How Zebra Technologies Uses Machine Vision to Transform Manufacturing Automation

Machine vision is in the spotlight Machine designTakeover Topic Event (August 12–16, 2024). Covering the entire gamut of computer vision, the show features case stories, trends, and interviews with notable players to watch in the space.

In the accompanying video, Andrew Zosel, senior vice president and general manager of Zebra Technologies, explains the vision and strategy behind the recent acquisitions and resulting solutions. Machine vision is a significant part of this investment and future growth at Zebra, Zosel said.

(In Parts 2 and 3 of this interview series focused on machine vision, Zosel explores the following questions: Given technological advances, do we still distinguish between machine vision and computer vision? How are the economics of machine vision changing? What is driving the ubiquity of machine vision?)

READ MORE: 10 Things to Consider When Designing a Machine Vision System

Research commissioned by Zebra found that 70% of survey respondents plan to implement computer vision in the next five years. The 2024 Manufacturing Vision Study, published in June 2024, found that 86% of manufacturing leaders said they struggle to keep up with technology innovations and securely integrate devices, sensors and technologies into their facilities and supply chains. Survey respondents include 1,200 CEOs, as well as IT and OT decision-makers across manufacturing sectors.

During a stop at the Zebra Technologies booth at Automate 2024, Machine designn I learned from Zosel how much Zebra is leaning toward building solutions that address such gaps. The variety and number of technology applications on display revealed the pace at which Zebra was filling its automation ecosystem as part of its efforts to become a full-solutions company that can orchestrate what manufacturers can’t do alone.

Evolving Zebra Technology: Growing in Scope and Scale

Those familiar with Zebra Technologies are likely to recognize the company for its industrial printing capabilities, including scanning, tracking and tracing, and mobile computing and software solutions. “Our industrial printers are used all over the world,” said Andrew Zosel, senior vice president and general manager of Zebra Technologies. “The majority of the labels that are put on boxes around the world, identifying and giving items a digital voice, are printed on Zebra printers because they are incredibly reliable and durable. That’s where the original Zebra brand came from.”

In 2014, the $3.45 billion acquisition of Motorola Solutions’ Enterprise business, which includes handheld scanners and mobile computers, transformed the scope and scale of Zebra’s offering, then expanded it to include asset visibility, tracking, barcode scanning and printing.

Another transformational year was 2021. Zebra acquired Fetch Robotics, Adaptive Vision and Antuit AI this year, effectively launching its machine vision and continuous industrial scanning portfolio. 2022 saw the acquisition of Matrox Imaging, a manufacturer of frame capture cards, vision controllers and imaging software.

“We’ve been investing significantly, both organically, leveraging our own internal solutions developed by our existing teams, in new camera and vision systems and machine vision, fixed industrial scanning products, but we’ve also made some significant acquisitions over the last few years in the vision sector, including Adaptive Vision from Europe, as well as a company called Matrox from Canada.”

Combining Hardware and Software to Deliver Machine Vision Solutions

Although the company is relatively new to the computer vision and image processing market, combining advances in robotics, image processing, automation and digital decision-making has proven to be a sure-fire way to simultaneously increase stealth and agility.

“With these acquisitions, the biggest and most important part is still the software, the algorithms and the capabilities there,” Zosel said. “That’s where we’ve been focused on developing and acquiring, especially the traditional machine vision capabilities, like deterministic algorithms or whatever you want to call them. And a lot of developments in AI, specifically deep learning.”

Zosel pointed out that there are many ways to make vision systems work. The simplest method might be to use a vision sensor or smart camera, which can include a fully integrated lens, light, and all-in-one computer in a single product.

“We offer these types of products all the way up to completely distributed systems where you have multiple parts coming together, like a frame grabber on a high-end computer with a separate set of high-end cameras with separate lighting,” Zosel said. “It’s the same basic architecture of lighting, lenses, processing, and compute. But it depends on both the integration and the performance requirements of the application.”

From Zebra’s perspective, Zosel said, the intention is to offer all of the above. “The advantage of the Zosel platform is that it has the scalability to offer everything from basic, stripped-down, lug-and-play products all the way up to very complex, advanced systems that do either extremely high resolution or extremely fast processing or both,” he said.

Long-term strategies: automation and expansion in production

Zosel said Zebra first and foremost wants to help its customers find ways to solve the challenges they face.

Sustainable packaging is an example of this. Some of Zebra’s customers are moving away from plastics, opting for thinner plastics and paper packaging. Zebra is helping them with inspections and helping customers evolve toward sustainability by providing scalable products and platforms, Zosel explained. For example, a customer might start with 2D machine vision but “plan to scale to 3D and use more advanced AI. A lot of our systems and efforts are focused on building that scalability and providing that path for our customers,” Zosel explained.

Zebra’s study found that manufacturing leaders plan to implement a variety of automation technologies over the next five years, including robotics (65%), machine vision (66%), radio frequency identification (RFID) (66%) and fixed-line industrial scanners (57%). The study highlighted that adoption of these automation solutions is being driven by several factors: the need to assign high-value tasks to workers (70%), achieving service level agreements (69%) and increasing flexibility on the shop floor (64%).

Zosel said Zebra is looking at ways to make products “easier to implement, easier to integrate with robotics, easier to automate and integrate with material handling systems, and developing clever ways to use deep learning, artificial intelligence and various imaging technologies to help customers with their applications.

Watch additional parts of this interview serieses with Andrew Zosel:

Part 2: Deep Learning Algorithms Help Vision Systems See Better

Part 3: How Deep Learning Complements Machine Vision Solutions