Shortwave infrared (SWIR) imaging is quietly earning a growing place in industrial machine vision for quality inspection. SWIR imagers, sometimes also referred to as NIR imagers, can see objects and events that vision and thermal cameras cannot. Moreover, they’re’ smaller and lighter than all thermal cameras, and cost far less than many of them (See table below). Furthermore, most indium gallium arsenide (InGaAs) SWIR cameras are all solid state, with no shutters, cooling systems or other moving parts. Some come factory set, with no need for user nonuniformity corrections throughout their entire service lives. All SWIR imagers work with plain glass optics, avoiding the thermal camera requirement for silicon or germanium lenses, which can cost 10 times as much. The installed base for InGaAs SWIR detection is steadily rising in both military surveillance and industrial imaging primary due to the low noise and simple operation of these cameras and arrays. These devices permit detection in the SWIR band with minimal cooling and electronic overhead, making the cameras similar in operation to silicon CCDs and CMOS imagers. With the increasing use of InGaAs cameras, people’s confidence in using these cameras 24/7. SWIR-based machine vision systems have proven themselves, economically and technically, across the industrial landscape.
To illustrate how imaging in the short-wave infrared with InGaAs can work, some recent applications re described below:
In metal production, SWIR-based machine vision systems have become the technology of choice for improving yields during molten-metal processing. Emission differences between hot metal and slag show up more clearly in the 900–1,700 nm bandwidth than in the visible or IR ranges. This difference in an image tells operators precisely when to end the process in order to maximize yield without contaminating the metal with slag. Because SWIR imagers work through glass, the cameras can monitor the 1,899–3,000° F process safely from within protective enclosures. The option to use glass windows makes SWIR cameras much more economical than thermal cameras, which require silicon, sapphire or germanium windows. In more than 15 modern metal mills, resulting gains in yield and throughput have more than justified such SWIR-based machine vision systems in just months.
Wavebands for Mainstay Imaging Systems |
This diagram shows vis, SWIR, MWIR and LWIR wavebands, and the point at which equipment can work through glass. Note that various InGaAs formulations cover the entire SWIR range. The labeled bar across the bottom represents the wavelengths that work through glass optics. Above that range, more expensive silicon and germanium optics are required. |
Not surprising, SWIR’s ability to see through glass also qualifies it for process control in glass processing. Manufacturers of glass hollowware have long sought a means to pick out defective product at the “hot end” of their process, when the pieces are still at a temperature 200–700° C. At this stage, rejects can be shunted aside and reprocessed efficiently before creating a mountain of scrap.
Because SWIR works through glass, the camera sees into and through the bottles in process, all from one side. InGaAs imagers can also be calibrated to accurately measure emissions vs. temperature, thus monitoring the object’s temperature uniformity and cooling rate. With this data, glassmakers can optimize the process and prevent shattering due to uneven cooling, thereby improving yield and quality while saving energy and reducing costs. InGaAs cameras have also proven useful in detecting major defects in bottle forming. Sometimes a defect in manufacturing creates a web of hair-like glass filaments that crisscross the inside of the bottle. When cooled and hardened, these filaments fracture into shards that fall to the bottom of the container, perhaps winding up in someone’s finished drink. Because the InGaAs camera sees through the molten glass, it picks up the temperature difference between those filaments and the bottle wall itself. This triggers an error signal that diverts the bad bottle off the line for recycling and flags a need for a process correction.
In pharmaceutical processing, SWIR-based machine vision and hyper spectral spectroscopy systems monitor liquid fill levels within opaque containers and deliver real-time chemical analyses on products moving on belts or through pipes. For example, many liquid medicines are dispensed in white plastic bottles, making visual camera inspection of fill level impossible. As water in the liquid absorbs light in the 1,440 and 1,940 nm bands, a SWIR camera in combination with incandescent backlighting of the vial easily reveals the fill level inside.
Many pharmaceutical applications involve SWIR or NIR spectroscopy for real-time or continuous measures of the presence of water, proteins, carbohydrates, fats and oils, or various hydrocarbon molecules. NIR Spectroscopy has the advantage of minimal sample preparation compared to spectroscopy in other wavebands. This is a significant advantage over FTIR techniques in both the mid- and near-infrared, in addition to NIR spectroscopy’s relative immunity to vibration or temperature drift. And in a new application to protect product authenticity, SWIR cameras monitor the placement and readability of NIR fluorescing authenticity marks on-line.
