Luxonis OAK-D IoT-40 S2 - module with camera IMX378 + 2xOV9282 and ESP32

Description

OAK-D IoT S2 is an advanced camera module designed for vision systems in IoT (Internet of Things) applications, combining stereoscopic depth vision with a high-resolution RGB camera, offering perception similar to that of humans. With built-in neural network processing and computer vision capabilities, the camera allows for advanced real-time image analysis. It uses a USB Type-C port for power and connectivity, providing fast data transfer via USB 3.2 Gen1 (5 Gbps). The OAK-D IoT is based on the RVC2 architecture, offering 4 TOPS of computing power (including 1.4 TOPS dedicated to AI tasks). The OAK-D supports any AI models, including custom ones, after prior conversion. The device supports encoding in H.264, H.265, MJPEG formats at 4K/30FPS and 1080P/60FPS. In terms of computer vision, it offers distortion correction, resizing, image cropping, edge detection, and feature tracking. Additionally, the camera supports both 2D and 3D object tracking via the ObjectTracker module.

The module is equipped with a color camera featuring a 12MP IMX378 sensor (4056x3040 resolution) and two stereo cameras with a 1MP OV9282 sensor (1280x800 resolution). It also includes an ESP32 module, allowing wireless connectivity via WiFi and Bluetooth Low Energy (BLE). Additionally, it features 16 MB of eMMC memory, providing sufficient space for data and software. The kit also includes an integrated speaker and 6 MEMS microphones.

Power consumption in the OAK-D depends on the load of the modules, including the vision processing unit (VPU). Basic power consumption during video streaming is 2.5W - 3W, with additional power requirements of up to 1W for the AI system, up to 0.5W for the stereoscopic system, and up to 0.5W for the video encoder. The total power consumption can reach up to 5W when all functions are used simultaneously. This makes it an ideal solution for projects requiring precise robotic vision, AI, depth analysis, and real-time motion tracking.

Features

• Computing power: 4 TOPS (1.4 TOPS for AI - RVC2 neural network performance)
• Support for any AI models
• Encoding: H.264, H.265, MJPEG - 4K/30FPS, 1080P/60FPS
• Computer vision: distortion correction (warp), resizing, cropping, edge detection, feature tracking
• Stereo depth perception with filtering, post-processing, and RGB depth alignment
• Object tracking: 2D tracking via ObjectTracker node
• Color camera:
  • Sensor: IMX378
  • Field of view (DFOV/HFOV/VFOV): 81° / 69° / 55°
  • Resolution: 12MP (4056x3040)
  • Focus: AutoFocus (AF) from 8cm to infinity
  • Maximum frame rate: 60 FPS
  • Spectral range: Visible light only
  • Sensor size: 1/2.3”
  • Aperture: 1.8 ± 5%
  • Focal length: 4.81mm
  • Pixel size: 1.55µm
• Stereo cameras:
  • Sensor: OV9282
  • Field of view (DFOV/HFOV/VFOV): 89° / 80° / 55°
  • Resolution: 1MP (1280x800)
  • Focus: Fixed Focus (FF) from 19.6cm to infinity
  • Maximum frame rate: 120 FPS
  • Spectral range: Visible light only
  • Sensor size: 1/4”
  • Aperture: 2.0 ± 5%
  • Focal length: 2.35mm
  • Pixel size: 3.0µm
• ESP32 with WiFi and BLE
• 6x microphones
• Speaker
• 16 MB eMMC memory
• MicroSD card slot (up to 2TB)
• Power consumption: up to 5W