The ICORE-1126BQ38 utilizes the Rockchip RV1126B AI vision processor (3 TOPS NPU). This compact module (38mm × 38mm) supports 2 billion multimodal models and deep learning. Equipped with a 12M AI-ISP, it enables excellent day and night imaging, 4K video processing, multi-camera AI stitching, 6-axis digital image stabilization, and multi-source fusion (RGB/IR/ToF). 

Furthermore, it features AOA low-power sound detection and built-in encryption (SM2/SM3/SM4) security mechanisms, making it an ideal choice for AIoT applications such as facial recognition, smart security, and industrial vision.

• Product Datasheet: Click Here
• Product Wiki: Click Here

Feature

•  AI Vision Processor: Its performance is more than twice that of similar chips, providing powerful support for high-intensity computing and multitasking.
• Supports Multimodal Models: With advanced optimization techniques, including quantization and Transformer acceleration, it delivers powerful on-device AI capabilities for image understanding and cross-modal fusion in lightweight scenarios.
• Built-in 8-Megapixel AI-ISP: It features a new generation hardware-level image signal processor (ISP) that supports up to 12 megapixels and integrates accelerators for key algorithms such as HDR, 3A, LSC, and noise reduction. It also features an AI image signal processor (AI-ISP).
  
• Simultaneous Input From 5 Cameras: It features two MIPI-CSI/LVDS/SubLVDS DPHYs (each DPHY supports 4 channels: 1×4 channels or 2×2 channels) and one DVP interface (BT.601/BT.656/BT.1120), supporting up to five simultaneous input cameras.
  

Product Overview