KR20220147951A - Apparatus and method for tracking the movement of public transportation users - Google Patents

Abstract

The present invention relates to an apparatus and method for tracking movements of public transportation users. The apparatus and method for tracking movements of public transportation users can track movements of public transportation users (or passengers) continuously and extensively, greatly reduce an image investigation time for comparing massive video images with each other, have high reliability of a route tracking result by utilizing transportation card information, departure time of public transportation, arrival time, and log data of stations or stops at the same time, and can be useful when being applied to contact freight charges that may occur due to different operating companies sharing the same system network as a metropolitan urban railway network.

Description

Apparatus and method for tracking the movement of public transportation users}

The present invention relates to an apparatus and method for tracking the movement of a public transportation user, and more particularly, it is possible to continuously and extensively track the movement of a public transportation user, shorten the tracking time, and also improve the reliability of the route tracking result. It relates to a high public transport passenger movement tracking device and method.

In general, public transportation is a means of transport that can move passengers to a destination, and a bus, urban railroad (subway), or BRT is representative. Currently, a number of routes are operated in large cities such as buses, urban railroads, or BRTs, and a transfer system is used to move between routes.

In particular, with the introduction of the transfer system, recently, rechargeable transportation cards and post-paid transportation cards are mainly used, and the transportation card records the user's boarding and disembarking station information.

Tracking the movement of passengers using public transportation is a very important issue in terms of public transportation operation and enhancement of user convenience, and the necessity is urgently required due to the use of contact fare settlement, means sharing ratio, and route-based information provision.

As prior art related to such passenger movement tracking, Patent Registration No. 10-1136366 (Reference 1), Registration Patent No. 10-1548639 (Reference 2), and Patent Registration No. 10-2074894 (Reference 3) are proposed. has been

First of all, Reference 1 relates to an object automatic tracking system, which zooms in on an object detected by a fixed camera using a dome camera and automatically tracks it through image processing. An object automatic tracking system that continues tracking by taking over is proposed. The system of Reference 1 automatically performs the pan-tilt function without the user's manual operation, and in detailed tracking using a dome camera, it can automatically respond to tracking failure or deviation from the monitoring range due to a blind spot caused by a geographical feature. , it can be used to track the movement of an unspecified person, but it is difficult to obtain additional information about the object to be tracked other than image information, so it is difficult to identify the same object in different detection areas.

Next, Reference 2 relates to an object tracking apparatus and a method for a surveillance camera system, and transforms an image of a three-dimensional space plane obtained from a plurality of cameras into an image of a two-dimensional space plane using a predefined transformation matrix, , an apparatus for synthesizing an image of a surveillance camera system and a method thereof for generating a single synthesized image by synthesizing the transformed images have been proposed. When the compressed images input through the image input unit are restored to 3D coordinate system images, the object tracking device extracts the feature points of the object of interest from the restored 3D coordinate system image and separates it from the background, weather and ambient illuminance change An object recognition unit that removes noise and tracks the state change of the image and applies the correction value to the image; a filter unit that tracks the object of interest by applying a sensitive Kalman filter; and a controller for reflecting the tracking data on the object of interest to an image synthesized according to a predetermined image synthesizing procedure of an image matching system in order to analyze the cause of the occurrence.

However, although the object tracking method of Reference 2 is suitable for a role as an image synthesis device for objects on a connected shooting section, there is no procedure for storing/confirming the identity of a moving object, so it integrates footage from non-continuous sections. so it is difficult to obtain information.

And, Reference 3 relates to a method for estimating the travel route of a public transportation user, a computer program implementing the same, and a system configured to perform the same. And from RP (Revealed preferences) data including transportation card information including ticket time data and train log information including train departure time, arrival time and station data, the OD sorting unit analyzes the traffic volume through basic statistical analysis an OD screening step of the existence of a traffic volume of extracting an existing model point and setting an analysis range therefrom to select an OD (origin and destination); Analyze the number of travel routes by OD, in which the route calculation unit extracts the distribution of the travel time of users by model point for each selected OD, and sets the range and number of alternative routes for each selected OD using the reliability index of the travel time step; a route generation step of estimating, by a single route approach time estimator, a boarding train for a single route model point among the alternative routes for each OD based on the number of the alternative routes; a single-path access time distribution estimating step of estimating, by a single-path access time estimator, a train boarding for a single-path model point among the alternative routes for each OD based on the number of alternative routes; a single-path transfer time distribution estimating step of generating, by a single-path transfer time distribution estimator, a cumulative distribution function of an approach time and a transfer time for each station from the estimated boarding train; and estimating, by a travel route estimator, a train boarding for each user for the alternative route for each OD by using the cumulative distribution function, and estimating the travel route for the multi-path model point using this.

