基于深度学习的机坪视频语义检索方法  

Research and Implementation of Apron Video Semantic Retrieval System Based on Deep Learning

作　　者：郝家祺 甘雨 张洁盈 HAO Jiaqi;GAN Yu;ZHANG Jieying(Civil Aviation University of China,Tianjin 300000,China)

机构地区：[1]中国民航大学,天津300000

出　　处：《信息与电脑》2022年第18期36-38,42,共4页Information & Computer

摘　　要：为能够在一段视频中检索特定事件,提出一种基于深度学习的视频语义检索方法,使用你只看一次(You Only Look Once,YOLO)模型检测视频中的对象,通过聚类分析提取对象的语义轨迹,结合检索到的对象与语义轨迹实现特定事件的视频语义检索功能。实验结果表明,该方法能够满足应用要求。In order to retrieve specific events in a video,a video semantic retrieval method based on deep learning is proposed.Use the You Only Look Once(YOLO)model to detect objects in the video,extract the semantic trajectory of objects through clustering analysis,and combine the retrieved objects and semantic trajectory to achieve the video semantic retrieval function of specific events.The experimental results show that this method can meet the application requirements.

关 键 词：视频搜索 语义分析 聚类算法 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]