施工工程现场入侵监测系统设计  

Design of Intrusion Monitoring System for Construction Site

作　　者：金国栋[1] 杨钦 杨大健 

机构地区：[1]中国建筑第八工程局有限公司,上海

出　　处：《计算机科学与应用》2022年第12期2736-2743,共8页Computer Science and Application

摘　　要：随着社会的快速发展,隐私性和保密性问题日渐受到人们的关注,在施工领域也不例外。工程项目的图纸资料、施工现场的进程与技术本身就带有一定的保密性,而一些特殊的工程项目如机场等则更具有防泄密和防入侵的需求。针对这种问题,提出一种针对施工人员泄密行为和施工现场防入侵的监测系统设计思路。在设计中,分别从网络入侵和场地入侵两方面来进行系统设计,并采用一种融合卷积神经网络和长短时记忆网络的混合深度学习模型,用于自动处理施工现场环境中的可能泄密行为,通过模型识别训练证明了算法的可行性。With the rapid development of society, privacy and confidentiality issues are increasingly concerned by people, and the construction field is no exception. The drawing data of the engineering project, the process of the construction site and the technology itself have a certain degree of confidentiality, and some special engineering projects such as the airport are more anti-leak and anti-intrusion needs. In order to solve this problem, this paper puts forward a design idea of monitoring system aiming at the leakage behavior of construction personnel and the intrusion prevention of construction site. In the design, two aspects of network intrusion and site intrusion were respec-tively used to design the system, and a hybrid deep learning model integrating convolutional neural network and short and long time memory network was used to automate the possible leakage behavior in the construction site environment. The feasibility of the algorithm was proved through model recognition training.

关 键 词：施工现场 入侵监测 行为识别 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]