However, the travel route estimation method of Reference 3 uses only the origin and destination data to estimate the events occurring between the terminal points, such as transfers, using the statistical cumulative distribution function, and the performance of the cumulative distribution function is The route tracking result calculated only by simulation is derived without considering the data (image data, etc.) generated from the actual train boarding position.

As a result, the prior art related to the above-mentioned user (passenger) movement tracking utilizes the data collected at the actual site by using an imaging device, but there is a limit to continuously tracking the wide range of discontinuous movement routes of customers only with the data collected by the imaging device. Therefore, it is necessary to increase the reliability of the route tracking result by using the traffic card information, the departure time, arrival time, and log data of the station or stop at the same time.

In addition, the reliability of such high route tracking results is essential for the processing of contact fares that may occur due to the existence of different operating companies, even though they share the same transportation network as the metropolitan urban rail network.

Reference 1: Registered Patent No. 10-1136366 Reference 2: Registered Patent No. 10-1548639 Reference 3: Registered Patent No. 10-2074894

Accordingly, the present invention is to solve these problems, and the present invention can continuously and extensively track the movement of public transportation users, shorten the user's tracking time, and increase the reliability of the user's route tracking results. The purpose is to provide a device and method for tracking the movement of users of public transportation that can be used.

In addition, the present invention provides an apparatus and method for tracking the movement of public transportation users that can be usefully utilized for contact fare settlement processing that may occur due to the existence of different operating companies although sharing the same transportation network as the metropolitan urban rail network. There is a purpose.

The present invention in order to solve this technical problem;

A collection unit that collects traffic-related data related to the operation of public transportation and log data generated when a user uses public transportation, an optical unit installed at a stop where public transportation stops to take an image, and an image captured by the optical unit A storage unit for classifying and storing files by element, and linking the classified and stored image files in the storage unit with the user log data of the bus stop confirmed by the collection unit, integrating the information of the user into each image file to create an information integrated image file Public transportation, characterized in that it comprises a linkage unit for generating and storing, and an analysis unit for tracking the movement of the public transport user using the traffic-related data identified in the collecting unit and the information integrated image file collected in the linkage unit. Provides a device for tracking the user's movement.

In this case, the traffic-related data includes the daily number of passengers getting on and off at each stop where public transportation stops, a list of buses by route, a list of buses stopped by each stop, and data of corresponding stop, arrival and departure times of each bus; The log data is characterized in that it includes information on arrival and departure of a stop through a transportation card or other payment means of a user using public transportation.

In addition, the storage unit stores the image files photographed by the optical unit by classifying them by photographing location or time; The linking unit links the image files organized in the storage and the stop arrival and departure information, which is log data of the stop users checked in the collection unit, with the same tag, shooting time, same tag, and shooting location data to each image file of the user. It is characterized in that the information is integrated to generate and store the information-integrated image file.

In addition, the analysis unit is characterized in that the travel-related data identified in the collection unit and the information integrated image file collected in the link unit are used to identify the movement of the public transportation user by using the Person Re-ID method.

In this case, the Person Re-ID method is characterized in that it grasps the movement of the public transportation user based on deep learning.

In addition, the analysis unit searches in the order of the public transportation that arrives early after the public transportation user arrives at the stop, and searches in the order of the major stops with the most daily average number of people getting off among the stops in the direction where the public transportation user moves, which is the analysis target. .

In addition, the present invention;

A first step of collecting traffic-related data related to the operation of public transportation and log data generated when the user uses public transportation; a second step of classifying and storing the image files photographed at the stop where the public transport stops by elements; a third step of generating and storing an integrated information image file by linking the classified and stored image file for each element with the user log data of the bus stop; and a fourth step of tracing the movement of the public transportation user by using the traffic-related data and the information-integrated image file.

