Intelligent Detection Of Safety Helmet Wearing Of Construction Personnel For BIM Application

As the cornerstone of economic construction,the field of construction plays an important role in promoting China’s sustainable development.However,the production and supervision methods of the traditional construction industry can no longer keep up with the development of the information society.It is necessary to upgrade and innovate the technology of the construction industry in combination with various modern information technology means to realize the information transformation of the construction industry.In recent years,with the vigorous development of information technology,Internet of things,Building Information Model(BIM)technology,industrial Internet and other technologies,from building design to construction management,have played an important role in improving project quality,optimizing personnel management and accelerating construction period,and significantly improved project efficiency.In order to protect the life safety of constructors and maximize the benefits while realizing safe production,based on the application and research of BIM technology,and combined with the research of deep learning algorithm,this paper designs an intelligent detection method for constructors’ helmet wearing for BIM application,and puts forward a practical scheme to solve the problems of safety supervision of construction site and the visualization of building information monitoring.The specific work of this paper is as follows:(1)In BIM modeling,Dynamo for Revit software is used to realize the rapid creation of threedimensional bridge information model through visual programming,including horizontal curve design,bridge centerline framework,creating contour parameter cluster,creating bridge substructure,and finally importing the bridge entity into Revit software;(2)In terms of helmet detection,YOLOv4 algorithm is adaptively improved based on the underground scene of the construction site,the recognition rate of targets below 32×32 scale is increased by 6.02% and the overall recognition rate is increased by 2.91% with the final weight model,improve the accuracy of helmet detection system and the robustness of intelligent detection system in site scene;(3)In the linkage between BIM model and intelligent detection system,it is mainly divided into the development and research of BIM near part and BIM far part.Through site information network,the monitoring information obtained from the BIM far part is transmitted to the local information database at the BIM near part,and then the monitoring information is visually displayed on the BIM model through the secondary development of Revit,so as to realize the intelligent supervision of safety helmet wearing detection on the construction site;At the same time,screenshots of violation information are saved and uploaded to the local resource database to give early warning of unsafe behaviors.Through the integration and application of BIM Technology and deep learning algorithm,the safety helmet wearing detection of construction personnel designed and studied in this paper can effectively solve the disadvantages of traditional human safety supervision and improve the intelligence and informatization of site supervision system;At the same time,the improved YOLOv4 algorithm improves the recognition accuracy of the detection system for low-size targets,combined with the linkage work of BIM near end and BIM far end,it can effectively improve the scope of site safety supervision,investigate potential safety hazards,and provide guarantee for the life safety of construction personnel.