Join or Sign in

Register for your free membership or if you are already a member,
sign in using your preferred method below.

To check your latest product inquiries, manage newsletter preference, update personal / company profile, or download member-exclusive reports, log in to your account now!
Login asmag.comMember Registration[%]26S&utm_medium=topbanner&utm_campaign=WN7_SEAcampaign

Unleashing the potential of Industry 4.0 with machine vision

Unleashing the potential of Industry 4.0 with machine vision

Machine vision can do more than just read barcodes or find defects on the production line. It can also read text, organize in-trucking for logistics and allow humans and robots to work and interact safely in the same areas.

Traditionally, machine vision in manufacturing was used for tasks like barcode reading and defect detection. However, “with the advent of technologically-advanced, low-power machine vision chips, new use cases are appearing in the manufacturing area,” said Jerome Gigot, Senior Director of Marketing at Ambarella.

Warehouse management, Gigot said, was a great example of where the latest machine vision technologies were being used to enhance operational efficiency and safety, as well as ensure manufacturing quality control.

Previously, a barcode reader scanned a package, then a human worker or robot used the collected data to carry the package between different points within the warehouse or to a loading area.

“Both methods have limitations,” said Gigot. “Due to safety concerns, robots do not typically roam in the same area as humans -- ‘robot zones’ and ‘human zones’ tend to be separate.”

With the latest progress in machine vision, robots powered by 3D vision and neural network processing are able to move around freely and safely in the same work areas as humans. Such systems can distinguish a person from an inanimate object, as well as take actions to maintain safe distances and avoid collisions.

“Collaborative robot and autonomous mobile robot on the factory floor now have machine vision capabilities to help them identify objects and obstacles, pick up objects and help them to navigate in unstructured environment,” said Lian Jye Su, Principal Analyst at ABI Research.

“Ultimately, manufacturers will benefit from either or [both of] these solutions as they can push carts and deliver parts within or between the factories, optimizing workflows, minimizing workplace hazards, and freeing up valuable staff resources,” said Su.

Gigot offers the example of machine vision robots loading packages inside a truck. “With the latest advances in machine vision, a robot will be able to perform a 3D scan of the inside of a shipping truck, and then decide on the optimal placement for each package.”

Such robots are able to scan packages that are already in the truck, and find and decrypt barcodes. It can then decide whether to organize packages by final destination or to load-balance the vehicle.

More possibilities

Other areas, such as product and component assembly, can also be enhanced using machine vision.  

The ultimate goal of manufacturing is to ensure products and components coming from the production line meet quality, safety and production guidelines. For packaging operations, machine vision can provide a 360-degree view to make sure products are placed in the right position for packaging in terms of cap closure, position, label and print quality.

Outside of barcodes or a GTIN (Global Trade Item Number), a packaged product often has descriptive text on it. When this printed text can be read by machines, the system can further check against a database to ensure the printed labels are valid. If the product code doesn’t match with the text, the package will be rejected on the production line.

“The increasing number of stock keeping units (SKUs) and short product life cycles necessitate the deployment of industrial solutions that can be automate and augment different manufacturing processes. As such, one of the most important elements in the factory floor is to be able to recognize objects, something that only humans could do well until machine vision became more advanced,” said Su.

Share to:
Comments ( 0 )