摘要
为给人群疏散问题研究提供准确可靠的依据,本文提出一种基于深度卷积神经网络的人群疏散运动仿真模型。为获取神经网络训练所需数据,采用CSRNet神经网络和DBSCAN算法从监控视频中提取真实人群轨迹数据,通过深度卷积神经网络的训练,对真实的人群行为模式进行深度学习,并利用训练出的深度卷积神经网络建立人群运动仿真模型。结果表明,该模型可准确预测人群的运动行为,真实模拟人群的运动轨迹,可为应急疏散策略的制定和公共场所疏散通道的设计提供依据。
In order to provide accurate and reliable fundamen⁃tals for the study of crowd evacuation,a crowd evacuation mo⁃tion simulation model based on deep convolutional neural net⁃work was proposed.To obtain the data required for neural net⁃work training,the CSRNet neural network and DBSCAN algo⁃rithm were deployed to extract real crowd trajectory data from surveillance video.By the training of deep convolutional neu⁃ral network,the real crowd behavior pattern was deeply ex⁃tracted so as to establish the crowd motion simulation model.The results show that the model can accurately predict the movement behavior and truly simulate the movement trajectory of the crowd,thereby providing a basis for the formulation of emergency evacuation strategies and the design of evacuation channels in public places.
作者
王宗尧
吕子龙
徐欣然
毕容珲
隋聪
WANG Zongyao;LV Zilong;XU Xinran;BI Ronghui;SUI Cong(Collaborative Innovation Center for Transport Studies,School of Maritime Economics and Management,Dalian Maritime University,Dalian 116026,China)
出处
《大连海事大学学报》
CAS
CSCD
北大核心
2024年第2期101-108,共8页
Journal of Dalian Maritime University
基金
国家自然科学基金资助项目(72072018,71831002)
中国博士后研究基金会(2019M651101,2021T140081)。
关键词
人群疏散
仿真模型
深度卷积神经网络
深度学习
crowd evacuation
simulation model
deep conv⁃olutional neural network
deep learning
