The invention discloses a passenger behavior recognition method based on a support vector machine. The passenger behavior recognition method comprises the steps that S1, an infrared device detects the state of an elevator door; s2, monitoring the key point state of the human body by a camera; s3, key point coordinate difference values are analyzed and compared; s4, feeding back a signal to the system according to the coordinate difference value and a vector machine classification model signal processor; s5, the system receives the signals and then recognizes and processes passenger behaviors, the infrared device combined with human body key point recognition is provided for overcoming the defects of an existing elevator infrared device to recognize falling-down and other actions of people, the problem of the detection precision of the existing infrared device can be solved, and the safety of the elevator is improved. And the safety of passengers in the elevator when accidents occur is further ensured.
本发明公开了一种基于支持向量机的乘客行为识别方法,包括S1红外线装置检测电梯门状态;S2摄像头监测人体关键点状态;S3分析比对关键点坐标差值;S4根据坐标差值以及向量机分类模型信号处理器向系统反馈信号;S5系统接收信号后识别乘客行为并作出处理,针对现有电梯红外线装置的不足,提出一种结合人体关键点识别的红外线装置实现对人摔倒及其他动作的识别,能够解决现存的红外线装置检测精度的问题,进一步确保乘客在电梯内发生意外时的安全性。
Passenger behavior identification method based on support vector machine
一种基于支持向量机的乘客行为识别方法
2023-07-18
Patent
Electronic Resource
Chinese
IPC: | B66B Aufzüge , ELEVATORS |
Analyzing passenger train arrival delays with support vector regression
Elsevier | 2015
|Analyzing passenger train arrival delays with support vector regression
Online Contents | 2015
|