基于可编程SOC的图像边缘检测系统  

Image Edge Detection System Based on Programmable SOC

作　　者：胡龙宝 李丕丁[1] HU Long-bao;LI Pi-ding(School of Medical Devices and Food Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China)

机构地区：[1]上海理工大学医疗器械与食品学院,上海200093

出　　处：《软件导刊》2020年第3期244-247,共4页Software Guide

摘　　要：随着人工智能的不断发展,云计算面临数据量大、实时性低等问题。为了解决该问题,一种新的计算方式——边缘计算进入了人们的视野。边缘计算推动人工智能走向嵌入式系统,与传统嵌入系统相比,边缘人工智能要复杂得多,传统嵌入式设备计算能力弱、效率低的问题更为突出。基于Xilinx公司Zynq-7000系列的全可编程SOC完成系统设计,实现对视频的边缘检测。测试结果显示,硬件边缘检测速度比纯软件边缘检测速度快10倍,为边缘人工智能在嵌入式应用方面提供了可行方向。With the continuous development of artificial intelligence,cloud computing is facing the problems of large amount of data and low real-time.To solve this problem,a new computing method,edge computing,has come into people’s vision.Edge computing drives AI to embedded systems.Compared with traditional embedded systems,edge AI is much more complex.Traditional embedded devices have weak computing power and low efficiency.In this paper,based on Xilinx Zynq-7000 series of fully programmable SOC to complete the system design,to achieve video edge detection.The test results show that the speed of hardware edge detection is 10 times faster than that of software edge detection,which provides a feasible direction for the application of edge artificial intelligence in embedded system.

关 键 词：边缘人工智能 边缘计算 可编程SOC 软硬件协同 边缘检测 嵌入式应用 

分 类 号：TP317.4[自动化与计算机技术—计算机软件与理论]