The invention discloses an artificial intelligence inspection and detection method for an elevator, which comprises the following steps of: capturing a specific action process through a camera, and then extracting different feature information from the specific action process by using different algorithms; using machine learning or deep learning to identify and judge action processes related to the extracted different feature information; the machine learning or the deep learning is a convolutional neural network (CNN) and a recurrent neural network (RNN); and finally, outputting a judgment result. According to the method, feature information extraction is carried out on the action process of elevator dynamic inspection and detection items, then recognition and judgment are carried out, and finally wanted judgment information is output. For different action processes, different pertinent algorithms need to be developed. For different action processes, due to different algorithms, the extracted feature information is different.
本发明公开了一种电梯人工智能检验检测方法,它是通过摄像头捕捉特定的动作过程,然后使用不同的算法从中提取不同的特征信息;使用机器学习或者深度学习,对上述提取的不同特征信息涉及到的动作过程进行识别和判定;所述的机器学习或者深度学习,为卷积神经网络(CNN)和循环神经网络(RNN);最后输出判定结果。本发明通过对涉及电梯动态检验检测项目的动作过程进行特征信息提取,然后进行识别和判定,最后输出想要的判定信息。对于不同的动作过程,需要开发针对性的不同算法。不同的动作过程,由于算法不同,所提取的特征信息是不同的。
Artificial intelligence inspection and detection method for elevator
一种电梯人工智能检验检测方法
2024-05-03
Patent
Electronic Resource
Chinese
Artificial intelligence inspection and detection method for elevator
European Patent Office | 2024
|Artificial intelligence inspection and detection device for elevator shaft data
European Patent Office | 2024
|Artificial intelligence inspection and detection device for elevator machine room data
European Patent Office | 2024
|Artificial intelligence inspection and detection device for comprehensive data of elevator landing
European Patent Office | 2024
|