摘要
针对医院设备间安全管理的重要性,以及医疗改革的信息化需求,结合图像处理方法,提出一种目标检测与跟踪的设备间视频监控及报警系统。其中,运用YOLO网络实现设备间走廊行人的检测,然后结合多目标跟踪算法,实现行人轨迹的跟踪提取;最后通过位移差的长度判断行人是否徘徊和是否跨越警戒线,以此实现异常行为的报警。结果表明,通过以上方案,可清晰查看医院设备间走廊实时界面,同时通过测试,准确报警率可达到98%以上。由此得出,设计的医院设备间视频监控及报警系统可行。
Medical equipment is expensive and dangerous,and in order to ensure that medical equipment is not damaged and to ensure the safety of non-specialists,the entrance to the hospital equipment room and the critical area in the equipment room are the focus of monitoring and vigilance.Therefore,this paper proposes a video surveillance alarm system between hospital equipment based on intelligent video surveillance system,deeply studies the core technologies of pedestrian detection,behavior recognition,and trajectory recognition in intelligent video surveillance system,and designs an alarm system based on pedestrian target crossing the cordon and wandering and staying abnormal behavior recognition,detects and identifies pedestrians who break into the hospital equipment room and linger at the entrance of the equipment room,and issue an alarm at the first time to ensure the safety of equipment and pedestrians.Through the simulation results,it can be seen that the video surveillance alarm system between hospital equipment designed in this paper is effective and feasible.
作者
徐飞
邓亚萍
陈兴
XU Fei;DENG Yaping;CHEN Xing(Shenzhen Nanshan District Medical Group Headquarters Smart Health Department.Shenzhen,Guangdong 518052,China)
出处
《自动化与仪器仪表》
2023年第4期138-142,共5页
Automation & Instrumentation
基金
南山区2020年度卫生科技计划项目(202060)。
关键词
视频监控系统
报警系统
行人检测
行为识别
轨迹识别
video surveillance system
alarm system
pedestrian detection
behavior recognition
movement trajectory recognition
作者简介
徐飞(1978-),男,辽宁沈阳人,中级。
