摘要
为了保证船舶的安全航行,有效预防和打击犯罪活动,在船舶的公共区域广泛安装了视频监控系统。目前,这些系统只能记录下拍摄的信息供日后调取,却不能实时的判断画面中的行为,对犯罪活动及时预警。采用物联网技术将拍摄的信息传送给云平台,在云平台下通过CNN(convolutional neural network,卷积神经网络)模型来提取特征向量,同时应用RNN(Recurrent Neural Networks,循环神经网络)对拍摄的信息进行判断,及时将危险预警发送给相关人员处理。实验表明本文提出的方法是可行、有效的。
In order to maintain the safety of ship navigation, effectively prevent and combat illegal crime, video surveillance systems are widely installed at the public area of the ship. At present, these systems can only record the videos for sequential use, and cannot judge real-timely the actions in the videos or alarm illegal crime in time. This paper introduces a method that transfers information to cloud platform upon internet of things, and CNN(convolutional neural network)model is used to extract the feature vector, RNN(Recurrent Neural Networks) is used to judge the actions at the same time. Experiments show that the method proposed in this paper is feasible and effective.
作者
宋全记
SONG Quanji(Department of Electromechanical and Information Engineering,Sichuan College of Architectural Technology,Deyang,Sichuan 618000,China)
出处
《微型电脑应用》
2020年第4期153-156,共4页
Microcomputer Applications
关键词
船舶安全航行
物联网
云平台
卷积神经网络
循环神经网络
safety of ship navigation
internet of things
cloud platform
convolutional neural network
recurrent neural networks
