JP7664880B2 - Traffic control device - Google Patents

Description

An embodiment of the present invention relates to a traffic control device.

Automatic ticket gate devices, which act as passage control devices and are installed at each station of a railway or other transport facility, are equipped with height measurement sensors to determine whether or not a user passing through the ticket gate passage is an adult. When detecting the height of a user using a height measurement sensor, it is thought that the accuracy of determining whether or not the user is an adult can be improved by installing multiple height measurement sensors in the passage.

However, conventional automated ticket gates use height measurement sensors to determine height, assuming that users will walk through the passage in a natural posture. For this reason, conventional automated ticket gates have the problem that if a user intentionally walks through the passage in a low posture, the device cannot detect the user as an adult.

JP 2013-69090 A

In order to solve the above problems, the present invention aims to provide a traffic control device that can detect the passing status of people.

According to an embodiment, the passage control device includes an interface and a processor. The interface acquires a captured image from a camera that captures an image of a capture area including a passage. The processor determines a posture of a pedestrian passing through the passage based on an image of the pedestrian extracted from the captured image captured by the camera. The processor calculates posture feature amounts of the pedestrian from the captured image captured by the camera each time the pedestrian passes through a plurality of predetermined detection positions, and determines that the pedestrian has passed through the passage in a constant posture if a change in the posture feature amount of the pedestrian who has passed through the plurality of detection positions is within an allowable range, and determines that the pedestrian has passed through the passage while changing his/her posture if a change in the posture feature amount of the pedestrian who has passed through the plurality of detection positions exceeds an allowable range.

FIG. 1 is a diagram illustrating an example of the configuration of a ticket gate system including a ticket gate as a passage control device according to an embodiment. FIG. 2 is a side view showing an example of the appearance of a ticket gate as a passage control device according to the embodiment, on the passage side. FIG. 3 is a diagram showing an example of setting a detection position by a height sensor serving as a height measurement sensor installed in a ticket gate as a passage control device according to the embodiment. FIG. 4 is a block diagram showing an example of the configuration of a control system in a ticket gate as a passage control device according to the embodiment. FIG. 5 is a flowchart for explaining an example of the operation of a ticket gate process (passage control process) by a ticket gate as a passage control device according to the embodiment. FIG. 6 is a diagram for explaining examples of postures of a user when passing through a passage of a ticket gate as a passage control device according to the embodiment. FIG. 7 is a diagram for explaining examples of postures of a user when passing through a passage of a ticket gate as a passage control device according to the embodiment. FIG. 8 is a flowchart for explaining a processing example as a first operation example of a posture determination process in a ticket gate as a passage control device according to the embodiment. FIG. 9 is a diagram for illustrating a second operation example of the attitude determination process in a ticket gate as a passage control device according to an embodiment. FIG. 10 is a flowchart for explaining a processing example as a second operation example of the attitude determination process in a ticket gate as a passage control device according to the embodiment. FIG. 11 is a diagram for explaining a specific example of the attitude determination process according to the second operation example in the ticket gate as the passage control device according to the embodiment. FIG. 12 is a diagram for explaining a specific example of the attitude determination process according to the second operation example in the ticket gate as the passage control device according to the embodiment. FIG. 13 is a flowchart for explaining a processing example as a third operation example of the attitude determination process in a ticket gate as a passage control device according to the embodiment. FIG. 14 is a diagram for explaining the attitude determination process of the third operation example by the ticket gate as the passage control device according to the embodiment. FIG. 15 is a diagram for explaining the attitude determination process of the third operation example by the ticket gate as the passage control device according to the embodiment.

Hereinafter, embodiments will be described with reference to the drawings.
FIG. 1 is a diagram showing an outline of the overall configuration of a ticket gate system 1 including a ticket gate 13 as a passage control device according to an embodiment of the present invention.
The ticket gate system 1 is a system that connects ticket gates 13 installed at the ticket gates of each station. In the configuration example shown in Fig. 1, the ticket gate system 1 includes a server 11, a monitoring panel 12, and ticket gates 13. For example, the server 11 is connected to the monitoring panel 12 installed for each monitored object such as a ticket gate or station. A plurality of ticket gates 13 installed at the ticket gates of each station are connected to the monitoring panel 12 of each station.

The server 11 is connected to the monitoring panel 12 via a network. The server 11 supplies the monitoring panel 12 with information to be provided to each ticket gate 13. The server 11 may also obtain, via the monitoring panel 12, information indicating the operating status of each ticket gate 13 and information indicating the results of the ticket gate processing at each ticket gate 13.

The monitoring panel 12 is connected to the server 11 and to multiple ticket gates 13. The monitoring panel 12 is connected, for example, to the ticket gates 13 installed at the ticket gates to be monitored. The monitoring panel 12 includes an operation panel with a display unit and an operation unit. The monitoring panel 12 monitors the operation status of each connected ticket gate 13, and displays information indicating the status of each ticket gate 13 on the display unit. The monitoring panel 12 also remotely supplies operation instructions to each ticket gate 13 when an attendant operates the operation unit. For example, the monitoring panel 12 sets the operation mode of each ticket gate 13 according to the attendant's operation instructions. The monitoring panel 12 also has a function of distributing information supplied from the server 11 to each ticket gate 13.

Each ticket gate 13 is an example of a passage control device. Each ticket gate 13 is installed at the ticket gate of each station, and performs ticket gate processing (entry processing and exit processing) at the ticket gate. The server 11, the monitoring panel 12, and the ticket gate 13 constitute a ticket gate system that realizes ticket gate processing at ticket gates at stations, etc. The ticket gate 13 switches operating modes in response to instructions from the monitoring panel 12, and performs ticket gate processing using information from the monitoring panel 12.

