No, there’s not a typo in the headline. Machine vision – an “industrial automation” technology best known for inspection – is indeed being used for track and trace-type applications in non-industrial settings such as fast food, sports, and more. Although, I can understand why you might have thought I meant to say “RFID” instead of machine vision.
Typically, when we talk about how best to track and trace anything, RFID is the star of the show. However, as you’ll learn when you tune into the latest episode of Industrial Automation Insider podcast below, there are some things better tracked using machine vision (or computer vision, CCTV, etc.).
Poultry feet are among them.
Yes, once again you read that right. Machine vision technology is being used by a very popular global fast-food chain to track the health of the chickens they source for your favorite nuggets and sandwiches. The AI powering the machine vision system’s cameras has been trained to look for lesions on the bottom of poultry feet, which could be indicative of inhumane living conditions that consequently impact the birds’ health. They hold their suppliers to a high standard and feel machine vison is critical to monitoring SLAs and maintaining accountability. Fascinating, right? (It’s also reassuring that fast food companies really do prioritize quality of life for our food and quality ingredients like they say they do.)
Something else that machine vision is good at is tracking balls. Volleyballs, tennis balls, squash balls, you name it. Oh, and motorsport competitors. Machine vision cameras are becoming the go-to technology tools for tracking when race car drivers or bikers cross the line, when balls hit a line, and so on. They’ve successfully mediated many a challenge in live events. Cool, isn’t it?
Who would have thought that machine vision was so versatile?
I’ve been in the industry 20+ years and was admittedly surprised when I learned about the many creative ways this industrial automation technology is being applied outside traditional industrial environments such as manufacturing and warehousing. However, I am equally exhilarated that there are people like my guest today who have spent the last 20+ years making non-traditional machine vision applications fairly commonplace. This technology is far from “old school” or a “one-trick pony” like some people may try to classify it. This underdog technology is helping people do things not possible with RFID or other AI technologies. It’s helping us see the little things – those moments in time that can’t be repeated – with greater clarity. It’s also helping us build confidence in time-sensitive decisions that could have serious consequences if misjudged.
That’s why I appreciate what Marcin Kowalski does every day, and what he shared on a recent day when we sat down to record this latest Industrial Automation Insider podcast episode.
Marcin may be a product manager for Avicon Vision Systems, a Zebra Machine Vision partner, but his job isn’t really to manage product roadmaps or guide product development. It’s to understand what people want to accomplish and then figure out a way that technology can be applied to make their dreams possible. To do that, though, he must first learn to speak their language because he knows that they most likely don’t speak his.
For example, no one in Poland’s premier volleyball leagues would have called him and said, “We want a machine vision system or a CCTV system that can help line judges review plays and validate they made the right call.” They would call and say, “We need a way to better review plays and validate that the right call was made when a team challenges the call.” It is Marcin’s job to then tell them how technology can help them accomplish their goal.
But you’ll probably never hear him say “machine vision is the better approach than CCTV” or vice versa. That’s because the best way to accomplish your goals is to think about technology as a team sport.
What does he mean by that?
You’ll have to tune into our 30-minute discussion to find out: