摘要
随着社会的不断发展,人工智能越来越深入的融入到人们的生活,而人工智能的一个应用智能家居大大提高了人们的生活水平。智能家居中最重要的部分就是人体安全监控,而这一技术的核心就是人类活动识别。传统的识别技术速度慢、精确度低,而度量学习能很大程度改善传统人类活动识别技术的不足。本文研究的是将大间隔最近邻应用到人类活动识别中以提高识别效率。
With the continuous development of society,artificial intelligence is more and more deeply integrated into people's lives,and an application of artificial intelligence in smart home has greatly improved people's living standards.The most important part of the smart home is human security monitoring,and the core of this technology is human activity recognition.Traditional recognition technology is slow and has low accuracy,and metric learning can greatly improve the deficiency of traditional human activity recognition technology.This paper studies the application of largespaced nearest neighbors to human activity recognition to improve recognition efficiency.
作者
王丽晓
WANG Li-xiao(Zhengzhou Vocational College of Electronic and Information Technology,Zhengzhou Henan 451400)
出处
《数字技术与应用》
2020年第2期91-92,94,共3页
Digital Technology & Application
关键词
大间隔最近邻
人类活动识别
度量学习
large-spaced nearest neighbor
human activity recognition
metric learning