SWIR IMAGING
not using SWIR because they cannot get a camera, so what we are hoping for is that because we can offer SWIR imaging at orders of magnitude lower price, then new applications will pop up.’ Malinowski mentioned agriculture as one
area that could benefit from the new SWIR sensors, to use on harvesting robots or for monitoring humidity or moisture. He also believes the sensors could be integrated into smartphone cameras or for applications such as augmented reality. ‘In inspection, they could be used for food or plastics sorting, and in surveillance for low-light cameras with better contrast,’ he said. ‘Additionally, by enabling feature distinction in bad weather or smoke conditions, one can envision firefighting applications and, in the future, advanced driver assistance systems.’ Te first generation of Imec’s quantum
dot sensor has a resolution of 758 x 512 pixels and a pixel pitch of 5µm. According to Malinowski, however, the second- generation chips, currently being tested, will have a pixel pitch as low as 1.8μm. He noted that the typical pixel pitch of an InGaAs sensor is between 15μm and 20μm. Despite the lower fabrication cost and
higher resolutions achievable with the new sensor technology, Malinowski said quantum efficiency – the performance achieved for the amount of light – will only be around 30 to 40 per cent; InGaAs sensors are able to offer 80 to 90 per cent quantum efficiency. He added: ‘I think that InGaAs will remain unbeatable in terms of high-end performance for the time being. Tis won’t be same story as CMOS versus CCD, however – there will always be space for both quantum dot and InGaAs sensor technologies.’ Imec aims to have its first wafer-level active NIR and SWIR sensors next year.
Complete cameras Quantum dot-based SWIR imaging technologies are also available from US-
based SWIR Vision Systems, which has been selling its Acuros colloidal quantum dot (CQD) VIS-SWIR cameras from Q3 2018. Te cameras are available in VGA (640 x 512 pixels), one-megapixel (1,280 x 1024 pixels), and full HD (1,920 x 1,080 pixels) formats, and as of recently have had a thermoelectric cooler added to the CQD sensors in order to improve their signal-to-noise ratio. ‘Demand for these cameras has been
increasing throughout 2019,’ said George Wildeman, CEO of SWIR Vision Systems, who remarked that the 1,920 x 1,080-pixel model, which has a resolution six times higher than the current standard 640 x 512-pixel InGaAs cameras, is the first of its kind to be commercially available. ‘Tere are a few high-resolution InGaAs cameras with 1,280 x 1,024-pixel sensor arrays, but these are very high cost,’ he said. ‘It is a big challenge to scale InGaAs cameras to larger array sizes without a large increase in their price point.’
‘It is a big challenge to scale InGaAs cameras to larger array sizes without a large increase in their price’
One of the primary markets taking
advantage of these higher resolution CQD cameras, according to Wildeman, is silicon chip and silicon wafer inspection, where manufacturers want to identify finer defects or cracks in silicon substrates. Silicon is transparent in the SWIR
wavelength band. For example, a bonded wafer pair can be backside illuminated with SWIR light to identify alignment marks and also inspect for void defects. Tis is not possible in the visible band where silicon is opaque. ‘Our full HD SWIR value proposition is definitely resonating here,’ Wildeman said.
The high-resolution SWIR images that can be captured with emerging quantum dot-based sensors can be used to identify defects in semiconductor chip production
Another application expected to benefit
from high-resolution SWIR cameras is maritime imaging, where they can be used to see through haze and other obscuring weather conditions. In addition, Wildeman expects this
technology to be of interest to laser beam profiling, optical alignment, hot glass inspection, liquid fill level measurement and art restoration applications. ‘Wood moisture detection in lumber facilities is also an emerging application for this technology, one that we weren’t originally anticipating,’ he added. Similar to Malinowski, Wildeman
believes that the high quantum efficiency of InGaAs SWIR cameras will maintain the technology’s dominance in high-end imaging applications, in addition to those involving a low number of photons – such as nighttime imaging. However, SWIR Visions Systems has so far found that the combination of its full HD sensor with LED- or halogen-based infrared lighting is able to deliver high-quality images for most machine vision applications. Te firm plans to move to smaller pixel
pitches in the future – its first product is designed on a 15µm pixel pitch, but the company has demonstrated sub-3µm pixel pitch in the lab. It will also work to increase quantum efficiency. ‘Te quantum efficiency for CQD sensors will not reach 70 per cent anytime soon, so we expect there will always be room for InGaAs cameras, especially in less cost-sensitive applications,’ Wildeman said. Finally, SWIR Vision Systems is working
on cameras with broader bandwidths. Its Acuros cameras already cover the entire visible and SWIR band with one sensor – from 400nm to 1,700nm. It will now extend the bandwidth out beyond 2µm, which, according to Wildeman, will open up new applications, for example in chemical detection, plastic sorting, hyperspectral imaging, methane detection and mineral detection.
A SWIR image taken of a shoreline, which can be seen clearly through low-visibility rain
www.imveurope.com | @imveurope
Graphene-enhanced detectors Te Acuros cameras will soon be joined on the market by another VIS-SWIR sensor
g DECEMBER 2019/JANUARY 2020 IMAGING AND MACHINE VISION EUROPE 11
SWIR Vision Systems
SWIR Vision Systems
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40
