Artificial Intelligence for an Automated Payment Terminal

To develop the live demo, our embedded vision team worked closely with NXP Semiconductors. The result is an automated retail payment terminal based on AI software. The embedded vision system identifies and classifies the products in the shopping basket and displays the respective prices. Customers can then choose which products they want to put in the shopping basket. A trained neural network consequently recognizes the products via the video stream and displays the total price - the principle is similar to a face recognition application.

Learn more about our embedded vision solutions Download the complete Use Case now

Our retail solution is based on the Basler Embedded Vision Kit, which consists of the following components:

• dart BCON for MIPI camera module from Basler with ON Semiconductor’s AR1335 sensor
• i.MX 8QuadMax SoC from NXP
• i.MX 8QuadMax based processing board
• SMARC 2.0 carrier board

The complete demo setup is close to a real-world application and incorporates all vision relevant factors.

Discover the Embedded Vision Portfolio

The software is an essential part of this application. It consists of two main components: the system software and the application software.

Let's look at the system software first: The different elements of the system software must be connected to each other and form a coherent system for the embedded system to work. This is where Basler can use its many years of vision experience to combine software elements such as NXP's Yocto Kernel and BSP to create a lean embedded system.

Our AI application software is based on a customized Convolutional Neural Network (CNN). The model uses the latest deep learning and device-internal processing techniques for fast and reliable response in a seamless scenario.

Go to Software for Embedded Vision

With Basler's embedded vision solutions, our customers can always rely on one contact who will take over all relevant coordination tasks. Our retail application offers not only our customers, but also the retail industry a number of advantages:

• This system is scalable. Products can be added easily to the system: a host-sided training allows for an easy deployment to the edge systems using IoT technology
• This robust system is optimized for vision, has high accuracy and a fast inference time
• Hardware is industry-proven and long-lasting
• Cost-efficient, lean design due to embedded and IoT technology
• Small form factor allows for different integration options
• Retail stores benefit from lower labor costs as well as significantly improved customer experience through instant checkouts, minimized queues and 100% checkout capacity at all times, even when the shop is open 24/7.