Research And Implementation On The Algorithm Of Detecting The Number Of Elevator Persons Based On Images

With the rapid progress of urbanization process,the demand for construction elevators is growing,and the use of its safety is getting more and more attention.In the construction elevator's application,there is a widespread phenomenon of overcrowding or a mixture of people,which is the main reason for the frequent accidents of construction elevator.To this end,the paper studies the technique of calculating the number of elevator people based on images,and realize the algorithm on the ARM processor board.The paper discusses the target detection and tracking algorithm based on image,and the workers in the elevator can be detected by detecting the color of the helmet.The target trace is generated by the method of adjacent matching,and the number of people entering and exiting the elevator is calculated based on the detection line set near the door.Based on the construction elevator number detection hardware platform,this paper presents the process of the number detection algorithm.The program modules of image acquisition,target extraction,target trajectory processing,and gate number of people calculation are designed in detail.The real-time detection of elevator numbers is achieved on the ARM processor.Through practical testing,the results show that the design requirements of calculating the elevator people numbers are achieved.