Edge-based video analytic systems are significantly improving business operations across industries, says Schneider Electric. Video has, over the recent years, become an integral part of our lives.
Edge-based video analytic systems are significantly improving business operations across industries, says Schneider Electric. Video has, over the recent years, become an integral part of our lives. Going forward, even machines will make more use of videos to recognize objects and activities, and take action.
According to Jamie Bourassa, VP of Edge Computing for the IT Division of
Schneider Electric, emerging technologies that enable integrated video analytics (IVA) are impacting a broad set of applications across a wide variety of environments including buildings, hospitals, vehicles, and factories. In a recent post that was published on the company’s site, Bourassa said that with high-definition cameras that enable advanced analytics, real-time decision making becomes much more efficient.
“Examples abound of how IVA is impacting people’s lives,” Bourassa noted. “The US Department of Homeland Security uses multiple high definition cameras to perform facial recognition in airport security lines. High definition cameras capture images of the people who pass through from multiple angles. The cameras are strategically placed so that an image of a person’s face can be quickly assembled. That captured data is then compared to the facial data of persons of interest, and emergency security measures are then taken if required.”
Such technologies can also be seen deployed at several public places like museums and stadiums. A single major attractive feature in such contexts is that they ensure protection without causing inconvenience to people who are simply there to enjoy an event or the visit.
Popular in retail and manufacturing
The retail industry was one of the early adopters of IVA, with some high-end chains using the technology to recognize VIP customers as soon as they enter the store. Bourassa also gave an example of a pilot project from Amazon Go that uses IVA with cameras placed on ceilings. With input from indicators on a customer’s mobile phone, the system can recognize the products that are being selected from the aisle. These products are charged automatically, enabling the store to do away with checkout counters. Manufacturing sites are also making use of IVA to boost efficiency.
“In the food and beverage industry, for example, the amount of food wasted due to processing defects, breakage, or contamination has always been a cost driver,” Bourassa said. “Now, food items, such as potato chips, are filmed by high definition cameras as they make their way down the conveyor belt production line. The images reveal whether the potato chips are too dark (overcooked) or too salty. The system then sends instructions upstream to where the ingredients are being prepared or to the ovens and heaters so that proper adjustments can be made. Algorithms process video-based data, which drive adjustments to the automation of the manufacturing process, thus limiting waste and boosting output efficiency.”
Why edge matters
As the data collected by various sensors continue to increase, transmission to cloud-based centers could turn out to be a costly affair. This is when local processing with the help of micro data centers become important.
“In many cases, integrated video analytics applications are supported by local processing supported through micro data centers,” Bourassa said. “The software supporting such applications drives hardware requirements that then feed the specifications for the micro data center. IT server processing power and storage ideally should be bundled with power, cooling, rack, UPS, and monitoring solutions so that the integrated video analytics applications can run in a reliable, predictable, and safe manner.”
As video analytics on the edge continue to become more and more popular, businesses across the globe could find it easier to boost their product quality and safety. This will, in turn, lead to more satisfied customers and improved bottom lines.