READY FOR PROJECT START

phyBOARD®-Pollux AI kit

  • Reference design for the development of simple to complex AI applications
  • Development examples for face and object recognition are included.
  • Reference platform with NXP i.MX 8M Plus, phyCAM-M camera and optional display
  • Additional add-ons for increased AI performance via Hailo or Ara240 DNPU
  • Step-by-step instructions for creating facial recognition software
  • Ready-to-use Full Feature Board Support Package (BSP)
  • Easy integration of the AI ​​toolset into your own Yocto BSPs
  • NXP eIQ® Machine Learning Software integrated into the BSP
  • Using the NPU directly via driver API, Python and GStreamer

phyBOARD-Pollux AI Kit for Edge AI Development

The phyBOARD-Pollux AI Kit serves as both a learning platform and a reference design for unlocking the embedded vision and machine learning capabilities of the NXP i.MX 8M Plus processor. It is specifically designed for developing simple to complex AI applications and includes all the necessary hardware and software components for object and face recognition applications. The AI ​​computing power can also be expanded with Hailo or Ara240 DNPUs as needed via our add-ons.

AI Use Case: Object and Face Recognition

The kit is based on the phyCORE-i.MX 8M Plus System on Module (SoM), which leverages the capabilities of the NXP i.MX 8M Plus processor. With its quad-core Arm Cortex (1,6 GHz), integrated NPU with 2,3 TOPs of performance, and HDR-capable Image Signal Processor (ISP), the NXP processor is ideal for object and face recognition applications. The necessary camera is included: The proven phyCAM-M camera system consists of a cost-effective and industrial-grade VM-016 HD camera with 1 MP resolution and a MIPI-CSI-2 interface, an M12 lens, and a lens holder. Image data is captured and processed directly by the integrated ISP. The advantage: No post-processing of the image data is required. 

Demo application for facial recognition included

The provided demo application facilitates the use of neural networks for object comparison using machine and deep learning algorithms. Specific facial features are extracted from live-recorded images of faces using various neural networks. These features are then compared to pre-trained images of faces belonging to celebrities. The face with the fewest deviations from the user's own face is displayed as the best match.

Software ecosystem

Integration into your own BSPs is designed to be as simple as possible. You can also use the ready-to-use, full-feature BSP provided by Phytec, which includes the development environment for NXP's eIQ.®Machine learning software is also already integrated. Expanding AI performance with additional DNPUs is also easily possible via driver API, Python, and GStreamer.

Sales manager Ralf Orschau explains the possibilities of the i.MX 8M Plus platform.

Celebrity Face Match AI Demo presented by Nico Bendisch.

Module phyCORE-i.MX 8M Plus
SOM Mounting Connector insertion
Processor i.MX 8M Plus Quad
Architecture ARM Cortex-A53 / Cortex-M7
Frequency 4x 1.6 GHz + 800 MHz
SPI NOR Flash 64 MB
eMMC 8 GB
LPDDR4 RAM 2 GB
EEPROM 4 kB
Ethernet 2x 10/100/1000 Mbit/s
Wireless 802.11 b/g/n (ac) 2.4 GHz / 5 GHz (via optional Wi-Fi Adapter)
Bluetooth BLE 4.2
USB 2x 3.0 Host
Serial 1x RS-232/RS-485 Full Duplex/RS-485 Half Duplex
CAN 2x CAN FD
Digital I/O optional via Expansion Connector
PCle 1x miniPCIe
Mass Memory microSD Card Slot
Display LVDS (1x 4 lanes or 1x 8 lanes), MIPI DSI (1x 4 lanes), HDMI
Audio SAI via A/V Connector
Camera 2x MIPI CSI-2 (phyCAM-M)
Expansion Bus I²C, SPI, SDIO (8-bit), UART, JTAG, USB
JTAG via Expansion Connector
User Control Elements 1x Reset Button, 1x On/Off, 1x Boot source select
Boot Source eMMC, SD Card, NOR, USB Serial Download, Internal Fuses
RTC Real Time Clock (mounted on SoM) with Gold Cap Backup
Power supply 12 V - 24 V
Dimensions 160 mm x 77 mm
Temperature range -40°C to +85°C
Betriebssystem Linux
Distribution Yocto

phyBOARD-Pollux KI Kit

Didn't find an answer? Just ask your question in Live Chat

phyBOARDPollux AI kit

Item No. KPB-03123-AI-Alph

symbol online shop.svg

phyBOARD Pollux
AI kit


Item No.: KPB-03123-AI

Available variants:

430.00  plus VAT.

Quantity:

Online seminars _ Helpful embedded knowledge explained in a nutshell in short video sessions

In informative online seminars with our experts and partners, we will inform you about exciting topics from the embedded industry.

You will get a free insight into new hardware and software solutions and learn more about special offers.

Education + Training _ Use our know-how for yours Productsdevelopment

With the know-how transfer from our experts to your developers, you will reach your goal faster!

 

Participants in our training courses receive a solid knowledge of professional hardware and software development.

Do you have questions about an embedded topic? Arrange an online meeting now!

 

Within 30 minutes you will learn new solutions for your project.
You can also secure exclusive offers after your conversation.