Featured Article Excerpt from AIA Vision Online
In devising comprehensive solutions for complex vision and imaging applications, system integrators see first-hand the pressures confronting their customers across a variety of industries. One user may need 3D vision to measure the volume of a food item, while another requires component traceability for automotive air bags.
Above all, customers rely heavily on system integrators as experts. As Starke Farley, senior sales engineer for Integro Technologies (Salisbury, North Carolina) puts it, “Customers typically come to us when they have a vision application that nobody has been able to solve.”
That means system integrators must be at the forefront of in-demand machine vision applications and technologies, as well as training and support of both the customer and vision professional. This makes integrators great bellwethers for the machine vision industry, including where markets are going today and tomorrow.
3D Vision Grows Up
Across many market segments, inspecting for quality remains a top driver for adopting machine vision systems. However, today customers are looking for more dimensional data to make better automated decisions.
As the technology behind 3D vision has improved, so too have adoption rates. “We were working in 3D vision 10 years ago, and we jumped off the end of the dock into some deep water,” says Brian Durand, president of i4 Solutions (St. Paul, Minnesota). “It was a lot to implement 3D vision successfully. We had to write our own software not only to communicate with cameras but to calibrate them from pixels into world units, as well as develop our own physical calibration tools.” Today, new offerings from camera and sensor suppliers “have made it easier for everyone to integrate 3D,” Durand adds.
Much of the demand for 3D vision has come from the food industry, Durand said. To illustrate his point, he explained a recent project designed to inspect a food kit that comprises several items. In addition to confirming presence of all items, “perhaps there’s a cup in the kit that’s filled with sauce, and 3D vision allows us to verify the correct amount is there,” Durand says. “You could try to weigh the entire kit but you don’t know if the right amount of cheese or pudding is there, or whether the shape of a snack bar is correct.”
The automotive sector continues to find new uses for 3D vision as well. In the last six months alone, Integro Technologies has fielded requests from four different tire manufacturers for a 3D inspection system. “Customers are looking for 100% inspection of the tire sidewalls for defects and thread inspection,” says Integro’s Farley. “Automotive suppliers also want 3D character recognition on parts that have been cast or dot peened.”
Integro holds a patent on a tire inspection system that reads the code required by the U.S. Department of Transportation. “Traditionally, that’s been a very difficult application because of the black-on-black on the sidewall of the tire,” Farley says. “You get very little contrast with a 2D vision system, so you have to use 3D.”
However, plenty of opportunities exist for 2D vision in automotive. One area of expansion is component and part traceability. “More and more components down to the smallest level are being marked and tracked all the way through the assembly process,” says Steve Wardell, director of imaging at ATS Automation(Cambridge, Ontario). “That opens up a lot of ID reading, OCR and track-and-trace type applications.”
Industry 4.0 Comes Knocking
The increased connectivity, automation and digitization of machines is resulting in sophisticated data collection and analytics capabilities. In manufacturing and other industrial sectors, this phenomenon goes by many names, among them Industrial Internet of Things (IIoT), Industry 4.0, Smart Manufacturing, and the Connected Enterprise.
As more executives consider and eventually implement IIoT practices and strategies, machine vision integrators are formulating ways to help customers achieve the cost and efficiency gains promised by a digital enterprise. The collection of data offered by machine vision “gives the manufacturer information in order to make decisions on what areas are working well and what areas they can improve upon,” says Nick Tebeau, manager of the Vision Solutions product group at LEONI Engineering Products & Services, Inc. (Lake Orion, Michigan). “If we are educating them on anything, it’s the nuances of how to make Industry 4.0 an effective and real solution.”
The biggest nuance, according to Tebeau, is how to control data. “Some people don’t understand what’s involved in taking information from a localized device and have it be collected and sorted automatically with a large scale of devices,” he says. To that end, LEONI has developed a platform that allows customers to work with their current IT infrastructure.
Sometimes considered an enabling technology of IoT, machine learning allows computers to derive insights from existing data without being explicitly programmed to do so. “The ability to use more learning type of applications or algorithms in a real-time setting has opened up certain quality inspections that were cost prohibitive or not very functional in the past,” says ATS’ Wardell.
ATS has developed toolsets and products delivered with their vision systems that collect and analyze data, which produces information on “how well the system is working and see a problem before it happens,” Wardell says. “Customers can make use of more advanced analytics algorithms that are available to deal with large sums of data from various places sources.”
Know the Customer, Know Thyself
System integrators consider customer training and ongoing support essential components of the services they provide. ATS recently developed a multipart automation training course known as ATS University, which teaches various aspects of automation at the operator, maintenance and engineering levels “so that people are more apt to maintain the systems we provide,” Wardell says.
In tandem, ATS is providing certification for large-scale customers who go through more expansive automation training.
