JP6712010B1 - Drop-off station information system and drop-off station information program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To efficiently and accurately identify and predict a passenger's exit station with a simple configuration. In a disembarking station information system S including a terminal device 1 owned by a user U who is on a train, the terminal device 1 includes a photographing unit 12, and photographing inside the train photographed by the photographing unit 12 is performed. Determination means for determining that the passenger P in the train has stood based on the image, and position information acquisition means for acquiring position information of the user U when the determination means determines that the passenger P has stood. Based on the exit station specifying means for specifying the exit station of the passenger P based on the position information of the user U acquired by the position information acquiring means, and the information of the exit station of the passenger P specified by the exit station specifying means in the past. Then, the dismounting station predicting means for predicting the dismounting station of the passenger P currently on the train is provided. [Selection diagram] Fig. 16

Description

The present invention relates to an alighting station information system and an alighting station information program capable of providing information on a passenger's alighting station.

In Patent Document 1 and Patent Document 2, it is possible to provide a system that can provide information about seats that are scheduled to become vacant in a train, and a passenger who is standing can know at an early stage which position is likely to become vacant. Various systems are disclosed.

Japanese Patent Laid-Open No. 2009-175953 JP, 2009-301499, A

However, the system disclosed in the above document specifies a passenger's exit station based on commuter pass information and boarding history information recorded on an IC card or the like, and the passenger can use such an IC card or the like. If you don't have it, it won't work.
In addition, it is necessary to install a reading device such as an IC card in each seat, which is very expensive.
If the IC card or the like is covered with a shield, it may not be possible to retrieve the commuter pass information and the boarding history information.

In order to achieve the above object, the disembarkation station information system of the present invention is a disembarkation station information system including a terminal device owned by a user who is in a train, wherein the terminal device includes a photographing unit, and the photographing unit. Determination means for determining that a passenger in the train has stood on the basis of a captured image in the train captured by, and the position of the user when the determination means determines that the passenger has stood A position information acquisition unit that acquires information, an exit station identification unit that identifies an exit station of the passenger based on the position information of the user acquired by the location information acquisition unit, and an exit station identification unit in the past. An exit station predicting unit that predicts an exit station of the passenger currently on the train based on the information of the specified exit station of the passenger.

According to the present invention, it is possible to efficiently and accurately identify and predict a passenger's exit station with a simple configuration.

It is an external view of a spectacles type terminal which is an example of a terminal device which constitutes the alighting station information system of the present invention. (A) is a figure which shows the case where a terminal device is comprised by several apparatuses, (b) is a figure which shows the case where an alighting station information system is comprised by several apparatuses. It is a block diagram which shows the structure of a terminal device. It is an example of station information. It is a figure which shows the usage mode of a terminal device. In the disembarkation station identification process, when the terminal device is used in the manner shown in FIG. 5, it is an object visually recognized through the lens at a certain time point t1, and a photographed image photographed by the photographing unit is schematically shown. FIG. In the disembarkation station identification process, when the terminal device is used in the manner shown in FIG. 5, it is an object visually recognized through the lens at a time point t2 after t1, and a captured image captured by the image capturing unit is displayed. It is a figure which shows typically. It is a figure for demonstrating the determination method of the passenger having stood. It is a figure for demonstrating the identification method of the alighting station in case the position of a user when a passenger stands up corresponds with the position of a station. It is a figure for demonstrating the specification method of the exit station in case the position of the user when a passenger stands is between stations. It is a chart showing the 1st example of getting off station DB. It is a chart showing the 2nd example of getting off station DB. FIG. 6 is a diagram schematically illustrating a captured image captured by a capturing unit, which is an object visually recognized through a lens when the terminal device is used in the mode illustrated in FIG. It is a display example of prediction information of an alighting station, (a) is an example of displaying a station name of a predicted alighting station, and (b) is an example of displaying the number of stations to a predicted alighting station for each range. is there. It is a display example of prediction information of an alighting station, and is an example in which the predicted number of stations to the alighting station is displayed in different colors for each range. It is a flow chart which shows a specification method of an alighting station (a registration method of an alighting station DB). It is a flowchart which shows the prediction method of an alighting station.

An embodiment of an alighting station information system S of the present invention will be described with reference to the drawings.
As shown in FIG. 1, the getting-off station information system S of the present embodiment can be configured by a glasses-type wearable terminal which is an example of the terminal device 1 of the present invention.
The terminal device 1 can also be applied to other devices such as a smartphone.
Further, as shown in FIG. 2A, the terminal device 1 is not limited to a single unit, and is configured by a combination of a plurality of devices, such as a configuration in which a glasses-type wearable terminal and a smartphone can be connected via a communication line. You can also do it.
Further, as shown in FIG. 2B, the disembarkation station information system S of the present invention is configured by a combination of a plurality of devices such as a configuration in which the terminal device 1 and the server device 2 can be connected via the communication line 3. You can also do it.
Hereinafter, the getting-off station information system S configured by a single glasses-type wearable terminal as the terminal device 1 will be described.

