A method and a device for detecting passenger density abnormity in a rail transit carriage

Abstract

The invention relates to a method and a device for detecting passenger density abnormity in a rail transit carriage, the method comprises an offline training step and a real-time online detection step, and the real-time online detection step specifically comprises the following steps: collecting image frame data in a monitoring video in the current carriage in real time, and preprocessing the image frame data; Inputting the preprocessed image frame data into an acquired image recognition model to detect the number of people, obtaining the number of current carriages at the current moment, andstoring the number of current carriages at the current moment; And obtaining the current number of carriages at the previous moment, and judging whether the passenger density in the current carriage is abnormal or not according to the current number of carriages at the previous moment and the current number of carriages at the current moment. The method can detect the situation that the number ofpassengers in the carriage is suddenly increased or decreased suddenly in real time, early warns safety events appearing in the carriage, and guarantees the safety of the passengers in the carriage.

Description

technical field [0001] The invention relates to the technical field of computer vision image processing, in particular to a method and a device for detecting abnormal passenger density in rail transit carriages. Background technique [0002] Today, with the rapid development of urban rail transit, the subway has brought many conveniences to people. The subway car is a special semi-enclosed environment. When the number of people in the car increases or spreads rapidly, passenger crowding or even stampede incidents will easily occur, endangering the safety of passengers. When the train arrives at the terminal station, it is necessary to determine whether the passengers are staying in the compartment. Now the staff enters each compartment to check whether there are passengers who have not left. If there are passengers who have not left, the staff will assist them to leave. This method works Efficiency is low, and increase the residence time of train at terminal. [0003] At p...

AI Technical Summary

Problems solved by technology

The subway car is a special semi-enclosed environment. When the number of people in the car increases rapidly or spreads, it is very easy for passengers to be crowded or even stampede, which will endanger the safety of passengers.

When the train arrives at the terminal station, it is necessary to determine whether the passengers are staying in the compartment. Now the staff enters each compartment to check whether there are passengers who have not left. If there are passengers who have not left, the staff will assist them to leave. This method works Inefficient and increases train dwell time at terminal

[0003] At present, in the existing technology, there are mainly three ways to detect abnormal passenger density in subway cars. One is to calculate the passenger density by setting a pressure sensor in the car, and estimate the number of passengers by the weight of the passengers on the pressure value of the car. This method ignores The weight difference between passengers can only be estimated by the number of passengers in each car. At the same time, this method is limited by the accuracy of the pressure sensor. When the train arrives at the terminal station, it is impossible to detect a very small number of passengers staying in the car abnormally. Inside the situation

The second is to install an infrared detector in the carriage to detect the passengers getting on and off the train, and obtain the number of passengers. The infrared detector uses the optical principle, and the light is easily blocked when passengers get on and off the train, resulting in missed detection, which affects the accuracy of passenger density statistics.

The third is the method of counting the number of passengers passing through the entrance and exit gates. This method is set at the entrance or exit, and cannot be counted to obtain specific statistics on the number of people in each carriage.

Images