Safeguarding workers and quality with Accenture and Coral

New industrial applications from Accenture leverage Coral intelligence to do more

ic_share_24px Created with Sketch.
Close Icon Created with Sketch.

An automotive component supplier in the U.S. autonomously monitors safety and productivity on its assembly lines. A life sciences company in France sorts thousands of Petri dishes an hour to speed drug discovery. And a French manufacturer checks the integrity of food containers as they speed down the production line.

Edgel AI from Coral implemented by Accenture makes it all possible.

Multinational consultancy, Accenture, has been helping corporate clients realize the benefits of cutting-edge technology for 70 years. Now, through its partnership with Google, the company is working to bring AI-powered visual inspection capabilities to factories, utilities, and other industrial settings.

Such settings often require enhanced privacy and security to protect intellectual property and workers. That’s why Accenture has turned to Coral’s edge AI capabilities. For example, Coral lets factory floor managers analyze video locally instead of streaming it to the cloud for AI processing, reducing data transfer costs while keeping potentially sensitive data on-site.

“The idea is to share just segmented data on the cloud for your applications,” says Sudhindra K. Ghanathe, Global Industry Innovations Lead for the Accenture Google Business Group. “That’s exactly the breakthrough Coral allows.”

Initial proofs of concept in manufacturing plants in North America and Europe have already demonstrated positive results, setting the stage for more deployments to come, for example, at the automotive component manufacturer.

Visual inspection without compromise

The automotive manufacturer wanted to check workers for safety equipment such as hardhats and safety vests without putting personal identifying information on the cloud or requiring supervisors to continually watch monitors.

Conventional cloud processing was too slow and expensive. “When you capture video of processes or monitor for safety issues, you need high bandwidth when uploading to the cloud — especially when running encryption for privacy,” Sudhindra explains. “You are also taking data about people outside the premises and slowing response time.”

The Accenture team built an object tracking solution that keeps tabs on worker locations using a camera with a Coral Mini PCIe Accelerator on board mounted above the production line. A worker without proper safety gear triggers an alert for a manager to check on the situation. Data doesn’t leave the building, and the system doesn’t identify workers.

In a separate use case, the French consumer packaged goods (CPG) manufacturer wanted to check the quality of food containers in real-time without the burdens of high data transfer and slower response times that can come with cloud-based processing..

The company added a thermal camera to the Accenture-provided, AI-powered tracker on their assembly line to help catch defects in the containers. “They wanted to check for welding defects at the bottom of tubes of whipped cream,” Sudhindra explains. “And also if there were any cracks in the cap or in the tube itself.”

In these and other cases, only Coral could provide the right mix of affordability and advanced ML processing at the edge, according to Sudhindra.

The Accenture and Google Partnership

The Accenture team works closely with its customers to source the right sensors, processors, and other components for its AI solutions.

For situations where bandwidth is constrained, privacy is paramount, or processing speed is of the essence, Accenture brings Coral to the table — a natural extension of Accenture’s ongoing work with Google.

The Accenture Google Business Group is a partnership between Accenture and Google focused about 80% on the Google Cloud Platform, according to Ghanathe. “I lead the industry innovation practice. What that means is we take Google’s technology and create industry solutions.”

For local AI, the Accenture team always recommends Coral to clients. “The major advantage with Google’s Edge TPU is that it’s cost-effective,” Sudhindra says. “The second one is they’re open-source, so they have a robust developer community.”

What’s Next

So far, the Accenture team is still helping its customers test the waters with the relatively new capabilities Coral brings.

The next step will be to take these proofs of concepts into production. And while the COVID-19 crisis slowed the process, Sudhindra says he’s looking forward to seeing how Accenture and Coral can help provide solutions on the other side.

“Wherever you need quite advanced models to run on the edge, definitely Edge TPU-based solutions are worth it,” Sudhindra says of the processor at the heart of Coral hardware. “For the client, it’s a great value.”

If you’re interested in how Coral can help you build your next product, please contact us. If you want to learn more about the Accenture Google Cloud Business Group please visit accenture.com/google.