The invention discloses an elevator door lock meshing depth detection system based on three-dimensional point cloud deep learning. The method comprises the steps that S1, elevator door lock meshing feature point cloud data are collected and preprocessed; s2, carrying out point cloud anti-normalization processing on the preprocessed feature point cloud data, and extracting a point cloud boundary; s3, according to given point coordinates of the point cloud boundary, the relative distance between the two points, namely the engagement depth of the door lock, is obtained; and S4, judging whether the meshing depth of the door lock meets the standard meshing depth or not according to the meshing depth of the door lock, if not, giving an alarm, and sending the landing door lock meshing depth information to a control room in a control room. A point cloud boundary is extracted by using a point cloud boundary extraction method, and point cloud coordinates of a lock hook and a hook block are extracted to calculate the engagement depth of the elevator door lock, so that the engagement depth of the elevator door lock is monitored, and the safety risk during manual detection of the engagement depth of the door lock is reduced; and by adopting the optimized point cloud data processing method, the detection efficiency and accuracy are greatly improved.
本申请公开了基于三维点云深度学习的电梯门锁啮合深度检测系统,方法包括:S1.采集电梯门锁啮合特征点云数据并进行预处理;S2.对预处理后的特征点云数据进行点云反归一化处理,提取点云边界;S3.根据点云边界的给定点坐标,得到两点的相对距离,即门锁的啮合深度;S4根据门锁的啮合深度判断是否符合标准啮合深度,若不符合,发出警报,并将层门门锁啮合深度信息发送至控制室内的控制室中。利用点云边界提取方法提取点云边界并提取锁钩和钩挡的点云坐标计算电梯门锁啮合深度的计算,实现对电梯门锁啮合深度的监测,降低了人工对门锁啮合深度检测时的安全风险,采用优化过后的点云数据处理方法还极大地提高了检测效率与准确度。
Elevator door lock engagement depth detection system based on three-dimensional point cloud deep learning
基于三维点云深度学习的电梯门锁啮合深度检测系统
05.09.2023
Patent
Elektronische Ressource
Chinesisch
Hand-held elevator door lock engagement depth detection equipment and detection method
Europäisches Patentamt | 2024
|Europäisches Patentamt | 2018
|Europäisches Patentamt | 2023
|