Compartment congestion degree detecting method, device and system based on 3D radar scanning

Abstract

An embodiment of the invention provides a compartment congestion degree detecting method, a compartment congestion degree detecting device and a compartment congestion degree detecting system based on3D radar scanning. The compartment congestion degree detecting method comprises the steps of: acquiring point cloud data of passengers getting on and off scanned by means of 3D radars installed abovecompartment doors, acquiring two-dimensional feature data from the point cloud data, and marking target two-dimensional feature data corresponding to the passengers from the two-dimensional feature data by means of a pre-trained identification model; and tracking a walking direction of the passenger corresponding to each piece of target two-dimensional feature data, further counting numbers of people getting on and off at a current station according to the walking directions of the passengers, calculating a number of passengers in the compartments when a train leaves the station, transmittingthe number of passengers to a next station, and displaying the number of passengers in the next station. According to the compartment congestion degree detecting method, the counted number of the passengers in the compartments is displayed before the passengers arrives at the next station, such that the waiting passengers can select platforms according to the number of passengers in the compartments, thereby avoiding blindly waiting for the train, and improving the carrying capacity of the train.

Description

technical field [0001] The present invention relates to the technical field of train transport capacity, in particular to a method, device and system for detecting congestion degree of carriages based on 3D radar scanning. Background technique [0002] At present, during the operation of the train, before the train arrives at the station, the passengers on the station cannot know the number of passengers in each compartment. Thereby when waiting for a train, due to the unpredictability of the crowded situation of each compartment, one can only wait blindly. Especially during peak hours, the parking time of vehicles is short, and passengers cannot predict the degree of congestion of each train compartment, which may cause some compartments to be very crowded, while some compartments are very loose, resulting in waste of space and reducing the carrying capacity of the train. When there are too many passengers, it will also cause passengers to stay in the station to a certain ...

AI Technical Summary

Problems solved by technology

[0004] Embodiments of the present invention provide a 3D radar scanning-based compartment congestion detection method, device, and system to solve the problem that passengers cannot know

Images