云梯系统宠物乘梯预警分析数据

依据告警数据，捕捉到宠物进入、乘坐以及离开电梯的全过程，确保宠物在乘梯过程中的安全，避免宠物因不当行为而受伤；设备抓图便于在发生突发状况时，快速了解现场状况，定位电梯及乘客所在位置，对宠物入梯进行安全管理；定期对告警数据设备抓图所反映的现场情况进行分析，方便物业对宠物入梯行为进行日常监管，判断采取何种安全措施，制定更加科学、合理的宠物管理规定；根据一段时间内告警的数据规模、现场图片、时间等，帮助电梯生产销售企业研判是否需要开发定制化的电梯服务；公开报道一段时间内宠物入梯告警数据，可以提高公众对宠物入梯安全的重视程度。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、目标匹配groupid； 4.对输出的目标类型进行识别，若目标类型category=8，表示目标结果为宠物，系统会向平台发送宠物乘梯告警。

Based on alarm data, this dataset captures the entire process of pets entering, riding, and exiting elevators, to ensure the safety of pets during elevator rides and prevent them from getting injured due to improper behaviors. The images captured by the devices enable quick understanding of on-site conditions, positioning of elevators and passengers during emergencies, and implementation of safety management for pets entering elevators. Regular analysis of on-site situations reflected by alarm data and device-captured images facilitates daily supervision of pets entering elevators by property management teams, helping determine appropriate safety measures and formulate more scientific and reasonable pet management regulations. Additionally, based on metrics such as the scale of alarm data, on-site images, and timestamps over a specific period, elevator manufacturing and sales enterprises can evaluate whether to develop customized elevator services. Publishing pet elevator entry alarm data within a certain period can also enhance public awareness of pet elevator riding safety.
1. Detect elevators in operation using a sensor-based real-time elevator door opening and closing state detection algorithm.
2. Enable the in-elevator camera image capture service with a capture frequency of 1 FPS, and upload the captured images to Huawei Cloud OBS. After the cloud algorithm server pulls the images from Huawei Cloud OBS, it performs image detection via a target detection algorithm trained on deep neural networks, and outputs relevant image attributes.
3. The output image attributes include: target category, target bounding box (denoted as box), target confidence score (denoted as score), and target matching group ID (groupid).
4. Identify the target types from the output results: if the target category equals 8, indicating the detected target is a pet, the system will send a pet elevator riding alert to the platform.

搜集汇总

数据集介绍

特点

该数据集名为“云梯系统宠物乘梯预警分析数据”，由浙江新再灵科技股份有限公司自行产生，包含751条记录，每日更新。数据集通过电梯内摄像头抓图和深度学习算法检测宠物乘梯行为，用于宠物乘梯安全管理和物业监管。