摘要
随着我国经济的快速发展,各类大型工程层出不穷,对起重机吊装作业的需求不断增加。然而,吊装作业过程中依然存在众多的安全隐患,极易造成人员伤亡等安全事故。因此,该文提出一种基于深度学习和机器视觉的起重机吊装安全监测方法。将深度学习与机器视觉相结合对监控图像中的被吊物和工人进行识别和定位,同时可自主判断工人是否佩戴安全帽。根据监测模型的识别和定位信息,获得工人与被吊物之间的空间关系,为起重机吊装过程提供安全预警信息。为了提高所提方法的实用性和便携性,开发一个起重机吊装安全智能监测系统,不仅可以实时显示监控图像的识别结果,而且能够输出场景的语义描述、发出安全预警信号。
With the rapid development of China's economy,various large-scale projects emerge one after another,and the demand for crane hoisting operation is increasing.However,there are still many safety hidden dangers in the process of hoisting operation,which can easily cause casualties and other safety accidents.Therefore,this paper proposes a crane hoisting safety monitoring method based on deep learning and machine vision.The combination of deep learning and machine vision is used to identify and locate the suspended objects and workers in the surveillance images,and at the same time can independently judge whether workers wear helmets or not.According to the identification and positioning information of the monitoring model,the spatial relationship between the worker and the suspended object is obtained,which provides safety early warning information for the hoisting process of the crane.In order to improve the practicability and portability of the proposed method,an intelligent monitoring system for crane hoisting safety is developed,which can not only display the recognition results of monitoring images in real time,but also output the semantic description of the scene and send out safety early warning signals.
出处
《科技创新与应用》
2024年第2期1-5,共5页
Technology Innovation and Application
基金
国家市场监督管理总局科技计划项目(2021MK044)
江苏省高等学校自然科学研究面上项目(21KJB470031)。
关键词
深度学习
机器视觉
吊装监测
智能监测
安全预警
deep learning
machine vision
hoisting monitoring
intelligent monitoring
safety early warning
