AI image recognition with digital learning can detect far more detail in MRI and X-ray images than the human eye, thus improving the accuracy of diagnoses and helping to prioritize treatment.
AI technology in embedded systems is typically incorporated into service model innovation, rather than the manufactured product itself, and exhibits a high degree of sophistication in the various tasks it performs, including data collection, data analysis, pre-formed models, and inference. Regarding machine learning, programmed algorithms rely on reliable and powerful computing units to process big data. Therefore, edge computing plays a crucial role in connecting the cloud and sensor devices.
Advantech offers an embedded platform (SOM-5871/AIMB-228) equipped with the latest AMD Ryzen™ Embedded V1000 processors, combined with the Mentor® Embedded Linux operating system. The AMD Ryzen Embedded V1000 processor supports the frameworks, libraries, tools, and compilers for e-sight applications that leverage the full power of the Ryzen processor with Radeon™ Vega graphics processing unit technology. Importantly, all three companies embrace open source, such as OpenVX and OpenCL API support in the Linux® kernel, enabling users to migrate digital learning across diverse hardware architectures for a wide range of AI applications.
Rex Lee, director of Advantech's Embedded IoT Group, stated: “Everyone can find many open-source AI technologies. However, integrating AI technology into embedded systems remains a considerable challenge. From data collection and model development to the implementation of edge inference systems, no one can tackle AIoT alone. That's why Advantech works closely with our co-creation partners to accelerate the deployment of AI in embedded systems.”.
“By combining the power of the AMD Ryzen™ Embedded V1000 processor, eSight software, and solutions from our partners Mentor and Advantech, we are helping our customers take advantage of a future where they can easily create and deploy AI-enhanced applications within their embedded ecosystem,” said Stephen Turnbull, Director of Product Marketing, Embedded Solutions, AMD. “This will open the door to countless unique scenarios and use cases that will not only help the embedded industry embrace AI but, more importantly, will also help businesses run seamlessly and securely.”.
“Mentor has a long history of providing commercial operating system platforms for a wide variety of embedded applications,” said Scot Morrison, general manager, embedded platform technology business unit, Mentor. “By partnering with AMD to leverage its hardware-optimized AI capabilities, we are jointly delivering a comprehensive embedded AI deployment solution with Advantech’s embedded hardware platforms.”.
Advantech offers a broad portfolio of embedded and automation products and professional redesign services.
Product offering:
SOM-5871 (COM Express)
- High-performance "Zen" CPU, 4 cores/8 threads, with Radeon™ Vega graphics, 11 compute units, and 4K displays
- Abundant super-speed I/O and expansion bus. PCIe® gen3 and USB 3.1 gen2 up to 10 Gbps
- DHCS thermal solution optimized for high TDP design
- Supports secure boot and fast boot via custom BIOS service
- Supports secure memory encryption and built-in TPM2.0 function
AIMB-228 (Mini-ITX)
- High-performance Zen CPU with 4 cores/8 threads and 32 GB DDR4 SODIMM
- Vega GPU with 11 compute units and support for Quad 4K displays
- Supports PCIe x8, 2 M.2 expansion slots, 6 COM ports, 4 USB 2.0 ports, 2 USB 3.1 Gen2 connectors
- Supports WISE-PaaS/embedded software API and RMM
- Supports wide voltage 12-24V DC input
