Event
On November 3, NORCO (North China Industrial Computer) released the embedded AIoT motherboard EMB-3582, aimed at AI inference, multi-stream video analysis, and multi-sensor integration for smart streetlights. Industry observers see the smart-streetlight race shifting from “integrating hardware” to “edge compute.”
Technology / Product Details
EMB-3582 is powered by a Rockchip RK3588 processor with an integrated 6 TOPS NPU, designed for wide-temperature outdoor operation and resistance to electromagnetic interference, supporting local processing and aggregation of multi-stream video and multiple sensor types for long-term stable operation. As the compute base of the “urban nerve endings,” it moves the video and sensing computation that used to be offloaded to the backend into the streetlight node itself, cutting backhaul bandwidth and latency.
From the Industry Landscape…
The “brain” of the smart city is being pushed down to streetlight, surveillance, and roadside edge nodes. Edge AI penetration is still low; whoever completes video, sensing, and AI inference at the edge locally gains the voice in scaled smart-city deployment. Edge hosts, AIoT boards, and edge gateways are becoming the key carriers connecting the “city brain” to “terminal sensing.”
Junyi Business Link
Bringing multi-stream video analysis and AI inference to the edge is exactly Junyi Microelectronics’ product focus. Core boards J-3568 / J-3576 / J-3588 and edge motherboards in the K series support local inference for smart streetlights and surveillance, edge computing hosts 5G-EDGE-3399Pro / AI-BOX-3588 provide higher compute bases, the industrial mini PC IPC-i5-1235U serves rack-level applications, and the all-in-one touch PC offers visualization terminals for O&M and dispatch.
Source: NORCO launches EMB-3582 for smart streetlight edge compute (2026-11-03)