imaging systems, according to Jens Hashagen, product manager at Allied Vision (Stadtroda, Germany). Led by military and followed by electronics and related industries, including photovoltaics industry, infrared cameras can be found in most major industrial markets, including chemical, pharmaceutical, metal and glass production, agriculture, plus scientific, medical, traffic and security industries. “Hyperspectral imaging is also a very strong market for SWIR cameras to identify specific material properties,” Hashagen says.
Picking up where the visible spectrum drops off at around 750 nm, the infrared spectrum is split into five sections: near (750 nm–1.0 µm), short (1.0-2.7 µm), midwave (3-5 µm), longwave (8-14 µm), and far (14 to 1000 µm, or 1 mm). In summary, the infrared spectrum is huge compared to the approximately 400nm-wide visible spectrum. (Hashagen points out that CCD/CMOS sensor-based near-infrared (NIR) cameras are only sensitive up to 1100 nm, as silicon becomes transparent beyond this wavelength.)
“There’s always someone coming up with a new application in the infrared, particularly in the shortwave,” says Chris Johnston, president of Sierra-Olympic Technologies (Hood River, Oregon), which specializes in infrared imaging systems. “Shortwave infrared cameras are in applications all over the place, in pharmaceuticals, agriculture, airborne imaging and forest-fire detection.”
In the midwave, “the trend is confined primarily to military applications because of the exotic semiconductor sensors and the expensive coolers that midwave requires,” Johnston says. “Within the midwave, the military always has a new requirement, most recently high-definition midwave cameras at 1920 x 1080 pixels. One high-definition midwave camera with a suitable optic can cost $100,000 USD, and the military is one of the few buyers with that sort of budget.”
Johnston also cites a trend toward smaller pixels in the midwave, with the standard being 15 µm today and 12-µm, 10- µm, and even smaller pixels either available or in development. “The government is also interested in two-color sensors where each pixel is sensitive to both the mid- and long-wave infrared,” he adds.
But it’s the longwave infrared that represents a big growth area. “I would estimate that uncooled microbolometer-based cameras represent 98% of all IR cameras, maybe even more,” Johnston says. “Dominant applications are thermography, firefighting, enhanced vision, security/surveillance, R&D, and machine vision. These range from the smaller quarter VGA cameras from DRS and others up to 3 megapixels for some military applications with multimillion-dollar budgets.”
“While the volume comes from less-expensive, low-resolution long-wave cameras, there’s also a push to develop the most cost-effective optics and optical materials,” says Johnston. “The industry is pushing to bring down the cost of both the sensor and optics. You see the same trends in IR optics that you see in the visible optics world. To achieve low cost, IR optical vendors are exploring novel molded optics technologies. Low-cost optics in molded chalcogenide glasses and sintered zinc selenide are now available for low-cost, high-volume applications.”
Machine Vision Compatible IR Cameras
Microbolometer-based cameras can be found in many first-responder kits thanks to their ability to identify hot spots and thermal signatures from injured persons, and they are being used for machine vision as well.
Sierra-Olympic Technologies has recently introduced compact uncooled VOx microbolometer thermal IR imagers for easy configuration into systems for machine vision tasks such as process monitoring and other industrial imaging applications. Sierra-Olympic’s Viento cameras provide 14-bit Camera Link digital output, feature analog video (NTSC and PAL), and have standard mountings, which make them easy to integrate and very affordable for industrial applications that require general-purpose thermal imaging, according to Johnston.
The near-infrared (NIR) and SWIR spectral bands are receiving the most attention from the machine vision field because of their utility and higher resolution for commercial imaging applications. Industrial camera manufacturers JAI and Teledyne DALSA have both recently introduced multi-sensor line-scan cameras that include both visible and NIR linear arrays inside the cameras.
“In 2011, we acquired infrared and scientific camera manufacturer VDS Vosskühler,” says Allied Vision’s Hashagen. “Last year, we released a new SWIR camera from that acquisition. The new camera platform is optimized for industrial applications. We added Power over Ethernet; machine vision compliant feature handling mechanisms (GenICam); multiple lens mounting options; and comprehensive I/O control functions for triggering and synchronization. We also added the latest VGA indium gallium arsenide (InGaAs) sensor from Sofradir in France, which allows us to run the camera at more than 300 frames per second at full resolution, or up to 13.75 kHz if you window the sensor to a lower resolution. This makes it the fastest GigE SWIR camera on the market today.”
Allied Vision also enhanced the image-correction handling capabilities required by all SWIR cameras that use InGaAs sensors. “Non-uniformities occur between the InGaAs active area and the indium bump-bonded silicon readout chip,” Hashagen explains. “Unlike other cameras, we’ve implemented non-uniformity, defective pixel, and background correction inside the camera, which gives you very good image quality.”
U.S. Eases ITAR Regulations
According to Sierra-Olympic’s Johnston, “historically, all InGaAs cameras in the U.S. have been classified as defense articles under the International Traffic in Arms Regulation (ITAR) law, and thus have required strict licensing controls for any shipment or transfer of information related to these items to any non-U.S. person. For commercial use and transactions related to these items, most commercial customers find ITAR severely restrictive and will tolerate the licensing controls only under limited circumstances. There recently has been some easing of restrictions of some classes of InGaAs cameras, and under the commodity jurisdiction process, have been reclassified as dual-use and thus regulated by the more favorable Export Administration Regulations (EAR) law under the U.S. Department of Commerce. While still a controlled item and subject to licensing for exports to most countries, there will be instances where a class of InGaAs cameras will be exportable under license exemptions. This is a very big deal for InGaAs. It is still early in this reclassification to EAR, and we consider this a positive development.”
Expanding the customer base for SWIR cameras will lead to larger volumes, pushing the price for InGaAs-based cameras down, which is good news for the machine vision industry.
Vision in Life Sciences
This content is part of the Vision in Life Sciences curated collection. To learn more about Vision in Life Sciences, click here.