IP67-rated CM4 AI camera uses Raspberry Pi Compute Module 4 for computer vision applications

EDATEC ED-AIC2020 is an IP67-rated, Raspberry Pi CM4-based industrial AI camera equipped with a fixed or liquid lens and LED illumination that leverages the Raspberry Pi Compute Module 4 to run computer vision applications using OpenCV, Python, And Qt.

We’ve previously written about Raspberry Pi Compute Module-based smart cameras such as the Q-Wave Systems EagleEye camera (CM3+) working with OpenCV and LabVIEW NI Vision and the StereoPi v2 (CM4) with stereo vision. But the EDATEC ED-AIC2000 is the first ready-to-deploy Raspberry Pi CM4 AI camera we’ve covered so far.

CM4 AI Camera

EDATEC “CM4 AI camera” (ED-AIC2020) specifications:

  • SoM – Raspberry Pi Compute Module 4  up to 8GB RAM, up to 32GB eMMC flash
  • Camera
    • 2.0MP global shutter or 5.0MP rolling shutter
    • Acquisition rate – Up to 70 FPS
    • Aiming point – Red cross laser
    • Built-in LED illumination (optional)
    • Scanning field
      • Electronic liquid lens
      • Fixed focal length lens
  • Networking
    • Gigabit Ethernet M12 port
    • Communication protocols – Ethernet/IP, PROFINET, Modbus TCP, TCP/IP, MC (MELSEC)
  • Expansion – M12 connector with 4x optical isolated GPIO and RS232
  • Misc – Power, Failure, Good Read, and Network LEDs
  • Power Supply – 10 to 30V DC up to 1.5A @ 24V DC (36W) via M12 connector (shared with expansion port)
  • Dimensions – 75 x 54 x 50mm (Industrial blue aluminum shell)
  • Weight – 150 grams
  • Temperature Range – Storage: -40°C to 70°C
  • Humidity – 5% to 90% relative humidity, non-condensing
  • IP Rating – IP65 and IP67
  • Shock Resistance – EN 60068-227, 30 g, 11ms, 3 impacts per axial direction
  • Vibration Resistance – EN 60068-26, 14 mm @ 2 to 10 Hz, 1.5mm @ 13 to 55 Hz; 2 g @ 70 to 500 Hz; 2 hours per shaft.


EDATEC provides a camera SDK with OpenCV 4.5.4, Python 3.7, and Qt pre-installed, and lists support for YOLO V5 and V8, third-party Machine Vision Libraries like Halcon, AVS and more. But the company’s CM4 AI camera appears to be especially suited to barcode/QR code and OCR applications with EDATEC highlighting support for one-dimensional codes (Code 39, Code 93, Code 128, I 2 of 5, UPC/EAN), two-dimensional codes (Data Matrix, DotCode, MaxiCode,  PDF417, Micro PDF417, QR Code, Micro QR), and optical character recognition capabilities with OCR-A, OCR-B, MICR, and trained OCR.

Samples of the EDATEC ED-AIC2020 CM4 AI camera can be purchased on Aliexpress for $238 and up plus pricy shipping ($77 to the US) because only DHL and FedEx shipping options are available.

There are three somewhat confusing options:

  • No Accessories – $238 for the CM4 AI camera only.
  • Accessories  – $55 for the accessories only, i.e. M12 cables and mounting attachments.
  • Accessories included – $293 for the CM4 AI camera and accessories.

M12 Ethernet & Power cables

We’re not told which Raspberry Pi CM4 module is included in the camera sold on Aliexpress. More details may also be found on the product page.

Share this:
FacebookTwitterHacker NewsSlashdotRedditLinkedInPinterestFlipboardMeWeLineEmailShare

Support CNX Software! Donate via cryptocurrencies, become a Patron on Patreon, or purchase goods on Amazon or Aliexpress

ROCK 5 ITX RK3588 mini-ITX motherboard

One Reply to “IP67-rated CM4 AI camera uses Raspberry Pi Compute Module 4 for computer vision applications”

Leave a Reply

Your email address will not be published. Required fields are marked *

Khadas VIM4 SBC
Khadas VIM4 SBC