基于计算机视觉的临边洞口坠落事故自动预警方法

An Automatic Warning Method of Falling from Edges and Openings Based on Computer Vision

作为高处坠落事故的主要危险源之一,临边洞口一直都是施工现场安全管理的重点。然而,施工现场环境复杂,临边洞口数量众多,仅靠管理人员很难做到全方位的管控。为此,本研究提出了一种基于计算机视觉的临边洞口坠落事故自动预警方法,协助管理者掌握现场信息,提前预防安全事故。首先,根据塔吊相机拍摄的视频信息,进行工人的识别与定位;然后,从图像数据中确定危险区域位置,并自动测算工人到危险区域的距离;最后,根据所设置的危险判别规则,分析工人所处的安全状态,并给出预警信息。该方法在新桥国际机场T2航站楼项目中进行了应用,结果显示:距离测量误差能有效控制在6%以内,因此能有效判断工人的风险状态,辅助安全管理。

As a major hazardous source of falling accident from height,edges and openings have always been the focus of safety management on construction sites.However,the complexity of construction sites and a large number of edges and openings make managers difficult to monitor them simultaneously.Therefore,this study proposed an automatic warning method for falling from edges and openings based on computer vision,to assist managers in mastering on-site information and prevent safety accidents in advance.Firstly,videos are captured by cameras installed on tower cranes and used to identify and locate site the workers.Then,the location of hazardous area is determined,and the distance between the workers and the hazardous area is measured automatically.Finally,the workers’safety state is judged and warning information is generated in a hierarchical way according to the established judgment rules.The proposed method was applied in Terminal 2 of Xinqiao International Airport.The results show that the overall measurement error is within 6%,which can effectively judge worker’s safety state and assist in safety management.