The terminal device 1 is assumed to be used by the user U while wearing it on the train while being worn by the user U.
FIG. 3 is a block diagram showing the configuration of the terminal device 1.
As shown in the figure, the terminal device 1 is configured to include a processor 11, a photographing unit 12, a storage unit 13, a display unit 14, a positioning unit 15, an operation unit 16, and a communication unit 17.
The processor 11 includes a CPU and executes various control processes by reading a program stored in the ROM.
The program also includes an alighting station information program capable of executing a specific operation and a prediction operation of the alighting station described later.

The image capturing unit 12 is an image capturing unit (camera) capable of capturing a still image or a moving image.
The image capturing unit 12 is provided, for example, in the bridge portion (see FIG. 1 ), and allows the user U to capture a still image or a moving image in the image capturing range corresponding to the field of view of the user U when the user U is wearing the terminal device 1. Is becoming
The storage unit 13 is a storage unit such as a memory or a storage that stores various information.
The storage unit 13 stores information about still images and moving images captured by the image capturing unit 12, station information described below, and an exit station DB.
The display unit 14 is a display unit that displays various kinds of information, and includes a projection unit 141 and a lens 142 that also functions as a projection target (see FIG. 1 ).
The lens 142 is made of a translucent member such as glass or plastic.
Therefore, the user U can visually recognize an actual person or object included in the visual field through the lens 142 as well as normal glasses, and can also visually recognize various information displayed on the lens 142. Yes (optical see-through method).
At this time, it is possible to link with an actual person (for example, the head) visually recognized through the lens 142 and display the information related to the person. As this display method, a known method is adopted.
Further, it is also possible to superimpose and display the photographed image photographed by the photographing unit 12 and the information to be displayed (video see-through method).
At this time, it is possible to display the image of the person included in the captured image in association with the information about the person. As this display method, a known method is adopted.
Further, it may be a non-transmissive type instead of a transmissive type, or a hybrid type of these. Further, either a single-eye system or a binocular system may be used.
Further, a separate body other than the lens 142 can be used as the projection target. For example, a virtual image projection method that uses a half mirror or a retinal projection method that directly projects on the retina of the eye of the user U can be adopted.

The positioning unit 15 is a positioning unit capable of measuring the position of the terminal device 1.
The positioning unit 15 is specifically a known satellite positioning unit such as GPS or a quasi-zenith satellite.
According to the quasi-zenith satellite positioning system, it is possible to position the current position with high accuracy within an error range of about several centimeters.
Also, a known positioning method using LTE, Wi-Fi (registered trademark), bluetooth (registered trademark), or the like can be adopted.
According to this positioning method, positioning can be performed even when the user U is located in a place where satellite positioning is difficult, such as indoors or underground where satellite waves do not reach. Therefore, for example, even when the user U is on the subway The position of the person U can be measured.
Since the terminal device 1 is mounted on and used by the user U, the positioning unit 15 is also a positioning unit that measures the position of the user U.

The operation unit 16 is an operation unit that allows the user U to perform various operations manually.
The communication unit 17 is a communication unit for performing bidirectional communication with another device, and transmits and receives various information between another terminal device (for example, a smartphone) and a server (server device 2 described later). can do.
The communication unit 17 supports various communication standards and communication systems such as bluetooth, Wi-Fi, LTE (4G, 5G, etc.).

Station information and alighting station DB stored in the storage unit 13 will be described.
The station information is position information of each station on each route and is stored in the storage unit 13 in advance.
For example, FIG. 4 shows station information of the Ueno Tokyo line (Utsunomiya line system), which is composed of position information (latitude/longitude) of each station of the Ueno Tokyo line (Utsunomiya line system).
Station information can be used for various lines such as JR, various private railways, and subways.
Further, as such station information, information stored in an external storage device or a server device can be used.
For example, necessary station information (station information of the route on which the user U board) is acquired from a Web server that stores station information of various routes via the Web when necessary, or station information previously acquired from the Web server is acquired. It can also be stored in the storage unit 13.

The getting-off station DB is data including the getting-off history of each passenger P, and is referred to in the getting-off station prediction described later.
The getting-off history is past information (history) of the getting-off station specified by the method described later.

(Specification of drop-off station and registration of drop-off station DB)
Hereinafter, the specification processing of the disembarking station and the registration processing of the disembarking station DB will be described.
FIG. 5 is a diagram showing a usage mode of the terminal device 1.
As shown in the figure, the terminal device 1 is assumed to be used by a user U who is on a train and is mounted in front of a passenger P seated in a seat.
Here, when the disembarking station information program is activated by a predetermined operation of the user U, the photographing unit 12 starts photographing.
The terminal device 1 may be attached and the alighting station information program may be started before boarding.
Specifically, the image capturing unit 12 captures a still image (photograph) or captures a moving image (including time-lapse photography) at regular intervals.

