Offline AI Binocular 3D Vision Recognition Sensor (Face / Palm Vein / QR Code)

This AI binocular vision sensor is a standalone, multi-modal recognition module engineered for high-security and versatile identity authentication. It operates on a powerful onboard AI processor, utilizing dual color and infrared cameras to perform all computations locally. The core of its functionality lies in its three-in-one identification capability, integrating robust Face Recognition, secure Palm Vein Recognition, and universal QR Code scanning into a single, compact device. Key features include advanced 3D Liveness Detection for anti-spoofing, complete offline AI processing that eliminates any computational load on the host system, and local storage for up to 1,000 users. With its simple UART interface, this offline AI vision sensor is ideal for rapid integration into applications such as secure access control, smart home systems, and automated retail terminals.

High-Security Biometrics with 3D Liveness Detection

Security is paramount in authentication systems. This face recognition sensor employs a binocular camera system (RGB + Infrared) to capture 3D depth information, enabling a sophisticated liveness detection algorithm. This technology effectively distinguishes a live person from a 2D representation, providing robust protection against spoofing attempts using photos or videos. With a Face Recognition False Acceptance Rate (FAR) as low as 0.001%, the module delivers a level of security suitable for critical access control applications.

Versatile 3-in-1 Recognition Capability

The AI binocular vision sensor offers unparalleled flexibility by combining three distinct recognition methods. The deep learning-based face recognition provides fast, contactless access. For enhanced security, the palm vein and palm print recognition algorithm identifies the unique subcutaneous vein patterns, a biometric marker that is extremely difficult to replicate. Finally, the integrated QR code decoding offers a convenient solution for granting temporary access, processing payments, or device pairing, making the sensor adaptable to a wide range of operational requirements.

Powerful On-Module Offline AI Processing

Equipped with a dedicated System-on-Chip (SoC) featuring a 0.5 TOPS Neural Processing Unit (NPU), the AI vision camera handles all complex AI algorithms internally. This "edge computing" design means it requires zero computational resources from the host device, allowing it to be controlled by any microcontroller, including basic models like an Arduino Uno. All user data, including 1,000 face and 1,000 palm vein templates, is stored and processed locally, ensuring system performance, data privacy, and functionality even without a network connection.

Seamless Integration via UART & USB

Designed for ease of implementation, the binocular camera communicates recognition results through a simple UART serial protocol, making it straightforward to integrate into both new and existing product designs. Furthermore, the USB interface supports the UVC (USB Video Class) protocol, allowing the module to stream video like a standard webcam. This dual functionality is invaluable for applications requiring video feeds for monitoring, visual confirmation, or building video intercom systems.

Figure: AI binocular vision recognition sensor hardware composition diagram

Features:

• Binocular 3D liveness detection resists spoofing via photos/videos
• Neural network-based algorithms: face recognition, palm vein recognition, QR code decoding
• Local storage: 1,000 faces + 1,000 palm vein templates
• Operates in complete darkness and non-direct sunlight outdoors
• Zero host-computation requirement; all AI processing occurs offline on-module

Applications:

• Smart home systems
• Smart door locks
• Self-service retail payment terminals
• 3D face/palm vein payment systems
• Palm vein authentication systems

Specification:

Hardware Parameters

• SoC: Arm CPU@900MHz, NPU@0.5TOPS, RISC-V@600MHz
• DDR2: 64MB
• Flash: 32MB
• Cameras: Dual 1/5” CMOS, 2MP, dual MIPI interfaces
• Lens: FOV: 83° diagonal; optimal focus: 60cm
• LED: IR@850nm, RGB@650nm (90° illumination)
• USB: UVC video transmission (expandable to UAC); MJPEG output (H.264/YUY2 expandable)
• Communication: UART@115200 baud

Power Parameters

• Supply voltage: 5-12V
• Operating current: 320-330mA@8V
• Standby current: 120–130mA (auto-detection mode)
• Shutdown current: 0µA (non-auto-detection mode)

Other Parameters

• Operating temperature: -20°C to +60°C
• Storage temperature: -30°C to +70°C
• Humidity: 10–93% RH (non-condensing)
• Audio I/O: Expandable (speaker/microphone)
• Boot time: 900ms–2.5s (varies with stored user count)
• User capacity: 1,000 faces; 1,000 palm veins
• Algorithms: Binocular liveness detection, deep learning face recognition, palm vein recognition, QR code recognition
• Liveness metrics:
• FAR (False Acceptance Rate) ≤1%
• FRR (False Rejection Rate) ≤1%
• Face recognition:
• Pass rate: 98.85%
• FAR: 0.001%
• Recognition angle: ±20° pitch/yaw; supports multi-angle enrollment
• Detection range: Face: 30–120 cm, palm vein / QR code: 15cm
• Dimensions: 57.8×20×10.12mm

Documents:

• Product wiki
• Development Manual

Product Includes:

• AI Binocular Vision Recognition Sensor (Face & Palm Vein & QR Code) ×1
• 1.25mm 4-pin to Dupont female connector cable ×1
• USB data cable ×1