摘要
安全生产是变电站日常生产过程中最核心的要求。在变电站施工现场中,要面对高压危险设备,施工人员进入现场佩戴安全帽是必要的安全措施。而且对于不同身份的人员,要求佩戴的安全帽颜色也不同。在变电站中,要求能够实时监视生产过程中的施工方式是否符合安全生产的要求。该文提出了一种基于行人检测和安全帽颜色识别的算法,实现了对监控画面中检测到的人,识别其是否按照安全生产规范佩戴了安全帽,并且根据安全帽的颜色,进一步确定进站施工人员是否按照施工要求到位。该算法准确率高,大大降低了人工一一排查监控视频数据的时间开销,节省了人力,也提高了工作效率。
Safe production is the core requirement of the substation daily production process. In the substation construction site, to face the danger of high voltage equipment, construction workers shall wear a helmet to enter the scene. Different identities of personnel required to wear a helmet with different color. In the intelligent substation, It's necessary to have a real-time monitoring of the production process construction methods, in order to meet the safety requirements. This paper presents a helmet based pedestrian detection and color recognition algorithm, according to the detected people and identifying whether the people is wearing a helmet. And depending on the color of the helmet, It can be further determined whether who is absent in the place. The algorithm shows very high accuracy, which not only reduces the time cost of artificial investigation overhead the surveillance video data, saving manpower, but also improves work efficiency.
出处
《电脑知识与技术（过刊）》
2014年第2X期1018-1020,共3页
Computer Knowledge and Technology
关键词
智能变电站
行人检测
安全帽
颜色识别
Intelligent Substation
Pedestrian Detection
Helmets
Color Recognition