FIG. 6 is an object visually recognized through the lens 142 at a certain time point t1 when the terminal device 1 is used in the mode shown in FIG. 5, and schematically shows a captured image captured by the imaging unit 12. FIG.
Further, FIG. 7 is an object visually recognized through the lens 142 at a time point t2 after t1 when the terminal device 1 is used in the mode shown in FIG. It is a figure which shows an image typically.
As shown in these figures, the captured images include captured images of the passengers P1 to P3 who are sitting on the seats.
Here, in the terminal device 1, the processor 11 determines that the passenger P in the train has stood up based on the captured image in the train captured by the capturing unit 12 (determination unit).
For example, by comparing the captured image of FIG. 6 with the captured image of FIG. 7, it can be seen that the passenger P2 was sitting at least at time t1, but stood at time t2.
8A is a captured image of the passenger P2 in FIG. 6, and FIG. 8B is a captured image of the passenger P2 in FIG. 7.
As can be seen by comparing (a) and (b) of FIG. 8, based on the fact that the height (for example, eye height) of the head (face) of the passenger P2 has changed by a predetermined value h or more. It can be determined that the passenger P2 has stood.
Based on such a viewpoint, the determination means is based on a known image analysis method, and is based on the fact that the height of the head (face) of the passenger P or a predetermined portion of the head becomes higher than the predetermined value h. It is determined that the passenger P has stood.

The processor 11 acquires the position information of the "user U" when it is determined by the determination means that the "passenger P" has stood (positional information acquisition means).
Specifically, the processor 11 measures the position information of the user U when the height of the head of the passenger P or a predetermined portion of the head becomes higher than a predetermined value by the positioning unit 15 based on the captured image. The position information of the terminal device 1 owned by the user U is measured and acquired.
The reason why the position information acquisition means acquires the position information of the "user U" instead of the position information of the "passenger P" will be described.
Since the position information acquired by the position information means is used to identify the disembarking station of the "passenger P", the position information is the position information of the "passenger P" or the position information of the train on which the "passenger P" gets on. Is preferred.
However, it is generally difficult to acquire such position information.
For example, the position information of the passenger P can be acquired by a smartphone with a GPS function possessed by the passenger P, but this position information is data that is normally available only to the passenger P and must be acquired by a user U who is another person. Is difficult
In addition, it is considered that the position information of the train is acquired by the positioning means provided on the train on some routes, and it is not considered that the positioning means is provided on all the routes. Even if positioning is possible, it is difficult for the user U to acquire such position information as in the case of the position information of the passenger P.
Therefore, in the present invention, focusing on the fact that the "passenger P" and the "user U" are on the same "train", the "user U" is used instead of the position information of the "passenger P" and the "train". Position information (position information of the terminal device 1 owned by the user U) is acquired, and this is used to identify the alighting station.

The processor 11 identifies the alighting station of the "passenger P" based on the position information of the "user U" acquired by the position information acquiring means (alighting station specifying means).
Hereinafter, a method of identifying the getting-off station of the passenger P based on a plurality of position information (first position and second position) will be described.

First, as shown in FIG. 9, a case will be described in which the position (first position) of the user U when it is determined that the passenger P has stood up matches the position of the station.
Specifically, the position information for the first position is "35°56'55.5s north latitude, 139°38'25s east longitude", and that position is "Higashiomiya station" on the Ueno Tokyo line (Utsunomiya line system). The case where the positions match (including the case where they substantially match) will be described.
In this case, it is possible to identify (estimate) that the passenger P's exit station is Higashi-Omiya Station.
This is because when the seated passenger P gets off, it often happens that he/she arrives at the getting-off station, and the position when the passenger P stands up coincides with any station. If there is, the passenger P is likely to get off at the station.
In particular, if the passenger P wants to sit as long as possible until he/she arrives at the exit station, sits near the entrance/exit door, or is asleep, etc. Often.
The passenger P may stand for a purpose other than getting off the vehicle, but there are many cases for the purpose of getting off the vehicle, which can be proved empirically and statistically.
In this way, when the position when the passenger P stands up and the position of the station match, it is possible to specify that the station is the exit station.

However, in general, the disembarking station differs depending on the moving direction (up/down) of the train, so it is necessary to specify the moving direction of the train.
Therefore, for example, position information of the position (second position) of the user U after a predetermined time has elapsed (for example, several seconds) from the time when the passenger P stood up is acquired, and based on the first position and the second position. Identify the direction of travel for the train.
For example, as shown in FIG. 9, since the position information of the “second position” is “35°56′34.3 seconds north latitude, 139°38′16.3 seconds east longitude”, there is a difference between Higashiomiya station and Toro station. It is assumed to be the position between stations.
In this case, since the user U is moving in the direction from Higashi-Omiya Station to Toro Station, it is possible to specify that the moving direction of the train is "up."
Therefore, in this case, the passenger P can specify that he/she was on the up train of the Ueno Tokyo line (Utsunomiya line system) and that the alighting station was Higashiomiya station.

