Machine vision industry looking to 2017 as transition to positive growth

The machine vision industry looks to rebound after a tough 2016 and while there are promising signs on the horizon, some challenges both domestic and abroad still remain as companies look to make a major transition.

By Winn Hardin February 1, 2017

Coming off of a tough year in 2016, machine vision insiders are looking to 2017 with high hopes—and a pinch of trepidation—as several key industries enter transition phases.

In the first three months of 2016, sales of machine vision systems and components fell 11% year-over-year. By Q3, sales had rebounded by 7%, reducing overall contraction to 3%. With projected overall sales of $2.3 billion in 2016, the market is expected to contract by 1%, which is still a 10% improvement over the start of the year.

Alex Shikany, AIA’s director of market analysis, said the downturn in the first half of 2016 was mostly cyclical, influenced by tightening capital expenditure (CapEx) budgets, a soft global semiconductor market, and a slower manufacturing industry.

Some suppliers, however, reported weak performance throughout the entire year. "We heard from many people that had a lot of variability month to month, and a lot of bad months," said John Merva, vice president, North America at Gardasoft LLC. "That will bleed into the first-quarter, but we expect things will start to improve through the year. Everything happens slow in this market, but we will see growth in general."

While 2017 is the "Year of the Rooster" according to the Chinese calendar, the machine vision industry might view it as the year of the transition. For example, semiconductor is transitioning from traditional processors and memory 18-month cycles to a new focus on embedded, low-power processing elements that could "flatten" some of the semiconductor industry’s wild cycles. Automotive is shifting to autonomous, while machine vision migrates from the plant floor to nonmanufacturing markets such as security and transportation.

What impact will the new U.S. President have on the industry?

Even with a generally upbeat mood going into 2017, some vision companies say that with a new presidential administration in the U.S. comes questions about its impact on industry.

"We don’t know what to expect, but one thing we know about the market is that extended uncertainty can create volatility/bearishness," said Greg Hollows, director of machine vision solutions at Edmund Optics. "You can see this going to into any presidential election. We need to wait three to six months to see what Donald Trump starts doing in office to determine if there’s certainty in one direction or another."

"The U.S. could increase reshoring and boost manufacturing in general if the tax breaks promised by the Trump administration occur and leaves money in the pockets of businesses and will give us an uptick in the automation sector, including vision robotics," Merva said.

IoT readies for primetime

A handful of applications and technological trends are expected to influence the vision industry in this coming year of transition. Ghislain Beaupré, vice president of R&D and operations at ‎Teledyne Dalsa, pointed to continued and increased focus on robotics, 3D imaging, and machine learning, along with growing emphasis on software to power artificial intelligence (AI) and artificial neural networks (ANN). Additionally, Beaupré projected continued expansion of applications for non-visible imaging, including multispectral and thermal.

One trend that vision could capitalize on in 2017 and beyond is the Industrial Internet of Things (IIoT). "Machine vision is becoming more ubiquitous as the IIoT becomes more of a reality," said Marc Marini, director of vision R&D at National Instruments. He cited the example of infrared cameras for asset monitoring.

With spending on the IIoT expected to reach $500 billion by 2020, enterprises have high expectations to boost productivity and revenues. "They want connectivity from the top down to all the machinery and controls so they can guarantee a product change and download the correct parameters to all individual components," Merva said. "This interconnectivity will continue to proliferate, enabled by the GigE Vision and GenICam standards, which will continue to develop and become more useful to end users, integrators, and OEMs alike."

Even with the promised gains of the IIoT, the market might not yet be ripe for picking. "The leading edge would likely happen on the production of these IoT devices," said Wallace Latimer, sales director, customized optical systems, FISBA LLC. "The use of them in our space to update or change the industry is still to be determined."

Autonomous vehicles are going to be smarter

Demand for machine vision applications outside the factory is expected to climb as well in verticals such as intelligent transportation systems, surveillance, and autonomous vehicles. For self-driving cars, machine vision could offer distinctive solutions to an industry on the cusp of deployment.

"Deep learning or artificial intelligence (AI) platforms and penetration, as leveraged by great investment for autonomous driving, should continue to drive interest and expectation in solving challenges in the decision-making space of automation," Latimer said.

Machine vision’s place in the autonomous vehicle market has yet to solidify, according to Hollows. "These cars have had cameras for a long time, but very few machine vision companies sell into that market because of the commoditization of the product," Hollows said. "And when you’re integrating things in the automotive industry, it’s difficult."

He pointed to the cell phone market as an example. "Machine vision companies sell all kinds of things into inspection systems within the industry, but how many people are integrating components on device?" Hollows said. "Is the autonomous vehicle market going to create a boom of new technologies that need to be inspected and machine vision grows? Do we have the resources and capability, or is it the adjacent industries like lift trucks, semis, and public transportation trains and buses where the volumes are lower and parts are integrated with higher margins?"

If the answer to these questions is "yes," another question arises: Are vision OEMs and suppliers equipped for the task at hand? "2017 marks the beginning of the transition point because in five to six years we’re going to be seeing new players in the market and companies that will change radically," Hollows said.

The last time such a sea change occurred was between 2007 and 2009 when the recession hit. "We went from expensive smart and regular cameras to low-cost GigE Vision cameras, and it changed the industry," Hollows said.

Adapting to customer demand

Although opportunities in non-manufacturing market segments for vision and imaging are abundant, machine vision manufacturers confront several challenges. According to AIA’s Shikany, these industries, "Typically demand cameras with the ability to change focus and aperture while operating in differing ambient conditions."

On the other hand, traditional manufacturing markets generally prefer camera systems with a fixed focus, distance, and lighting. By overcoming camera limitations, vision solution providers continue to penetrate non-manufacturing industries such as life sciences, agriculture, traffic, security, and surveillance.

Regardless of the markets they target, vision companies face another key challenge: satiating customers’ appetite for higher-bandwidth applications. Most often that requires cameras with increased resolution and speed. "Customers expect more functionality for a lower price," Shikany said. "This mandate puts more pressure on vision manufacturers, distributors, and integrators to deliver feature-rich solutions at a modest cost."

With a number of irons in the fire, the machine vision industry is poised for growth through the rest of the decade. "Between a healthy manufacturing sector and mounting interest in automation, North America will continue to be an appealing business environment for machine vision sales in the next three to five years," Shikany said. 

Winn Hardin is contributing editor, AIA. This article originally appeared on the AIA website. The AIA is a part of the Association for Advancing Automation (A3). A3 is a CFE Media content partner. Edited by Chris Vavra, production editor, Control Engineering, CFE Media, cvavra@cfemedia.com.

ONLINE extra

See additional stories from A3 linked below.

Original content can be found at www.visiononline.org.


Author Bio: Winn Hardin is contributing editor for AIA.