In this case, the fourth step is a step of tracing the movement of a public transportation user by using the Person Re-ID method with the traffic-related data and the information-integrated image file.

And, the Person Re-ID method is characterized in that it grasps the movement of public transportation users based on deep learning.

According to the present invention, it is possible to track a continuous and extensive movement path compared to the existing public transportation user (or passenger) movement line tracking method, and it is possible to significantly reduce the image search time for comparing vast video images with each other.

Further, according to the present invention, the reliability of the route tracking result is high by simultaneously utilizing the traffic card information, the departure time, arrival time, and log data of the station or stop.

In addition, the public transportation user (or passenger) movement tracking method according to the present invention is useful when applied to contact fare settlement that may occur due to the existence of different operating companies although sharing the same network as the metropolitan urban rail network.

1 is a view for explaining a technique for tracking the movement of public transportation users according to the present invention.
Figure 2 is a block diagram of a device for tracking the movement of public transportation users according to the present invention.
3 is an overall analysis flow chart for tracking the movement of public transportation users according to the present invention.
4 is a set and input definitions shown in the overall analysis flow chart of FIG. 3 .

Hereinafter, it will be possible to understand the features of the embodiment described in detail with reference to the accompanying drawings of the public transportation user's movement tracking device and method according to the present invention.

Prior to this, the terms or words used in the present specification and claims should not be construed as being limited to conventional or dictionary meanings, and the inventor should properly understand the concept of the term in order to best describe his invention. Based on the principle that it can be defined, it should be interpreted as meaning and concept consistent with the technical idea of the present invention.

Accordingly, the embodiments described in this specification and the configurations shown in the drawings are only the most preferred embodiment of the present invention, and do not represent all of the technical spirit of the present invention, so at the time of the present application, they can be replaced It should be understood that various equivalents and modifications may exist.

1 and 2, in the method of tracking the movement of public transportation users according to the present invention, a camera is installed at the entrance and exit of the stop 10 at which the public transportation 1 stops, such as a bus, an urban railroad (subway), or a BRT. By attaching it, it is possible to track whether the user (A, B, C, D) is the same person by using the biometric image (photo) of the public transportation user (A, B, C, D) to trace the entire movement of the same person. .

At this time, each photo of users (A, B, C, D) taken through a plurality of cameras installed at the entrance and exit of the stop 10 is a head, not a face, in order to minimize illegal elements such as unauthorized collection of personal biometric information. The movement is traced using the Person Re-ID (Person Re-Identification) method, which identifies whether the person captured in different places is the same person through appearance factors such as color, hair shape, clothes, and bag.

That is, the present invention can quickly analyze and track the movement of a public transportation user by utilizing the Person Re-ID technology, which is useful in analyzing CCTV images, which is widely used to identify the movement of a moving object.

The device for tracking the movement of a public transportation user according to the present invention includes a collection unit 100 that collects travel-related data related to the operation of public transportation and log data generated when the user uses public transportation, and the public transportation 1 stops An optical unit 200 installed at a stop 10 to take an image, a storage unit 300 for classifying and storing the image files photographed in the optical unit 200 by element, and the storage unit 300 Linking unit 400 for creating and storing an integrated information image file by linking the classified and stored image file in the collecting unit 100 and the user log data of the stop 10 confirmed by the collecting unit 100 and integrating the user's information into each image file ) and the traffic-related data identified by the collection unit 100 and the information-integrated image file collected by the linkage unit 400 using the Person Re-ID (Person Re-Identification) method to determine the movement of public transportation users. and an analysis unit 500 for tracking.

Hereinafter, the constitution of each part of the present invention will be described in detail.

The collection unit 100 collects two types of data, namely, traffic-related data related to the operation of public transportation and log data generated when a user uses public transportation.

Priority traffic-related data is data necessary for the operation of public transportation. The daily number of passengers getting on and off at each stop where the public transportation (1) stops, a list of buses by route, a list of buses stopped by each stop, and the corresponding stops, arrivals and departures of each bus. time data and the like.

In addition, the log data includes information on arrival and departure of a stop through a user's transportation card or other payment means using public transportation (1) such as a bus, urban railroad (subway), or BRT.