In addition, it is preferable to acquire the position (third position) of the user U a predetermined time before the passenger P stands up.
Thereby, the moving speeds of the respective sections of the third position to the first position and the first position to the second position can be obtained based on the position information of the first to third positions and the predetermined time, and the first position At, it can be determined whether the train is moving.
If it is determined that the passenger P is moving, the station is not specified as the exit station even if the position of the passenger P when the passenger P stands and the position of the station match.
In this way, when the passenger P stands at the timing when a limited express train or the like other than each station train passes the passing station, it is possible to prevent the passing station from being erroneously specified as the alighting station.

Next, as shown in FIG. 10, a case where the position (first position) of the user U when it is determined that the passenger P has stood is between stations will be described.
Specifically, the position information of the first position is "35°57'46.5"N 139°38'32.7"E"(35°57'46.5 seconds north latitude, 139°38'32.7 seconds east longitude), and its position A case where there is a station between "Hasuda Station" and "Higashiomiya Station" on the Ueno Tokyo Line (Utsunomiya Line system) will be described.
In this case, it is possible to identify (estimate) that the passenger P's exit station is Hasuda Station or Higashiomiya Station.
This is because the passenger P prepares for getting off when approaching the getting-off station, but the preparation is after the departure (passing) of the station in front of the getting-off station at the earliest, and the station in front of the getting-off station ( This is because it is extremely rare to stand up in preparation for getting off before passing.
Further, it is physically impossible for the passenger P to get off at the alighting station after passing through the alighting station.
Then, the seated passenger P stands at the time of getting off the train except when the train is located at the getting-off station (that is, except in the case of FIG. 9), when the train arrives from the station in front of the getting-off station to the getting-off station. This is the case when it is located between the stations just before the departure station, that is, immediately before the disembarking station.
Therefore, when the position (first position) of the user U when it is determined that the passenger P has stood is between the stations “Hasuta Station” to “Higashiomiya Station”, the passenger P's exit station is Hasuda Station. It can be identified as Higashi-Omiya Station.

However, in this case, even if two getting-off stations can be specified, one getting-off station cannot be specified, and it is also necessary to specify the moving direction (up/down) of the train.
Therefore, for example, position information of the position (second position) of the user U at a timing later than when the passenger P stands up is acquired, and the moving direction of the train is specified based on the first position and the second position. ..
For example, as shown in Fig. 10, since the position information of the "second position" is "35°57'37.7 seconds north latitude, 139°38'30.9 seconds east longitude", it is more than the first position at Higashiomiya station. It shall be located on the side.
In this case, the getting-off station of the passenger P can be specified as Higashi-Omiya station.
In other words, since the user U is moving from the first position to the second position between the stations of Hasuda station and Higashiomiya station, the train on which the passenger P is riding with the user U is from Hasuda station to Higashiomiya. You can see that it is an up train moving toward the station.
Therefore, if a passenger P sitting in the seat of this up train stands between Hasuda station and Higashi-Omiya station, the passenger P is likely to get off at the next stop, Higashi-Omiya station. It is possible to specify Higashi-Omiya station as the exit station.

In addition, the user U can also set the moving direction (up/down) of the train by manual operation in advance, and in that case, only the position information of the first position, that is, only one position information is sufficient.
Further, the moving direction can be automatically set based on the time and time zone when the program is started.
Specifically, if the user U is a general office worker or student, the boarding time when commuting to work or attending school falls within the range of a predetermined time zone. When the program is started, it can be automatically set to the uphill, and when the program is started in a predetermined time zone after the afternoon or the evening, it can be automatically set to the downhill.
Further, the position information of the second position may be acquired immediately after the user U gets on the vehicle.
Further, by acquiring the position information at regular intervals, it is possible to acquire the position information at a timing before the time when the passenger stands up. This position information can be used as the position information of the third position, and can also be used as the position information of the second position.

The processor 11 stores, as an alighting station DB, information in which the alighting station specified by the alighting station specifying means and the identification information of the passenger P are associated with each other (alighting station storage means).
FIG. 11 is a chart showing a first example of the getting-off station DB.
As shown in the figure, the getting-off station DB is configured based on information in which face information, which is the identification information of the passenger P, and the getting-off history are associated with each other.
The face information is composed of a face image (face photograph) of the passenger P and face feature information (numerical information such as shape, position, size of eyes, nose and mouth) extracted from the face information.
The getting-off history is a history of the getting-off stations of the passengers P specified by the getting-off station specifying means, and the past getting-off stations (for example, the past five getting-off stations) of the passengers P are recorded separately for ascending and descending. It has been configured.
Therefore, the getting-off history is registered and updated each time the getting-off station is specified.
Specifically, it is determined whether or not the face information of the passenger P is already registered in the getting-off station DB by using a known face authentication technology, and if it is not registered, the face information and the getting-off history (the specified getting-off point are specified). (Station) is newly registered (stored), and if already registered, the disembarking station specified by the disembarking station specifying means is added (updated) to the dismounting history corresponding to the face information.

