Viewpoint: Emerging markets
Frank Grube, president and CEO of Allied Vision Technologies, expects a revolutionary change in the transport sector based on machine vision technologies
M
achine vision concepts are hardly new; AVT has been designing cutting edge systems for more than 25
years. However, few could have imagined the capabilities, applications and uses of machine vision we currently see today. Who would have thought even 10 years ago we would have the technology capable of capturing a pictorial sequence of street images for every freeway, arterial, road, lane and cul- de-sac, let alone the capability of making this available to the world in real-time? Yet, even though there has been a proliferation of technologies, including significant advances in machine vision, this is no longer considered exceptional. One emerging market segment on the cusp
of a major change due to the introduction of machine vision is the transportation industry, an industry which has long suffered from an inability to collect quantifiably accurate infrastructure level information. With
The current array of sensors has obvious advantages and disadvantages
the introduction of MV technologies, the transportation industry could collect data multiple times per second, providing a quantity and quality of data not only cheaper than a magnetic loop, but considerably more robust than any other current sensor technology. Integrate MV capabilities with Big Data philosophies, where 100 per cent of all transport data can be collected multiple times a second from every road, arterial and freeway, and it becomes clear that this industry is set to undergo a revolutionary change, one perhaps even bigger than the introduction of signalised intersections. Te current array of sensors has obvious advantages and
disadvantages. Te most significant collective disadvantage is their inability to provide robustness, forcing agencies to set-aside sizeable maintenance budgets to constantly recalibrate, reconfigure, and replace faulty equipment. Nonetheless, there has been a
widespread proliferation of sensor technology implementation in the more mature markets. Te real question, therefore, is not how – but where will the greatest benefits of machine vision within the transportation sector first be realised? Te answer is likely to be found in
traditionally considered undeveloped countries, unencumbered by decades of significant investment in traditional sensor technology. For these countries this presents an initial deployment as opposed to a system-wide replacement; countries in Asia, the Middle East, and South America, who are building new infrastructure could quickly leapfrog the capabilities of their western counterparts, by installing this lower cost and yet more capable infrastructure. MV technology will lend itself to establishing
complete network coverage, delivering high quality data and providing an integrated environment with a level of automation hitherto only dreamt of.
Michael Lund, EMEA sales director at JAI, notes colour and spectral imaging as growing trends for 2015
C
olour and spectral imaging are focus areas for JAI and the combination of faster cameras and narrowband LEDs
gives increased flexibility. Food inspection is benefitting from it because it’s a segment where investments in the single unit typically is low per throughput, and therefore the higher the throughput the higher the investment. Outdoor applications, in which intelligent
transportation systems play a significant role, have been steadily growing. Lately entertainment, such as sports imaging, has entered the segment and will gain momentum in the coming years. CMOS sensors are coming of age and many
of their previous shortcomings belong to the past. Te main obstacles were image quality and rolling shutter. Today’s sensors with improved image quality and global shutter, paired with speed and high resolution, are well in line with the goals of machine designers and manufacturing personnel. Leading machine
vision camera vendors are now offering cameras that can capture more than 10 megapixel images at more than 200 frames per second. Design and manufacturing engineers
are always scanning the horizon for new technologies that can make their next machine faster and better than the previous one. Obviously both higher resolution and speed go hand- in-hand with higher bandwidth and are driving interfaces like USB3 Vision and CoaXPress – soon we’ll see migration to 10GigE gain momentum. Higher speed, combined with lower prices,
was only used in analysis. Now it can be integrated in the manufacturing process. Ease of use and
The challenge for the machine vision camera vendors would be to increase in-camera processing
is pulling applications where quality inspection previously could only be performed by random sampling into 100 per cent in-line testing. Until recently, detailed time-resolved imaging was something that could only be done using expensive and complex cameras and therefore
22 Imaging and Machine Vision Europe • Yearbook 2014/2015
development of standards will continue; acquisition interfaces which are independent of specific transport layers and hardware are helping this. Machine vision camera vendors have brought technology in image acquisition to a level, where the amount of data has become very high. Naturally, demand from
data processing is increasing and part of it may have a knock-on effect on the camera. Te challenge for the machine vision camera vendors would be to increase in-camera processing to perform functions that can be carried out faster in the camera than externally in order to drive overall system performance forward.
@imveurope
www.imveurope.com
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 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61