基于MPSoC的智能摄像头系统设计  

Design of smart camera system based on MPSoC

作　　者：闫彬洁 刘一清[1] YAN Binjie;LIU Yiqing(School of Communication and Electronic Engineering,East China Normal University,Shanghai 200241,China)

机构地区：[1]华东师范大学通信与电子工程学院,上海200241

出　　处：《电子设计工程》2021年第18期175-180,共6页Electronic Design Engineering

摘　　要：针对传统摄像头终端仅能存储监控画面,需花费人力时间成本回看或占用网络资源将视频上传云端计算才能捕捉画面有用信息的现状,文中提出了一种基于MPSoC(Multi-ProcessorSystem onChip)的智能摄像头终端系统。文中系统采用边缘计算技术,使用自主设计的软硬件开发平台,以Xilinx公司的MPSoCZU7EV芯片为处理核心,通过实时处理从MIPI(MobileIndustryProcessorIn⁃terface)图像传感器采集超高清(3840×2160 P)视频数据,实现对视频关心画面的全自动捕捉、快速识别及分析信息的本地显示和云端上传。该系统应用于车载实时视频信息智能识别时,对交通标志的识别处理时间延迟为10 ms。可运用于各种需要实时处理视频数据的智能监控系统。In current situation that traditional camera terminals can only store monitoring images,it takes human time and cost to look back or occupy network resource to upload videos to cloud computing to capture useful information of pictures.This paper proposes a design of intelligent camera terminal system based on MPSoC(Multi⁃Processor System on Chip).This system is developed by an independently designed software and hardware platform,using edge computing technology.It takes MPSoC ZU7EV chip made by Xilinx as the processing core,captures ultra⁃high definition(3840×2160 P)video data by MIPI(Mobile Industry Processor Interface)image sensor and processes images in real time.This design automatically captures and quickly identifies the concerned video images,displaying the analyzed information on local screen and uploading it to the cloud.When the system is applied to vehicle real⁃time video information intelligent recognition,the time delay of traffic signs processing is 10 ms.It can be used in a variety of intelligent monitoring systems which need real⁃time processing of video data.

关 键 词：MPSOC 智能摄像头 终端系统 边缘计算 交通标志识别 

分 类 号：TN871[电子电信—信息与通信工程]