As the face information of the disembarking station DB, for example, a face image (face photograph) of the passenger P taken in the train can be used.
Specifically, the determination means determines that the passenger P has stood by using the captured image in the train, and when it is determined that the passenger P has stood, the passenger P included in the captured image is determined. The face image and its characteristic information are registered (stored) as face information, and the alighting station specified by the alighting station specifying means is stored as an alighting history in association with the face information.
In this case, it is announced on the websites of service operators and train operators, advertisements in trains, etc. that the facial photograph of the passenger P is used for identifying the getting-off station and not used for other purposes. It is preferable that
Further, a person who desires to use the service can register his/her own face image as the face information of the alighting station DB.
For example, a person who wants to use the service can register only the face information of the person in advance by passing the face photograph of himself/herself taken by a smartphone or the like to the terminal device 1.
As a result, the registrant can use the service as a "user" and can be targeted as a "passenger" by another user to specify the exit station.
In addition, at the time of registration, the person who wants to use the service is notified in advance that the facial photograph of the user U will be used for identifying the disembarkation station and that it will not be used for any other purpose. It is preferable to obtain it.

FIG. 12 is a table showing a second example of the getting-off station DB.
As shown in the figure, the face information can be configured by only the feature information without including the face image.
That is, by storing only the characteristic information from the captured image of the passenger P and not storing the face image, the privacy of the passenger P can be protected.

(Prediction of getting off station)
The processor 11 predicts the getting-off station of the passenger P who is currently on the train, based on the information of the getting-off station of the passenger P specified by the getting-off station specifying means in the past (getting-off station predicting means).
For example, as shown in FIG. 11 or FIG. 12, it is assumed that face information and exit history of passengers Px, Py, and Pz are registered in the exit station DB.
Here, it is assumed that the user U boarded an up train of the Ueno Tokyo line (Utsunomiya line system) at "Higashiomiya station".
Next, the user U wears the terminal device 1 and activates the exit station information program, and then photographs the passenger P sitting in the seat.
The terminal device 1 may be attached and the getting-off station information program may be started before boarding.
FIG. 13 is a target visually recognized through the lens 142 when the terminal device 1 is used in the mode shown in FIG. FIG.
That is, the user U is a captured image captured while standing in front of the passenger P seated in the seat, and the captured image includes captured images of the passenger Px, the passenger Py, and the passenger Pz. ing.

The disembarkation station information program of the present invention includes a program related to the prediction processing of the disembarkation station and a program related to the specification processing of the dismounting station (the registration processing of the dismounting station DB).
The disembarkation station information program of the present invention is constructed in a mode in which a part of the processing in the prediction processing of the disembarkation station and a part of the processing in the above-mentioned specification processing of the disembarkation station (registration processing of the disembarkation station DB) are shared. can do.
For example, the captured image including the passenger P used to predict the disembarking station can be used to specify the dismounting station, and conversely, the captured image including the passenger P used to specify the dismounting station can be used to predict the dismounting station. Can be used.
As a result, the process of registering the disembarking station DB and the process of predicting the disembarking station can be executed in parallel.

Then, the processor 11 acquires the identification information of each passenger P from the captured images of the passengers Px, Py, and Pz.
Specifically, the face information (face image, feature information) of each of the passengers Px, Py, Pz is extracted.
Next, the processor 11 searches the getting-off station DB for the getting-off history using the acquired face information as a key.
Specifically, it is determined whether or not the same face information as the acquired face information is registered in the alighting station DB by using a known face authentication technology, and if it is already registered, the face information is handled. Extract the exit history.
It should be noted that if it has not been registered, the disembarking station cannot be specified, but in this case, the acquired face information can be newly registered in the disembarking station DB.

Similar to the process of identifying the disembarking station, the predicting process of the disembarking station also identifies whether the train on which the user U boarded is "up" or "down".
For example, when the getting-off station prediction application is activated, “up” or “down” is selected by the operation of the user U, and the selection can be performed by referring to the selection information.
In addition, a plurality of position information is acquired, such as position information is acquired during a predetermined operation and after a predetermined time after the operation, and “up” is performed based on the moving direction of the user U identified based on the plurality of position information. It is also possible to specify whether it is "down".
Further, as in the case of the disembarking station specifying process, the moving direction can be automatically set based on the time and the time zone when the program is started.

