摘要
随着人工智能技术的迅速发展,人脸检测已逐渐成为各行业主流应用技术,如果使用通用处理器对图像数据进行运算处理完成人脸检测,将会产生一定耗时。本文设计一种基于K210芯片中自带的KPU神经网络加速器,完成人脸检测神经网络模型计算,快速实现人脸检测功能。实验结果表明:该系统操作方便和可靠性强,具有一定的应用前景。
With the rapid development of artificial intelligence technology,face detection has become gradually to the mainstream application technology in various industries.If the image data is processed by a general processor to complete face detection,it will take a certain amount of time.In this paper,a KPU neural network accelerator based on the K210 chip is designed to complete the calculation of the neural network model for face detection and realize the face detection function quickly.The result shows that this system operation is convenient with higher reliability,and possesses a certain prospect of application.
作者
王浩
WANG Hao(Institute of Artificial Intelligence,Suzhou Chien-shiung Institute of Technology,Taicang 215411)
出处
《办公自动化》
2021年第11期58-59,39,共3页
Office Informatization
基金
2020年度江苏省现代教育技术研究课题“基于OBD车联网应用系统在技术技能型人才培养中的应用研究”(编号:2019-R-76578)
关键词
KPU
神经网络
人脸检测
Knowledge Proccess Unit(KPU)
the neural network
face detection
作者简介
王浩(1971-)男,硕士,副教授.研究领域:物联网工程应用研究。
