Raspberry Pi AI Camera for Artificial Intelligence Neural Networks Intelligent Vision Projects

SC1174 is an AI-powered camera module for Raspberry Pi, built around Sony’s IMX500 Intelligent Vision Sensor. This sensor integrates a 12-megapixel CMOS image sensor with on-chip neural network acceleration, allowing for advanced vision-based AI applications without the need for an external AI accelerator. It can embed tensor metadata directly into captured images or video streams, freeing up the Raspberry Pi’s processor for other tasks.

The SC1174 is fully compatible with all Raspberry Pi models. It works seamlessly with libcamera, Picamera2, and the rpicam-apps software suite, making it accessible to beginners while offering powerful customization options for advanced users. Its PCB layout and mounting holes match the Raspberry Pi Camera Module 3, although it has a greater depth to accommodate the IMX500 sensor and its optics.

Key Features:

• AITRIOS support for scalable edge AI development

• Manual focus adjustment and built-in IR cut filter

• Image resolution: 4056 × 3040 pixels (H/V)

• Sensor size: 7.857mm (Type 1/2.3), Pixel size: 1.55 × 1.55μm

• Lens: 4.74mm focal length, F1.79 aperture

• Field of View: 66 ±3° (horizontal), 52.3 ±3° (vertical)

• Input tensor formats: int8 / uint8, Max tensor size: 640 × 640

• Output formats: Bayer RAW10, YUV/RGB (via ISP), ROI data, and tensor metadata

• Memory: 8,388,480 bytes for firmware, model weights, and processing

• Connection: 200mm ribbon cable; supports 15 × 1 mm FPC or 22 × 0.5 mm FPC connectors

• Module dimensions: 0.98" × 0.94" × 0.46"

• Operating temperature: 0°C to 50°C