Improvements in sensors and networking are often cited as reasons for the rise of robotics and the Internet of Things. Machine vision in particular is drawing support from robotics suppliers and investors.
Industrial imaging is about “reducing reliance on subjective human measurement,” said Steve Varga, principal scientist for imaging and instrumentation research and development at The Procter & Gamble Co.
During his keynote address at the AIA Vision Show in Boston earlier this month, Varga noted that as robotics matures, machine vision will use 3D imaging, motion analysis, and big data to build to true perception and understanding.
As an example of how machine vision is enabling advances, researchers at Carnegie Mellon University have found that a robotic hand can be more precise if a 3D camera mounted on it is used to develop a model of its environment and relative position.
CMU is also developing software to enable robots to visually sort through clutter, and Cambridge Consultants is using Microsoft’s Kinect sensor and algorithms to gather color and 3D data for picking and placing fruit.
Other speakers at the AIA Vision Show and International Collaborative Robots Workshop noted that advances in machine vision will improve robot safety and make it easier for them to learn tasks.
“Vision guidance will lead to new applications outside the factory, including in life sciences and self-driving cars,” Jeff Burnstein, president of the Robotic Industries Association, told Robotics Business Review. “It’s a matter of educating users.”
The machine vision market is expected to grow from $8.08 billion in 2015 to $12.5 billion by 2020, according to Markets and Markets. Transparency Market Research predicts the global industry will grow from $15.7 billion in 2014 to $28.5 billion in 2021.
While the figures differ widely, the consensus is still in favor of expansion. The latest batch of corporate transactions around machine vision hope to capitalize on that growth.
ARM spends $350 million on visual processing
Chip designer ARM Holdings PLC last week bought machine vision company Apical Ltd. for $350 million. Both companies produce widely used processing technology, and ARM expects that Apical’s systems will help it expand into new areas, including robotics and artificial intelligence. “Computer vision is in the early stages of development, and the world of devices powered by this exciting technology can only grow from here,” said Simon Segars, CEO of ARM. “The ARM partnership is solving the technical challenges of next-generation products such as driverless cars and sophisticated security systems.”
Apical’s Spirit translates data for machine vision.
London-based Apical has developed camera and display subsystems that are used in more than 1.5 billion smartphones and about 300 million other devices. It has about 100 employees.
Apical’s Spirit product uses dedicated silicon blocks to convert raw sensor data or video into machine-readable images. “Apical has led the way with new imaging technologies based on extensive research into human vision and visual processing,” said Michael Tusch, CEO and founder of Apical. “The products developed by Apical already enable cameras to understand their environment and to act on the most relevant information by employing intelligent processing.”
ARM designs and licenses semiconductors, including those used in laptops, smartphones, and robots. The Cambridge, U.K.-based company claims to be the world’s largest intellectual property supplier for chips. Its technology is used in 95 percent of smartphones, 80 percent of digital cameras, and 35 percent of all digital devices.
Like Intel, Qualcomm, and others, ARM is hoping that robotics and the Internet of Things (IoT) will compensate for slowing chip sales in the mature mobile computing market.
“It’s not only its mature products with an evolving road map,” ARM Chief Technology Officer Mike Muller said in reference to Apical. “It also enables us [to move] into a whole row of connected devices.” This week, Intel Corp. bought machine vision firm Itseez in San Francisco for an unspecified amount.
Kuang-Chi looks to eyeSight for robots, smart homes
Kuang-Chi has invested $20 million in eyeSight Technologies Ltd. and plans to apply its machine vision and gesture-recognition expertise to robotics, IoT, and automotive applications.
Herzliya, Israel-based eyeSight makes software and gesture-recognition technology for consumer electronics, including smartphones, tablets, and virtual reality headsets. These controls could be used in robotics and “smart homes.”
“Every device in our homes, from the TV to the cable box to the thermostat, has its own control system and remotes, cables, or apps for operation, resulting in lots of clutter and fragmentation,” said Gideon Shmuel, eyeSight CEO. “To solve this problem, we developed ‘singlecue,’ a control center to bring together all of these devices through a natural, easy-to-use interface that leverages our years of experience in gesture-recognition technology to let you experience your home through your fingertips.”
The company expects the deal to help it expand in international markets. EyeSight already has offices in Tokyo, Hong Kong, and Cupertino, Calif.
“We examined the top companies in computer vision, gesture control, and embedded deep learning, and came to the conclusion that eyeSight has developed one of the most advanced and versatile platforms,” said Ruopeng Liu, chairman of Kuang-Chi. “The company’s core technology is already incorporated into IoT systems and automotive platforms, and together, we will bring it to robotics and many other applications.”
Shenzhen, China-based Kuang-Chi was founded in 2010 and develops advanced technologies, including metamaterials, photonics, Wi-Fi, and space technologies. It has more than 3,000 patents worldwide.
Kuang-Chi’s interest in eyeSight is part of a recent trend of Chinese investments in foreign high-tech capabilities, particularly in Israel.
Norwegian IoT firm gets funding
Disruptive Technologies AS received 50 million Norwegian krone (NOK; $6 million) in equity financing from Ubon Partners. Disruptive Technologies is developing networked wireless sensors.
“In 2015, there were 25 billion devices connected to the Internet. According to Stanford University, it is expected to be more than 500 billion by 2022,” said Erik Fossum Faerevaag, founder and CEO of Disruptive Technologies. “We have a world-leading sensor technology and are happy to have Ubon Partners to help us to ensure that Disruptive Technologies becomes a great success.”
Disruptive Technologies has only 17 employees scattered across the world, but it expects to grow with the investment. Ubon Partners has offices in OIso, Cyprus, London, and Madrid.
IntegraOptics invests in automation, IoT
IntegraOptics Inc., which makes fiber optic equipment for major telecommunications companies, is investing more than $1 million to add robots to its warehouse in Albany, N.Y.
IntegraOptics expects to continue hiring even as it adopts pick-and-place robotic arms. It has doubled in staff to 60 people since 2013 and is growing internationally. The company supports Time Warner Cable and Verizon Communications and competes with Cisco Systems Inc. and Alcatel-Lucent SA.
In addition, IntegraOptics has partnered with nfrastructure Technologies Inc., an “Internet of Everything” networking integration firm in Clifton Park, N.Y., to explore IoT applications.
Italian and German machine vision firms merge
Earlier this year, Ambienta SGR SpA, an “environmental investments” company, bought Mikrotron GmbH, a high-speed camera provider. Milan-based Ambienta will combine Germany-based Mikrotron with Tattile SRL, an Italian machine vision company it acquired in 2012, to create LakeSight Technologies.
More on Machine Vision and Robotics:
It hopes to become the European leader in machine vision systems used for industrial automation and product inspection.
“Machine vision is a €2 billion [$2.24 billion], fast-growing market characterized by high fragmentation, where 90 percent of suppliers have a turnover below €10 million [$11.9 million],” the company said. “Ambienta’s end goal is to create a unique player with sales in the €40 million to €50 million [$45 million to $56 million] range and global sales reach.”
Mikrotron reported €9 million ($10 million) in revenues, but Ambienta didn’t disclose how much it paid for Mikrotron.
