Technology is key to post-COVID resiliency as companies use digital transformation, robotics, and computer vision to help increase safety for employees and customers alike.
Every industry has undergone unprecedented change and disruption over the last few months. Supply chains, working processes, customer behavior and the economy itself have had to transform almost overnight. As we get ‘back to normal’, that normal is likely to be very different from before.
It presents a unique opportunity for organizations to reinvent and enhance the way they operate. By embracing technology to support industry best practices post-COVID, it’s possible to build resilience and ensure long-term success.
Digital transformation using edge computing, robotics and computer vision offers huge potential. These technologies can help address existing challenges, such as the need to automate and accelerate decision making across the organization. But they are also ideal tools to support new best practices like remote working, maintaining social distancing and dealing with unpredictable labor shortages.
As each industry evaluates the adaptations it must make, these technologies should be a key consideration. Using the latest vision technologies, backed by powerful analytics and deep learning technologies, organizations can make better use of their data. They can use the latest analytics techniques to power innovative business models and use cases. Intel and its ecosystem work closely with companies across industries to implement such solutions, based on Intel® Vision Products. In this article, we’ll explore a range of existing use cases that can help increase business performance and resiliency.
While city and regional governments work to help their citizens return to their familiar places and routines, they must also add new layers of protection. They must enable social distancing and do what they can keep shared spaces as hygienic as possible. Computer vision solutions can help support these aims. For example, they can help track and control the number of people in a given space, automatically closing ticket barriers to limit crowding. When collated at a regional or national level, data on the volume of passengers using public transport may also help inform public health decisions.
Solutions like this are already available. For example, one AI solution provider built a solution that is used in railway stations to help increase efficiency and improve the passenger experience. Footage from security cameras is analyzed in real time to monitor foot traffic around the station, and identify abnormal behavior that the station staff should investigate. Internet of Things (IoT) sensors also enable the detection of fires or intrusions into restricted areas.
In addition to these safety benefits, the solution helps the station staff improve the customer experience. It’s possible to adjust staffing in real time, based on current footfall in the station. For example, by opening more ticket windows when wait time gets too long. Staff can also apply platform entry restrictions at busy times so that platforms don’t become too crowded.
There are many use cases for artificial intelligence (AI) and computer vision in manufacturing and industry. Indeed, it has been a central feature of many of the transformative efforts taken over the last years in the move to Industry 4.0. These applications will become even more essential moving forward. By bringing intelligence to equipment on the factory floor, it’s possible to automate processes, helping minimize the time humans need to spend in close proximity.
A real-world example of this type of solution in action can be seen in the case of a leading tire manufacturer. The company used an industrial PC solution with deep learning, computer vision and image processing to help improve its quality control process. It delivered 99.9 percent accuracy in defect detection and saw customer complaints drop by more than 10,000 annually.1
Computer vision has played an important role in healthcare for some time, for example by helping to accelerate analysis of complex medical images. In the post-COVID world, other uses of AI at the edge will also be important, such as combining it with robotics. This can help enhance patient monitoring and enable specialists to be ‘present’ for procedures or consultations, from a distance.
Robotics can also be used to help keep health facilities clean and safe for patients and staff, as demonstrated by the germ-killing robot Violet, developed by Akara Robotics. The robot uses computer vision to safely navigate its way around obstacles, including people. It is also able to blast UV-C light, which has been clinically proven to kill complex viruses. While UV-C is dangerous to humans, Violet can enter a room and sanitize its contents with without humans being present. In this way, the hospital can maintain a sterile and safe environment, without endangering the people using it.
The retail industry has already made great progress when it comes to digital transformation. It uses data and AI to help enhance the customer experience, optimize supply chains and make smarter decisions about store layout and product placement. Computer vision plays an important role in all these areas. For example, a camera on a digital sign can capture audience impressions and behavior in response to certain messages. Inventory systems can use computer vision to more accurately track what’s actually on the shelves.
In a world of social distancing, computer vision can be used to track customer volumes in a store and the way people move around the space. This can help retailers set out one-way routes around their stores that feel natural for the shopper while maintaining safe distances between individuals.
For inspiration here, we turn to a solution that uses computer vision to track and analyze in-store customer behavior and product trends. Computer vision and deep learning algorithms identify human shapes as they move around the space. This anonymized data provides insights into how the store is performing and how the customer experience could be improved. For example, heatmaps show how shoppers walk around and interact with the space. This helps retailers make more informed decisions about overall layout and the placement of specific merchandise.