Beginning in Europe and now in the United States, recyclers have come to depend on SWIR-based spectroscopy for sorting plastics in the waste stream. Inexpensive SWIR 1,024 × 1 or 512 × 1 linescan cameras with wavelength sensitivities ranging from 1,100 to 2,200 nm are mounted on spectrographs. These instruments quickly identify the type of polymer in a container passing down a sortation conveyer. The identification output triggers a dam or paddle to divert the container into its proper bin.
Moisture is a key indicator of process control and quality in agriculture, textile processing and forest-product industries. Because water is opaque to SWIR illumination, it’s an easily measurable quantity. SWIR imaging has proven increasingly useful for gauging the health of crops, their ripeness or dryness for processing, and their overall quality by detecting differences in moisture that are invisible to the eye. One key example is the sensing of bruises under the skin of produce moving on a sorting line. Another agricultural application is simply gauging ripeness of picked produce passing on a conveyor belt. Farmers can justify the investment in SWIR machine vision solely on the basis of maintaining quality.
Likewise, in dyed textiles, moisture content indicates when a dyed fabric is dry enough for the next step, and whether dye coverage in an area is correct. In particleboard manufacturing, on-line SWIR-based machine vision systems measure moisture in chips to regulate heating and drying operations downstream.
For all of these examples, inspecting the purity of the incoming feedstock prevents contaminating expensive process equipment down the line. If incoming raw cotton or wood chips are contaminated with pieces of the plastic strapping mixed in, millions of dollars of direct damage and lost product may result. SWIR inspection of incoming goods is helping avert such disasters.
SWIR InGaAs cameras offer a combination of intrinsic benefits that thermal and visible-spectrum cameras don’t. This creates a corresponding performance/life-cycle cost advantage for SWIR, even in applications in which both thermal and SWIR can technically do the job.
In gritty industrial applications, for instance, lenses will inevitably be damaged and need replacement. Replacing a glass lens in a SWIR camera will always cost less than replacing a germanium or sapphire lens in a thermal camera. Like visible CCD and CMOS detectors, SWIR detectors respond to reflected light, which enables high-resolution images. IR cameras, by contrast, only sense heat, which results in lower resolution. The current InGaAs cameras and arrays are as simple to operate as a silicon CMOS imager, but they enable the imaging of a different wavelength band, which has proven critical in many applications.
Moreover, InGaAs-based SWIR cameras need neither shutters nor expensive cryogenic cooling of the detector array. Eliminating these mechanical components adds reliability and cost/size/weight advantages not available in most IR cameras or in SWIR cameras with InSb or HgCdTe detector arrays. They’re also relatively immune to errors due to vibration because there are no mechanical shutters or large mechanical cooling systems on-board.
In many machine vision applications, today’s InGaAs SWIR imagers are monitoring both physical and chemical properties of the product. This gives the process engineer more measures of product quality in fit, form and function.
Some of the greatest recent progress in SWIR has been in InGaAs cameras. First, linear arrays up to 1,024 pixels on a 25 m pitch provide higher resolution with a wider field of vision so fewer cameras cover more area. Second, today’s InGaAs SWIR cameras provide simultaneous selectable analog and digital output as a standard feature. Third, because they have no moving parts—no shutters, cooling fans or pumps—InGaAs cameras are growing smaller, lighter and much more economical than any IR or competing SWIR imagers.
The U.S. military recently received a demonstration of an InGaAs microcamera measuring only 5.9 × 2.8 × 1.7 cm—smaller than a Zippo lighter and weighing only 70 g with the lens. It’s an all solid-state camera; there are no moving parts to fail or need attention. Such compact cameras should be of interest for machine vision applications that require squeezing multiple cameras into small, hard-to-reach places. One example is a machine tool manufacturer that needs to fit a camera to monitor each of the three axes during the cutting operation. The operator needs to monitor both the dimensions of the work pieces and overheating, which makes this an ideal application for the SWIR microcamera.
For these reasons, a strong case can be made to use the SWIR waveband real-time quality assurance in your process environments. Virtually every object or event of interest reflects or emits some SWIR radiation, providing companies with a means to achieve the goal of all machine vision quality inspection projects: to find enough contrast to separate the good product from the bad—or literally, the wheat from the chaff. The result can be a more stable, reliable and lower-cost process provided by an InGaAs machine vision system, with a return on investment that will make the company controller your best friend.
Add new comment