Fujisoft is rolling out an advanced, AI‑driven site security system built on AMD’s Embedded+ platform that aims to cut false alarms and improve detection accuracy across factories, warehouses and other business sites.
The solution leverages AI‑based image recognition to move beyond simple motion detection to read patterns in video feeds from multiple security cameras and sensors to distinguish people, vehicles and normal versus suspicious activity.
It is based on AMD’s Embedded+ architecture, which combines an AMD Ryzen Embedded processor with a Xilinx Versal adaptive SoC on a single board. This mix of x86 compute and FPGA‑style programmable logic lets Fujisoft run AI inference locally, keep latency low, and flexibly connect to a range of sensors and I/O interfaces without redesigning the entire hardware stack. The pre‑integrated platform also reduces development time and system‑integration costs for Fujisoft’s hardware and software teams.
“By combining both adaptive and x86 embedded technologies into a single, integrated solution, the AMD Embedded+ platform fully leverages Fujisoft’s strengths in both hardware and software development. Through conversations with various customers, we have identified specific challenges, and we believe that Embedded+ is well positioned to address and solve these issues,” said Naoya Yanagitsubo, Manager of Fujisoft.
Globally, Fujisoft now joins a growing group of vendors offering AI‑enhanced site security products. Companies such as Hikvision, Dahua, Milestone Systems, and Axis Communications already sell camera and video‑management platforms that embed onboard AI for face recognition, perimeter intrusion alerts and behaviour analysis.
Cloud‑oriented security players such as Verkada and startups using NVIDIA Jetson‑ or Google Coral‑based edge boxes also push similar “AI‑filters first, cloud later” approaches, focusing on reducing storage and bandwidth costs while keeping sensitive data on‑site where possible.
Entelechy Asia 