In addition, the optical unit 200 includes an entrance (eg, a gate installation area) and an exit (eg, a gate installation area) of each stop 10 where the public transportation 1 such as a bus, urban railroad (subway) or BRT stops and the user gets on or off. For example, it may be composed of a camera 201 that takes an image in the vicinity of the screen door installation area). These cameras 201 take pictures when the user of each stop 10 passes through the entrance (eg, when performing an operation such as a tag at the gate) or when passing through the exit (eg, when entering the inside of the screen door) can proceed.

In addition, the storage unit 300 classifies and stores the image files photographed by the optical unit 200 for each element such as a photographing location or time.

On the other hand, the linkage unit 400 includes the image files organized in the storage unit 300 before entering the analysis unit 500 and the stop arrival and departure information (log data) of the stop user checked in the collection unit 100 . by using the same tag, shooting time, same tag, and shooting location data, and integrating the user's information into each image file to create and store an information-integrated image file.

In this case, it is preferable not to use personal information such as a user's name and address that can be identified through a transportation card or other payment means in the information integrated image file for the purpose of protecting personal information.

In addition, the analysis unit 500 uses the travel-related data identified by the collection unit 100 and the integrated information image file collected by the linkage unit 400 to identify the movement of the public transportation user.

At this time, since the analysis unit 500 has to compare vast images with each other due to the nature of public transportation with a large number of daily users, the present invention shortens the image search time in the following two ways.

First, a public transport user, an analysis target, arrives at the stop 10 and then searches for public transport (1) such as a bus (or train) that came in early.

And, among the stops in the direction to which the public transportation users move, the analysis target, the main stops 10 with the most daily average number of people getting off are searched in order.

At this time, the analysis unit 500 uses the Person Re-ID (Person Re-Identification) method for the integrated information image file to identify the movement of the public transportation user, and the Person Re-ID procedure is performed by Convolution Neural Networks (CNN), etc. based on deep learning of

Hereinafter, an example of tracking the movement of a public transportation user according to the present invention will be described with reference to FIGS. 1 to 4 .

First of all, travel-related data such as the number of daily boarding and disembarking at a stop required for the operation of public transportation (1), a list of buses by route, a list of buses stopped by each stop, and data on the corresponding stop, arrival and departure times of each bus, and when users use public transportation Log data such as generated stop departure and arrival information is collected through the collection unit 100 .

Next, a biometric image (photo) of a public transportation user is taken through a plurality of cameras 201 that are optical units 200 installed at the entrance and exit of the stop 10, and the photographed image file is recorded at the shooting location or time It is stored in the storage unit 300 by classifying it by elements, such as.

In addition, the image file organized in the storage unit 300 and the stop arrival and departure information, which is log data of the stop user checked in the collection unit 100, are linked with the same tag, the shooting time, the same tag and the shooting location data to the linking unit 400 ) to create and store an integrated information video file by integrating the user's information.

After that, the analysis unit 500 uses the information-integrated image file collected by the collection unit 100 and the travel-related data identified by the collection unit 100 in the analysis unit 500 to minimize illegal elements such as unauthorized collection of personal biometric information. It uses the Person Re-ID (Person Re-Identification) method to determine whether a person captured in different places is the same person through appearance factors such as hair color, hair shape, clothes, and bag to track the movement of public transportation users. .

At this time, the analysis target public transportation user arrives at the stop 10 and then searches for the public transportation (1) that came in early in order, and the main stop ( 10) Search in order to shorten the image search time.

The overall analysis process of such a method of tracking the entire movement of the public transportation user is shown in FIG. 3 , and the description of the variables of the process shown in FIG. 3 is as shown in FIG. 4 .

The method of tracking the movement of public transportation users according to the present invention is utilized as a passenger tracking technique for buses (BRT or Super BRT) running on a dedicated section, or passengers of a train running on a track section (including light rails such as general trains, weights and trams) It can be used as a tracking technique, used as a passenger tracking technique for public transportation that applies a pre-fare system such as a regular bus, or, if necessary, can be applied in various locations other than the station where it is needed.

In addition, the method of tracking the movement of a public transportation user according to the present invention extracts only a part of the collected data presented and tracks and evaluates the movement of a subject to be analyzed among public transportation users in the same way, or each component of the collection unit 100, the optical unit ( 200), the storage unit 300, the linkage unit 400, and the analysis unit 500 can be displayed by tracing and analyzing the movement of the public transportation user in a converged manner by integrating and closing any one or more.

