Internet of Things

Harnessing Computer Vision To Enhance Quality Control

Internet of Things | Computer Vision | DataThings

There’s more to business than raw numbers. For the factories who produce the goods that we depend on, success requires not just hitting sales goals or shipping products on time. Enduring success depends on securing customer satisfaction and loyalty. That’s why the most successful companies offer consistent, quality products that their customers can rely on. As manufacturing marches forward with its newest industrial revolution, connected devices are being harnessed to introduce unprecedented agility to manufacturing processes.

Djamila Aouada (SnT)

 

But quality assurance and control methods need to keep pace in order for these flexibly manufactured products to continue exceeding customer expectations. Here in Luxembourg, DataThings is positioning itself as a leading global provider of the cutting-edge artificial intelligence (AI) and machine learning (ML) tools manufacturers need to do just that.

 

“Industry 4.0, the growth of the Internet of Things, and the growing importance of cyber-physical systems all mean that businesses are becoming more and more complex,” says Grégory Nain, Co-Founder of DataThings. “We take the raw data and transform it into actionable insights for our clients.” The company, which was founded in 2017 as a spin-off of SnT, helps its clients to monitor their infrastructures and production systems, allowing them to detect issues sooner, predict outcomes faster, and maintain maximally efficient operations.

 

“We’re very excited that thanks to our new research partnership with SnT, we will soon be adding visual data to the arsenal we work with,” says François Fouquet, Co-Founder and Chief Technology Officer at DataThings. The new partnership, which was made official in July 2020, will explore how computer vision technology can be adapted to support quality control in a wide number of industrial applications. The addition will open up new markets for the young data science company. First up for the DataThings team is applying their unique data processing technology to visual data for quality control. 

"This partnership will open up a new branch of technology for us, which will allow us to gain experience and find unique solutions for our clients.”

Grégory Nain, Co-Founder DataThings Tweet
François Fouquet, Grégory Nain and Joe Lorentz (DataThings)

“Using computer vision technologies for automatic quality control in industrial manufacturing is necessary to meet the demands of modern manufacturing, or the so-called Industry 4.0. This partnership project is unique because it will allow us to execute the research in context, giving us access to proprietary domain knowledge and data that we otherwise wouldn’t have,” says Prof. Djamila Aouada, principal investigator of the project at SnT.

"This partnership project is unique because it will allow us to execute the research in context, giving us access to proprietary domain knowledge and data that we otherwise wouldn’t have.”

Aouada and her team will work with DataThings to build computer vision capabilities into the GreyCat tool, an open-source platform that forms the backbone of the business. “GreyCat was actually the subject of our founders’ research back when they were doctoral researchers,” says Joe Lorentz, current doctoral researcher on the project. It is a toolset that analyses big data in real-time and provides rich graphical representations of its results. However, up to now GreyCat has only been able to process numerical data from sensors. Interfaced with a smartphone, it would yield information about things like vibration, temperature, or geolocation. Once equipped with the capability to process visual data as well, it will be able to automatically identify flaws in products on manufacturing lines, as well as support operators in analysing the causes, to sustainably solve the issue.

Research team working with Datathings

“Since launching our start-up we have experienced first-hand how research can inform solutions that give a competitive advantage to our clients. This partnership will open up a new branch of technology for us, which will allow us to gain experience and find unique solutions for our clients,” says Grégory Nain. “SnT is a unique family of innovators and we have never lost contact with those we did our research with while we were there, including Prof. Aouada. It means a lot to our entire team to be ‘coming home’ now as a partner.”

People & Partners in this Project​

Djamila Aouada
Oyebade Oyedotun
Joe Lorentz
DataThings