【作者】 李健； 奥帅； 张在成； 何斌；

【Author】 LI Jian;AO Shuai;ZHANG Zaicheng;HE Bin;School of Electronic Information and Artificial Intelligence,Shaanxi University of Science & Technology;College of Electronics and Information Engineering,Tongji University;

【机构】 陕西科技大学电子信息与人工智能学院； 同济大学电子与信息工程学院；

【摘要】 装配式施工由于施工人员众多,施工过程动态变化等因素,导致事故频发不止。为解决装配式施工起重作业导致的施工人员安全问题,本文基于多模态数据,实现了装配式建筑起重施工伤害动态预警。首先分析起重施工场景下导致危险发生的因素;其次将惯性导航传感器信息与计算机视觉目标检测技术结合,构建装配式建筑起重伤害预警模型。实例应用结果表明:提出的预警模型能够准确检测施工危险因素并进行预警,从而降低危险发生的几率,丰富了装配式建筑施工起重过程中的安全管理手段。更多还原

【Abstract】 Due to the large number of construction personnel and dynamic changes in the construction process,prefabricated construction has caused frequent accidents.In order to solve the safety problems of construction personnel caused by the lifting operations of the prefabricated construction,the paper studied the dynamic early warning of the prefabricated building lifting construction injury based on multi-modal data.Firstly,the factors were analyzed that lead to danger in the lifting construction scene.Secondly,combined the inertial navigation sensor information with computer vision target detection technology,a prefabricated building lifting injury early warning model was constructed.The application results of the case showed that the early warning model proposed in this paper could accurately detect construction risk factors and give early warning,thereby reducing the probability of danger and enriching the safety management methods in the lifting process of prefabricated buildings.更多还原