Machine Vision: Is It a No Brainer or Too Good to Be True?
If you used machine vision years ago, you may be hesitant to go there again. But modern AI-based systems are giving many manufacturers and warehouse operators significant competitive advantages. Here’s how.
The advent of new technologies such as internet of things (IoT), machine learning and artificial intelligence (AI) have been the source of much speculation and debate among business leaders and front-line workers alike. These new innovations have been heralded as key parts of the ‘fourth industrial revolution’, or Industry 4.0, with the potential to have transformative impacts across the board.
But it can be difficult, at times, to disentangle the hype – which has a tendency to take on a life all of its own – from the reality. This has left many business leaders, especially those running small and medium-sized enterprises (SMEs), wondering just what all this is likely to mean for them, in concrete terms.
In my opinion, one of the standout innovations of the past few years, machine vision, is an important aspect of the new wave of industrial innovation for the manufacturing sector, and one that could have a far-reaching impact for businesses of all sizes.
But how significant is it likely to be for smaller businesses? Is it just blue sky thinking? And haven’t manufacturers ‘been there, done that’ with machine vision, which is technically a decades-old technology?
If you haven’t experienced machine vision technology recently, or ever, it involves the use of cameras, AI-powered software, and a computer capable of running the high-powered system to identify and process images of products, packaging and more. The machine vision system then analyses the data it has captured using those images to automatically make a decision about what to do next with that product or package coming down the line, only calling a person in for help when a quality issue is identified and a second opinion is needed.
That decision being made might be very simple, for example checking whether a label has been properly applied to a product, and flagging to an operator when an error is detected. Or it could be a highly sophisticated system that can identify missing components on a printed circuit board which triggers a robotic arm to remove the defective component from production. The scope of use is truly enormous. (You can hear about some very interesting applications in episodes of the Industrial Automation Insider podcast, such as this one.)
In fact, despite machine vision’s impact just now being given the public attention it deserves, this technology has long been used in a wide variety of industries, including the manufacturing, pharmaceutical and fast-moving consumer goods (FMCG) sectors.
Machine vision has existed in theoretical forms since the 1950s and was first used in practice in the 1970s, but its use has expanded dramatically in recent decades as technology has become cheaper and exponentially more powerful.
It is already used for visual inspection, material inspection, process control, monitoring expiration dates, tracking product serial numbers at different stages of the supply chain, and gauging measurements in manufacturing. And there are new use cases – very clever and creative use cases – emerging every day as we often hear about on Zebra’s Industrial Automation Insider podcast.
That’s because machine vision enables business leaders like you to drive up quality not only by identifying defective parts or products but also highlighting where problems exist on the production line, which helps with predictive maintenance alerting.
In short, machine vision systems have a huge amount to offer across a multitude of applications. So, surely, it’s a no brainer?
Absolutely.
But – and it’s a big but – for many smaller business leaders, this level of automation seems wildly out of reach and verging on sci-fi. I get it. Historically, machine vision systems came with a hefty price tag. Not just for the equipment, but more importantly paying for the setup, which could take days or even weeks.
That has created a mindset which confines machine vision as simply too costly and too complicated to seriously consider. Particularly in the smaller and more traditional manufacturing base of which there are many in the UK.
It’s akin to what we’ve seen unfold around RFID where owners and executives of smaller businesses long felt the technology was out of reach because of the cost and complexity. More recent advancements in both RFID and machine vision have made the technology more accessible – and relevant – to smaller business operations and budgets. The barriers to entry have disappeared.
That’s why, as a barcoding solutions specialist and Zebra machine vision partner, my team and I at Cobalt Systems are working hard to rectify the assumption that – like RFID – machine vision is only for the big companies with big budgets and lots of resources. It’s for anyone who wants to remain compliant and competitive in today’s market.
We also want to prove that machine vision isn’t the complex technology system it used to be.
When we get in front of customers and can demonstrate the capabilities of the Zebra VS range of cameras and the speed with which it’s possible to set them up alongside Zebra Aurora software, there’s almost a sense of disbelief.
“Is it really that simple?... and it’s how much?!”
In many ways, it’s a very understandable reaction. The speed with which machine vision technology has become readily available in terms of both price and usability for SMEs is staggering.
Inevitably the more people (like you) that see the new breed of Zebra machine vision solutions, the more will get on board and the more mainstream the technology will become in the SME space.
And with projections for the global machine vision market to reach $USD 18.4bn (£14.5bn) by 2028, in this case, the hype looks firmly set to become reality and an exciting and lucrative one at that.
For those of us on the inside looking out, machine vision is definitely a no brainer, and it’s up to us to show the skeptics and traditionalists that’s it’s not too good to be true. In fact, it’s better.
Bring it on.
###
Editor’s Note:
Stay tuned to upcoming episodes of the Industrial Automation Insider to hear more from the Cobalt Systems solution architects about how machine vision and fixed industrial scanning technologies are being used to make SMEs (and all companies) more efficient, compliant, and competitive. You can also catch up on past discussions with experts about the impact machine vision can have on your business:
