Skip to main content

HEALTH

SECO at embedded world 2023

Computer vision is enabling new, value-added services in systems ranging from vending machines to digital signage to smart city security applications. But developing and deploying these capabilities into end markets come with a diverse set of challenges, including thermal management, power consumption, ruggedization, and cost.

Maurizio Caporali, Chief Product Officer at the IoT-solution developer SECO, explains at embedded world 2023 how the Intel Atom® x6000 series processors can be paired with standards-based COM Express or SMARC modules to deliver scalable, two- to four-core video-compute platforms that add value at minimum resource consumption. He demonstrates this in a digital signage proof of concept based on the company’s standardized hardware and Clea AI platform that’s capable of running facial and emotion-detection algorithms at the edge; then recommending products based on those outcomes.

Transcript

Maurizio Caporali: Hello, I’m Maurizio Caporali, and I am Chief Product Officer of the SECO Group. My activity is mainly oriented on all the products of the shared products of SECO.

Now there is computer vision—a very important aspect for our customers. Also in the sense to give new features, new services in a different field. For example, starting from industrial automation to analyze the production to have a different kind of information, or to give a new kind of service. For example, for vending machines, digital signage—have information from in front of the device of the person that arrives, and this analysis can perform in different ways.

Also for security application in smart cities and smart buildings. We have many customers in different field that are using our technology to start to define a specific application like this—to define a specific vertical on this market where they need to analyze video, video flow, or video data using the potentiality and power of an Intel electronics-based GPU that can perform really fast and really in a good way the inference of AI model for computer vision.

The important aspect is to start from a standardized model. We believe very much in a specific product that is based on a standard model, like SmartCommerce PressComm HBC. This is the starting point, is our main products, based on an Intel chip where we cover all series of product for industrial application. This gives to our customer the maximum flexibility, the possibility to integrate our solution, our products on their own devices with all the IoTs, with all the peripherals that you need with specific cameras, with a specific sensor that is to create a valid solution for the specific market. Generally, we actually see that very interesting approach in two main solutions. On the part of low-consumption solution, for example, based on Atom series.

With Intel, we have many solutions for different kinds of capability. Also with the flexibility with the same pinout to choose from a Dual-Core simple processor to arrive at a very powerful four-core in the range of very low consumption. For example, starting from 6 facts to arrive at 12 facts maximal, with also the possibility to scale up with respect to the need of power. This is very important for our customer. We can give the opportunity to choose with different level of needs for also starting with a specific product. For example, our SMARC solution with the Dual-Core for the entry level to arrive on the Quad-Core for the top product of our customers. So this has given more flexibility. On the other hand, for COM Express we can go on a query with the last generation GPU with very impressive performance on AI capability and the processing of data flow from cameras of video process. This is an example of what are the potentiality and flexibility of solutions of Intel. This is the low-power consumption solution based on Atom Ixa. This is very good for this kind of application. We have a flow of only one camera. In this case we apply two different models. One model that recognizes my face, another model that is used for giving the emotion and this natural gender. I’m male, age 39, and with this information, in this case, it’s only a simple use case to understand what we can do, suggests me the right solution for me, and I can go ahead. In this case, I can pay also directly with my phone and I will receive the beverage. It’s only to showcase what are the capabilities that is our customer that will perform the right solution.

On top of this, we give the platform to enable our customers on top of the Intel R-Duo to simply develop this kind of application manager to age up remotely with software updates, with control of all data in a simple way can be performed, or transfer the data from the edge to the cloud thanks to the clear platform and the clear solution that is enabled on all our Intel solutions. Me, I’m focused on the design of the solution with respect to the market and with respect to the needs of the customers. Starting from software to arrive to the hardware solution that more in general for modular is more a general aspect to arrive to a system, a more complete solution based on HMI, based on fanless PC to have a complete solution for different market-specific characteristic for the customers.

With SECO, we have a large catalog of products with all the characteristic of the actual cutting-edge technology on the market. We have a solution obviously dedicated on the industrial market with a wide range of industrial solutions. For example, we work together with Intel to define also a functional safety application on top of this market solution to give to our customer a right solution, and also to be ready for certification level, certification point of view, where the safety is one of the main points for industrial devices, and we have proposed a solution that is a final solution of HMI with application with software on top. Another important aspect is the flexibility of the algorithm in the sense of capability and connectivity that actually is very important. That what we define is also a solution that gives Wi-Fi and Bluetooth mobile connectivity, and this is possible. So we have a dedicated team to identify all the solutions, and also a wireless system to give the opportunity to have a complete wireless-connectivity solution.

About the Author

Brandon brings more than a decade of high-tech journalism and media experience to his current role as Editor-in-Chief of the electronics engineering publication Embedded Computing Design. His coverage focuses on artificial intelligence and machine learning, the Internet of Things, cybersecurity, embedded processors, edge computing, prototyping kits, and safety-critical systems, but extends to any topic of interest to the electronic design community. Brandon leads interactive YouTube communities around platforms like the Embedded Toolbox video interview series and Dev Kit Weekly hardware reviews, and co-hosts the Embedded Insiders Podcast. Drop him a line at brandon.lewis@opensysmedia.com or DM him on Twitter @techielew.

Profile Photo of Brandon Lewis