









Binocular Camera Module integrates dual onboard CG0328 image sensors to deliver synchronized stereo image capture for depth perception and machine vision. Featuring 2MP UXGA resolution (1632×1232) and a 60mm baseline distance between lenses, the stereo imaging board enables precise depth calculation for robotics and embedded perception systems. Three 24‑pin FPC connectors allow flexible integration with embedded processors and vision controllers. Designed for compact embedded environments, the dual‑sensor vision module supports multiple output formats and resolutions, making the hardware suitable for applications involving AI vision, stereo imaging, and depth‑based environmental analysis.
Dual‑Camera Stereo Vision Architecture
Two CG0328 CMOS sensors positioned with a 60mm binocular distance allow accurate stereo disparity calculation and depth estimation. Such a configuration mirrors human binocular vision, enabling depth perception required for robotics navigation, obstacle detection, and spatial measurement. The stereo camera board captures synchronized frames from both lenses, providing reliable image pairs for stereo matching algorithms used in embedded vision processing platforms and edge AI systems.
High‑Quality UXGA Image Processing
Each onboard CG0328 sensor integrates a full single‑chip camera and image processor capable of UXGA (1632×1232) resolution. Integrated processing functions include exposure control, gamma adjustment, white balance, color saturation tuning, hue control, white‑pixel canceling, and noise suppression. Proprietary sensor technology minimizes common image artifacts such as fixed pattern noise and smearing, ensuring stable color reproduction and consistent visual output across varying lighting conditions.
Programmable Control and Flexible Output Formats
The stereo imaging module supports programmable configuration through the SCCB interface, enabling firmware‑level control of imaging parameters. Multiple digital output formats are available, including RGB565/555, GRB422, YUV (422/420), and YCbCr (4:2:2), as well as compressed or raw RGB data streams. Resolution scaling from UXGA down to smaller formats allows developers to balance image detail and processing load for embedded systems or AI inference pipelines.
Such stereo vision hardware supports robotics perception, depth mapping, gesture recognition, and smart vision experiments. The dual‑sensor camera board serves as a practical embedded imaging solution for machine vision development, academic research, and advanced prototyping where synchronized stereo image acquisition is required.