In particular, the analysis unit 500 may change the type of the deep learning method for Person Re-ID presented as an example, or change the shape of the flowchart of the analysis unit 500 , but the same principle may be applied.

Summing up the method of tracking the movement of a public transportation user according to the present invention as described above, the first step of collecting traffic-related data related to the operation of public transportation and log data generated when the user uses public transportation, and the stop at which the public transportation stops The second step of classifying and storing the image file photographed in (10) for each element, and the second step of linking the image file classified and stored for each element and the user log data of the bus stop 10 to create and store an integrated information image file Step 3 and the fourth step of tracing the movement of the public transportation user by using the traffic-related data and the information integrated image file.

Here, the fourth step consists of a step of tracing the movement of a public transportation user by using the Person Re-ID method for the traffic-related data and the information-integrated image file.

And, the Person Re-ID method identifies the movement of public transportation users based on deep learning.

The above description is merely illustrative of the technical idea of the present invention, and various modifications and variations are possible without departing from the essential characteristics of the present invention by those skilled in the art to which the present invention pertains. The scope of protection should be interpreted by the following claims, and all technical ideas within the equivalent range should be construed as being included in the scope of the present invention.

1: Public transportation 10: Bus stop
100: collection unit 200: optical unit
300: storage unit 400: linkage unit
500: analysis unit

Claims (9)

A collection unit that collects traffic-related data related to the operation of public transportation and log data generated when a user uses public transportation, an optical unit installed at a stop where public transportation stops to take an image, and an image captured by the optical unit A storage unit for classifying and storing files by element, and linking the classified and stored image files in the storage unit with the user log data of the bus stop confirmed by the collection unit, integrating the information of the user into each image file to create an information integrated image file Public transportation, characterized in that it comprises a linkage unit for generating and storing, and an analysis unit for tracking the movement of the public transport user using the traffic-related data identified in the collecting unit and the information integrated image file collected in the linkage unit. User's movement tracking device.
The method of claim 1,
the traffic-related data includes daily number of passengers getting on and off at each stop where public transportation stops, a list of buses by route, a list of buses stopped by each stop, and data of corresponding stop, arrival, and departure times of each bus;
The log data is a device for tracking the movement of a user of public transportation, characterized in that it includes information on arrival and departure from a stop through a transportation card or other payment means of the user using public transportation.
The method of claim 1,
The storage unit stores the image files photographed by the optical unit by classifying them by photographing location or time;
The linking unit links the image files organized in the storage and the stop arrival and departure information, which is log data of the stop users checked in the collection unit, with the same tag, shooting time, same tag, and shooting location data to each image file of the user. A device for tracking the movement of public transportation users, characterized in that information is integrated to create and store an information integrated image file.
The method of claim 1,
The analysis unit uses the Person Re-ID method for the traffic-related data identified by the collection unit and the information integrated image file collected from the link unit to identify the movement of the public transportation user. tracking device.
5. The method of claim 4,
The Person Re-ID method is a device for tracking the movement of public transportation users, characterized in that it identifies the movement of public transportation users based on deep learning.
The method of claim 1,
The analysis unit searches in the order of the public transportation that arrives early after the public transportation user arrives at the stop, and searches in the order of the major stops with the most daily average number of people getting off among the stops in the direction where the public transportation user moves, the analysis target. User's movement tracking device.
A first step of collecting traffic-related data related to the operation of public transportation and log data generated when the user uses public transportation;
a second step of classifying and storing the image files photographed at the stop where the public transportation stops by elements;
a third step of generating and storing an integrated information image file by linking the classified and stored image file for each element with the user log data of the bus stop; and
A fourth step of tracing the movement of the public transportation user by using the traffic-related data and the information integrated image file;
8. The method of claim 7,
The fourth step is a method of tracking the movement of a public transportation user, characterized in that it is a step of tracing the movement of the public transportation user by using the Person Re-ID method with the passage-related data and the information-integrated image file.
9. The method of claim 8,
The Person Re-ID method is a method of tracking the movement of a public transportation user, characterized in that it grasps the movement of the public transportation user based on deep learning.

Images