As shown in FIGS. 11 and 12, the getting-off station DB is composed of a plurality of getting-off histories (previously to five times ago) that the passenger P got off in the past, and the getting-off station of the passenger P based on the getting-off history. Predict.
In the case of the present embodiment, the most frequently alighting station is predicted as the alighting station of the passenger P based on the latest five alighting history.
As a result, the exit station of the passenger Px is specified (predicted) as "Omiya station", the exit station of the passenger Py is "Tokyo station", and the exit station of the passenger Pz is "Shinagawa station".
Note that the dismounting station may be specified based on the dismounting history of the previous month or the dismounting history of the same month of the previous year, instead of the latest 5 dismounting histories.
Further, the number of times may be used instead of the frequency, and the last getting-off station can be predicted as the getting-off station.

The processor 11 can display the prediction information of the alighting station on the lens 142 by the projection of the projection unit 141.
As a result, the user U can visually recognize the prediction information of the exit station displayed on the lens 142.
FIG. 14 is a display example of prediction information of an alighting station.
As shown in the figure, the prediction information of the disembarking station can be displayed (projected) in association with the passenger P.
Thus, the user U can predict which passenger P will get off at which station.
For example, in the display example shown in FIG. 14A, the user U has a high possibility that the passenger Px gets off at Omiya station, the passenger Py has a high possibility of getting off at Tokyo station, and the passenger Pz is It is possible to understand that there is a high possibility that you will get off at Shinagawa Station.
Thereby, the user U can understand that the passenger Px is likely to get off the earliest among the passengers Px to Pz, for example.
Therefore, the user U may be able to quickly sit in the seat by standing in front of the passenger Px.
Further, if the predicted disembarkation station of the passenger P included in the field of view of the user U (that is, the passenger P included in the captured image) is far, the passenger P who gets off early by moving the place or changing the line of sight. You can also use it to search for.
It should be noted that instead of the station name, it is also possible to indicate the number of stations up to which the passenger disembarks, such as what station the passenger dismounts.
Further, as shown in FIG. 14B, the number of stations up to the disembarking station can be shown for each range.
Thereby, the user U cannot grasp the getting-off station name, but can see whether the getting-off station is “close” or “far”.
By doing so, it is possible to indirectly indicate the exit station of the passenger P while protecting the privacy of the passenger P.
Further, as shown in FIG. 15, the number of stations up to the disembarking station can be displayed in different colors for each range. In the drawing, the inside of the head region is colored, but it is also possible to color the frame of the region by coloring.
With such a display method as well, the privacy of the passenger P can be protected while indirectly indicating the exit station of the passenger P.

A method of identifying an alighting station (a method of registering the alighting station DB) will be described.
FIG. 16 is a flowchart showing a method of identifying an alighting station (a method of registering an alighting station DB).
In the registration of the alighting station DB, first, the passenger P in the train is photographed (S11).
For example, the user U photographs the passenger P sitting in the seat (see FIGS. 5 and 6).
Next, it is determined whether or not the passenger P has stood (S12).
Specifically, based on the captured image of the passenger P included in the captured image acquired in S11, the passenger P stands when the height of the passenger P's head or a predetermined portion of the head changes by a predetermined value or more. (See FIGS. 6 to 8).
When it is determined that the passenger P has stood (S13-Yes), the position information of the user U is acquired (S14).
Specifically, position information at the timing when it is determined that the passenger P stood (position information of the first position), position information before or after the timing (position information of the second position), Is acquired (see FIGS. 9 and 10).
Next, the getting-off station is specified (estimated).
Specifically, the alighting station and the moving direction (up or down) are specified based on the position information acquired in S13.
More specifically, when the first position coincides with a station, that station is estimated to be the exit station, and when the first position is between stations, the next stop station is estimated to be the exit station.
Next, the face information of the passenger P is acquired (S15).
Specifically, the face information (face image, face feature information) of each passenger P is extracted from the captured image of the passenger P included in the video acquired in S11.
Then, the information in which the face image of the passenger P acquired in S15 and the exit station identified in S14 are associated is stored as the exit station DB (S16).
Specifically, the specified alighting station is divided into an uphill/downhill route and registered or updated as a past alighting history (see FIGS. 11 and 12).

Next, a method of predicting an exit station will be described.
FIG. 17 is a flowchart showing a method of predicting a getting-off station.
In predicting the disembarking station, first, the passenger P in the train is photographed (S21).
Specifically, the user stands in front of the seat and shoots an image including the passenger P sitting in the seat (see FIG. 13).
Next, the face information of the passenger P is acquired (S22).
Specifically, the face information (face image, face feature information) of each passenger P is extracted from the captured image of the passenger P included in the video acquired in S21.
Then, the getting-off station is identified (predicted) based on the getting-off station DB (S23).
Specifically, the getting-off station DB is searched by using the face image acquired in S22 as a key, the getting-off history of the passenger P having the same face information is extracted, and the getting-off station having a high frequency from the getting-off history is selected as the passenger. It is specified (predicted) as the station of getting off P.
Then, the prediction information of the disembarking station specified in S23 is displayed (S24).
Specifically, the station name or the like of the disembarking station is displayed in a manner associated with the corresponding passenger P (see FIGS. 14 and 15).

