Interesting that technology that started out in medical equipment and made its way to manufacturing is now being tapped to improve the quality of manufacturing that equipment. Another great example of how technology travels full circle. Given the amount of imaging that's utilized in medical equipment, it stands to reason there's much more opportunity to apply machine vision equipment for garnering efficiencies and working out quality kinks on the production floor.
It makes sense that medical would be a great growth area for this technology, given the fact that handling is an issue for many medical parts. With cost coming down and electronic performance rising, though, it's natural that it would find new applications in a variety of other industries, such as aerospace and defense.
Although several of the vision technologies mentioned in the article started in the medical industry, the origin of machine vision in inspection began in electronics. As the electronics content in other industries has risen, the need for more and better inspection has gone up. That's also happened as the need for higher quality of the end product has risen, even when electronics aren't a major part of the end product, such as consumer food containers.
I visited a production line yesterday at a plant that does a lot of precision assembly using adhesives and laminates, and machine vision is utilized heavily to ensure quality (check tolerances, etc.) I was particularly struck by how robust the MV equipment has to be to handle the production rate, temp, vibration, etc. A tall order for such precision equipment.
Yes, machine vision is extremely rugged hardware compared to even consumer equipment, which is one of several reasons it's always been a lot more expensive. That's started to change recently with the use of more open platforms, but it's still got to be highly durable.
Machine vision has come a long way in both quality and price. With off the shelf components and Windows based software, the abiilty to include vision on most products as a quality check has never been more accessible.
Machine vision is becoming so ubiquitous in so many different types of products that a new organization, the Embedded Vision Alliance, formed recently to help unite some of these far-flung industries and development silos:
Unlike previous vision trade associations, it's not limited either by industry or geography.
might be a prime candidate for integrated machine vision. The vision components would have to be extremely small to fit on a heart-crawling robot like this one, but cameras are getting tinier all the time. And the integration of machine vision with robots is definitely a growing trend on the factory floor. Seeing them in surgery may not be far behind.
You've definitely got a point there about OTS machine vision components. They've become much more prevalent since vendors have begun designing them using OTS chips and open-source or open-standards software, such as Windows and Linux. They've also become smaller and cheaper.
OTS modules are really reducing developement time. This will really act as a catalyst for innovation as many iterations and solutions are possible. machine vision will soar to new heights in the coming days..!
vumalkarp, thanks for the additional info re the vision-enhanced Da Vinci surgical robot, and the link to Given Imaging. I think vision-enhanced surgical robots make a lot of sense, just as they do in the factory for assembly, fabrication, welding, and stocking jobs.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.