云梯系统电瓶车入梯分析数据

依据对电瓶车入梯行为的监测，及时在轿厢内提醒乘客安全乘梯，有效防止电瓶车在电梯内引发火灾或其他安全事故；设备抓图便于在发生突发状况时，快速了解现场状况，定位电梯及电瓶车所在位置，对电梯内居民的安全起到实际保障作用；定期对告警数据设备抓图所反映的现场情况进行分析，方便物业对电瓶车入梯行为进行日常监管，合理地分配人力和物力资源，提前采取措施，预防电瓶车违规进入电梯导致的安全事故，保障居民安全，提升居民对物业服务的满意度和信任度；根据一段时间内告警的数据规模、现场图片、时间等，公开报道一段时间内电瓶车入梯告警数据，可以提高公众对电瓶车入梯危险行为的重视程度。
1.通过基于传感器的实时电梯门开关状态检测算法检出处于运行状态的电梯； 2.开启电梯内摄像头抓图服务(抓图频率1FPS),并将图片上传到华为云obs。云端算法服务器从华为云obs拉取到图片后，通过基于深度学习神经网络训练的目标检测算法进行图片检测，输出图片相关属性；3.输出的图片属性包含：目标类型category、目标坐标框box、目标结果置信度score； 4.对输出的目标类型进行识别，若目标类型category=0，表示目标结果为电瓶车，系统会向平台发送电瓶车入梯告警，告警时间为时间戳，更加精准。

Based on the monitoring of electric bicycle entry into elevators, timely reminders are provided to passengers in the elevator car for safe elevator use, effectively preventing fires or other safety accidents caused by electric bicycles inside elevators. The images captured by the equipment enable quick understanding of the on-site situation and location of both the elevator and electric bicycles during emergencies, effectively ensuring the safety of residents in the elevator. Regular analysis of the on-site situation reflected by alarm data and captured images facilitates the daily supervision of electric bicycle entry into elevators by property management, reasonably allocates human and material resources, takes proactive measures to prevent safety accidents caused by electric bicycles illegally entering elevators, guarantees resident safety, and improves residents' satisfaction and trust in property management services. Furthermore, publicly releasing the electric bicycle entry alarm data (including alarm scale, on-site images, and timestamps) over a certain period can enhance public awareness of the dangerous behavior of electric bicycles entering elevators.
1. Detect operating elevators using a real-time elevator door switch state detection algorithm based on sensors;
2. Enable the in-elevator camera image capture service (with a capture frequency of 1 FPS) and upload the images to Huawei Cloud OBS. After pulling the images from Huawei Cloud OBS, the cloud-based algorithm server performs image detection via an object detection algorithm trained on deep learning neural networks, and outputs relevant image attributes;
3. The output image attributes include: target category, target bounding box, and target confidence score;
4. Identify the output target category. If the target category equals 0, indicating that the detected target is an electric bicycle, the system will send an electric bicycle entry alarm to the platform, with the alarm time recorded as a timestamp for higher accuracy.