(Composition of multiple devices)
The characteristics of the terminal device 1 and the disembarkation station information system S of the present invention that are configured by a plurality of devices will be described below.
(1) Combination of Terminal Devices As shown in FIG. 2A, the terminal device 1 and the getting-off station information system S of the present invention can be configured by a combination of a glasses-type wearable terminal and a smartphone.
Such a configuration is effective when the other terminal device has a configuration or function that is lacking in one terminal device.
For example, when the glasses-type wearable terminal is not provided with the positioning unit or the image capturing unit, the present invention can be implemented by using the positioning unit or the image capturing unit included in the smartphone.
Further, when the processing capability of the CPU of the glasses-type wearable terminal is low or the storage capacity is small, for example, various processes based on face authentication or image analysis can be executed on the smartphone or stored in the smartphone.
Further, the wearable camera is not limited to the glasses-type wearable terminal, and for example, a glasses-type camera or a wearable camera of a type worn around the neck or attached to clothes can be adopted to combine the wearable camera and the smartphone.
You can also use a smartwatch instead of a smartphone.
(2) System configuration including terminal device 1 and server device 2 As shown in FIG. 2B, the terminal device 1 such as a glasses-type wearable terminal and the server device 2 are connected via a communication line 3 such as the Internet. The connected configuration will be described.
In this case, the information from the terminal devices 1 of a plurality of users U can be collected by the server device 2.
Specifically, the getting-off history of each passenger P on various routes on which various users U board can be aggregated by the server device 2, and the getting-off station DB can be constructed.
For this reason, the getting-off station can be predicted for each passenger P on various routes, and the prediction accuracy can be improved by accumulating a large amount of data.
Further, in this case, the terminal device 1 owned by each user U does not store the photographed image of the passenger P and the like, and the server device 2 centrally manages the privacy of the passenger P to facilitate protection. it can.

As described above, the disembarkation station information system S of the present invention includes the terminal device 1 owned by the user U who is on the train, and the terminal device 1 includes the image capturing unit (image capturing unit 12) and the image capturing unit ( A determination unit that determines that the passenger P in the train has stood based on the image captured in the train captured by the image capturing unit 12), and a user U when the determination unit determines that the passenger P has stood. The position information acquisition unit that acquires the position information of the passenger P and the disembarkation station specification unit that specifies the dismounting station of the passenger P based on the position information of the user U acquired by the position information acquisition unit.
As a result, the exit station of the passenger P can be efficiently and accurately specified with a simple configuration.
Further, based on the information on the exit station specified in the past, the exit station of the passenger P of the train on which the user U is currently riding can be predicted efficiently and accurately with a simple configuration.

On the other hand, in the conventional technology, as disclosed in the above-mentioned patent document, the passenger's exit station is specified based on the commuter pass information and the boarding history information recorded in the IC card or the like. However, there is a problem that it cannot be established unless the passenger has an IC card or the like.
In addition, it is necessary to install a reading device such as an IC card in each seat, which is very expensive, and if the IC card or the like is covered with a shield, the commuter pass information and boarding history information cannot be retrieved. was there.
According to the present invention, it is possible to solve the above problems of the conventional technique.

