People still look on machines with suspicion. But is it justified? We don’t think so!
Not so long ago, people were deeply suspicious of mass production. How could an automated process possibly be as good as a human at creating things? Although this perception in factory production has been squashed, some aspects of this suspicion have shifted to machine vision, which is still a relatively new phenomenon. People tend to think that machine eyes can’t do the job as well as human eyes.
Why does this view persist?
A lot of people have had bad experiences with machine vision: early adopters of complex technology found it expensive and difficult to set up, and if it isn’t set up properly, machine vision is unreliable.
Unlike simpler sensors such as inductive sensors, you can’t just place a vision system on the production line and expect it to work effectively. These days, products have improved, technology has stabilised, methodology is structured, and the people who do it for a living have had years of experience making the vision systems work.
The mistrust arises when organisations buy vision systems direct from manufacturers, but lack the experience required to implement them correctly – and when they don’t work as they should the system gets the blame.
So while the vision system is still not as simple as some other switchgear or decision-making processes, it is stable and supremely reliable if it’s done properly.
What’s the alternative?
It used to be the case that checking was carried out by sampling: human employees would take, say, five items off a line every hour to check them over. This is off-line testing. When products are rushing past at 300 items per minute, that’s a checking percentage of just 0.03% – that’s pretty low, and leaves a massive margin for defective items to slip through.
Machine vision will accomplish 100% testing which, in industries such as pharmaceuticals, is a legal requirement (as well as being a moral responsibility). Batch codes, date codes, and confirmation of inspection are crucial.
Other companies have a brand to protect, so any percentage of their products leaving the factory faulty is unacceptable. And badly-printed logos can damage brand identity too. Not to mention the waste of money for scrapping defective items.
What can machine vision do that humans can’t?
Following are just a few of the reasons why machine vision is a vital tool in modern manufacturing (for inspection, testing and robot guidance).
- 100% monitoring: when products are whizzing by at high speeds, and stringent accuracy is required, the only practical way to achieve such goals is with automated machine vision systems.
- Human fallibility: humans are subject to lapses of attention and inconsistency. Machine vision systems don’t get distracted, and they apply the same standards of inspection to each and every part they come across.
- Multiple views of objects: several cameras connected to the same system can be positioned on a line to look at objects from different angles. Humans are unable to do this; each individual person sees things slightly differently.
- Machine vision enables non-contact measurement for items that are fragile or too small to handle.
- There are many defects that are invisible to the naked eye. Machine vision systems will detect these flaws, which is absolutely vital in safety-critical parts.
- Accuracy and consistency: machine vision will achieve far higher accuracy and consistency than human inspection.
- Efficiency: as previously alluded to, machine vision systems can inspect hundreds of products per minute on production lines, massively reducing the time and resources required to carry out inspection.
- Machine vision helps to reduce paperwork, too, with data and records going straight into databases. With no need for clipboards and tick-boxes, organisations save time, money and resources.
Machine vision systems are customisable, reliable, accurate and adaptable – as long as they are implemented by experienced engineers. So if you’ve been unconvinced up to now, why not take a closer look at what they can do?