In particular, the present invention has the following features.
(1) It is a technology for specifying a disembarking station based on the fact that the passenger P stands up as a dismounting act or a dismounting preparation act.
Since the passenger P gets off after the train arrives at the getting-off station, in many cases, the passenger P sitting in the seat also moves to the rising door and gets off the train after the train arrives at the getting-off seat.
Therefore, the getting-off station can be specified by regarding that the passenger P stood as one getting-off action.
However, among the passengers P who are sitting, there are a considerable number of passengers P who get up and prepare for getting off before the train arrives at the exit station.
The present invention also identifies the exit station of the passenger P who stands up in preparation for such exit.
As a result, it is possible to accurately specify the disembarking station corresponding to various types of passengers P.
(2) The timing at which the passenger P stands is used as an element for estimating the exit station.
As explained in (1), among the passengers P who are sitting, there are a considerable number of passengers P who stand up in preparation for getting off before the train arrives at the exit station. Varies depending on the character of the passenger P, the habit, the situation in which the passenger P is placed, and the like.
For example, a passenger P with a cautious character stands at an early timing to prepare early, stands after an announcement that he will arrive at the exit station soon, or stands when the platform of the exit station is visible, or gets off. There are various situations such as standing at an intermediate point between the station and the station in front of it, or standing at an early timing because of unfamiliarity with the line or hurrying.
However, although there are various timings at which the passenger P stands in preparation for getting off, the train starts (passes) at the station in front of the getting-off station at the earliest when the passenger P starts preparing for getting off. It was possible to understand empirically and statistically that the latter was mostly.
In other words, the analysis made it possible to find out that the timing at which the passenger P stands for getting off is limited to the station in front of the getting-off station to the getting-off station.
The present invention identifies (estimates) the next station as an alighting station when the standing position is between stations based on such an analysis result, whereby the passenger P accompanying the exit preparation It was made possible to efficiently identify the exit station of.
(3) The getting-off station is estimated based on the position of the passenger P at the standing timing.
As described in (1), the passenger P of the train is the passenger P who starts the getting-off act after the train arrives at the getting-off station, and the passenger P who starts the getting-off preparation act before the train arrives at the getting-off station. Where and exist, even in the latter case, the act of getting off the train is at the earliest after the train departs (passes) from the station in front of the disembarking station.
Therefore, when the position of the train when the passenger P stands up coincides with the position of the station, that station is presumed to be the exit station, and when the position of the train when the passenger P stands is between the stations. , The next station is presumed to be the getting off station.
(4) The dismounting station is estimated based on the position information of the user U, not the position information of the passenger P.
It is difficult to obtain the position information of the passenger P.
Therefore, in the present invention, instead of the position information of the "passenger P", the position information of the "user U" who rides with the passenger P (position information of the terminal device 1 owned by the user U) is acquired, and Was decided to be used to identify the exit station.
(5) Predict the alighting station based on the history of the alighting station.
Although the purpose (reason) for the passenger P to stand is assumed to be other than the purpose of getting off, there are many purposes for getting off, so in the present invention, by specifying the getting-off station based on the frequency based on the history of the getting-off station, I try to improve the accuracy of.

As above, the exit station information system and the exit station information program of the present invention have been described with reference to the preferred embodiments, but the present invention is not limited to the above-described embodiments, and various modifications are made within the scope of the present invention. It goes without saying that implementation is possible.
For example, in the above-described embodiment, various methods have been described as measures for protecting the privacy of the passenger P. However, other than that, for example, the exit station DB (particularly face information) is masked or encrypted by a known method. This also protects the privacy of the passenger P.
In addition, when the disembarking station DB is stored in the server device 2, not only password management but also known masking processing and encryption processing can be performed, and the face characteristic information is stored in the storage device. It can be divided and stored separately.
As a result, it is possible to prevent unauthorized access or leakage to the alighting station DB.

INDUSTRIAL APPLICATION This invention can be utilized suitably for the information provision service regarding a train.

S Disembarkation station information system 1 Terminal device (glasses type wearable terminal)
12 Imaging unit 14 Display unit 141 Projection unit 142 Lens 15 Positioning unit U User P Passenger

Claims (6)

In the disembarkation station information system equipped with the terminal device possessed by the user on the train,
The terminal device includes a photographing means,
Based on a captured image in the train captured by the capturing means, a determination means for determining that a passenger in the train has stood,
Position information acquisition means for acquiring position information of the user when it is determined that the passenger has stood by the determination means,
And a destination station specifying means for specifying a getting-off station of the passenger on the basis of the position information of the user acquired by the position information acquisition means,
The determination means is
An alighting station information system, characterized in that it is determined based on the photographed image that the passenger has stood when the height of the passenger's head or a predetermined portion of the head changes by a predetermined value or more . The alighting station information system according to claim 1 , further comprising: an alighting station predicting unit that predicts an alighting station of the passenger based on the information of the alighting station specified by the alighting station specifying unit . The terminal device possessed by the user is provided with positioning means for measuring position information of the terminal device,
The position information acquisition means,
When the determination means determines that the passenger has stood, the positional information of the terminal device measured by the positioning means is measured,
The alighting station identification means,
The alighting station information system according to claim 1 or 2, wherein the alighting station information system of the passenger is specified based on the position information of the terminal device owned by the user acquired by the position information acquiring means. The position information acquisition means,
Acquiring a plurality of position information including the position information of the user when it is determined that the passenger has stood by the determination means,
The alighting station identification means,
The alighting station information system according to claim 1, wherein the alighting station of the passenger is specified based on the plurality of pieces of position information acquired by the position information acquiring unit. The alighting station information system according to any one of claims 1 to 4, wherein the terminal device is used in a manner worn by a user. A computer including a terminal device possessed by a user on a train,
Determination means for determining that a passenger in the train has stood, based on a captured image in the train captured by the imaging means provided in the terminal device,
Position information acquisition means for acquiring position information of the user when the determination means determines that the passenger has stood,
Getting-off station specifying means for specifying a getting-off station of the passenger on the basis of the position information of the user acquired by the position information acquiring unit, and then allowed to function,
The determination means is
A disembarking station information program, characterized in that , based on the captured image, it is determined that the passenger has stood when the height of the passenger's head or a predetermined portion of the head changes by a predetermined value or more .

Images