In the following explanation, the ticket gate 13 as a passage control device according to this embodiment performs ticket gate processing based on a contactless IC card held by the user (passenger). However, the ticket medium (hereinafter also simply referred to as a ticket) that the ticket gate 13 processes as a ticket may be any medium that records information that can be read by the ticket gate 13.

For example, the ticket medium may be a magnetic recording medium (magnetic ticket) in which information is recorded in a magnetic recording section, a recording medium (code ticket) in which coded information such as a two-dimensional code is recorded (printed or displayed), a contactless IC card that transmits and receives information via contactless communication, a mobile terminal such as a mobile phone that has the same functions as a contactless IC card, or a mobile terminal that displays coded information such as a two-dimensional code.

Next, the configuration of the ticket gate 13 serving as the passage control device according to the embodiment will be described.
FIG. 2 is a side view that shows a schematic configuration example of the ticket gate 13 as a passage control device according to the embodiment.
For example, two ticket gates 13 form one ticket gate (entrance or exit) passage in a set. The ticket gates 13 may be used as dedicated entry or exit gates that can pass through the ticket gate passage in only one direction, or as dual-purpose gates that can pass through the ticket gate passage in both directions. For example, the ticket gates 13 are set in advance as either dedicated entry gates, dedicated exit gates, or dual-purpose gates.

In the configuration example shown in FIG. 2, the ticket gate 13 has a housing 30 that forms an entrance or exit passage. The housing that forms the ticket gate 13 is provided with a ticket processing unit (passenger ticket processing unit) 32 that performs ticket processing such as reading information from a ticket held by a user. A reading unit (communication unit) of the ticket processing unit 32, through which the user presents the ticket, is provided at one end (the end on the entrance side) of the top surface of the housing 30 that forms the ticket gate 13. The ticket processing unit 32 reads information recorded on an IC card that the user presents to the reading unit as a ticket.

For example, the ticket processing unit 32 includes a card reader/writer that reads information from an IC card that the user presents as a ticket at a specified reading unit. The ticket processing unit 32 may also be equipped with a device that processes media other than IC cards used as tickets. For example, the ticket processing unit 32 may be configured to include a magnetic ticket processing unit that processes magnetic tickets that have ticket information recorded in a magnetic recording unit. The ticket processing unit 32 may also be configured to include a code reader that reads code information from a medium on which code information as ticket information is printed.

A display unit 33 for providing guidance to users or staff is provided on the top surface of the housing forming the ticket gate 13. The display unit 33 is, for example, composed of a liquid crystal display device. The display unit 33 displays a guidance screen that informs users passing through the passage of the results of the ticket gate processing. The display unit 33 may be provided in a position where the user can check the displayed content. The display unit 33 is not limited to being set at the end of the top surface of the housing on the exit side, and may be provided, for example, in the center of the top surface of the housing or adjacent to the reading unit of the ticket processing unit 32 on the top surface of the housing.

At both ends of the side of the case 30 that forms the ticket gate 13 facing the passage, there are provided doors 34, 35 that can be opened and closed to allow or deny passage of pedestrians. In the configuration example shown in FIG. 2, if the direction of arrow a is the direction of travel toward the exit, door 34 is provided at the end of the passage on the exit (entrance) side and prevents users from passing through the passage. Door 35 is provided at the end of the passage on the entrance (entrance) side and prevents users from entering the passage.

The side of the housing 30 that forms the ticket gate 13 is provided with multiple passage detection sensors (human detection sensors) 36. The passage detection sensors 36 are composed of reflective or transmissive detection sensors. The output signal from the passage detection sensors 36 is either "light" or "dark." The passage detection sensors 36 detect whether or not a person is present at a specified detection position. The ticket gate 13 detects the presence or absence of a person at each detection position in the passage using the multiple passage detection sensors 36. As a result, the ticket gate 13 is able to determine the position of a user passing through the passage based on the detection results of the multiple passage detection sensors 36.

Furthermore, the ticket gate 13 is provided with a proximity sensor 37 on the side facing the passage entry direction on the side where the reading unit of the ticket processing unit 32 is provided (the entrance side of the passage). The proximity sensor 37 detects a person approaching the ticket gate 13. The ticket gate 13 detects a person approaching the passage using the proximity sensor 37, allowing the person to smoothly undergo ticket gate processing. For example, the proximity sensor 37 may detect a person in a position where they can present an IC card as a boarding ticket to the reading unit of the ticket processing unit 32. In this case, the ticket gate 13 may control the ticket processing unit 32 to become operable according to the detection result of the proximity sensor 37.

In addition, when the passageway is used in both directions, the ticket gate 13 controls the direction of travel of the user (ticket gate direction) by controlling the opening and closing of the door 35 in response to the detection results of the approach sensor 37. In addition, the ticket gate 13 may be configured to control switching between a power saving mode, which reduces power consumption by turning off the display unit 33, and a normal operation mode (operation mode in which ticket gate processing is performed), in response to the detection results of the approach sensor 37.

The ticket gate 13 may also be provided with a passage guidance display unit on the side facing the passage entrance direction on the side where the reading unit of the ticket processing unit 32 is provided (the entrance side of the passage). For example, the ticket gate 13 may display on the passage guidance display unit whether or not entry to the passage to the ticket gate is possible.

The housing 30 forming the ticket gate 13 is provided with a number of height sensors (height measurement sensors) 38. The height sensors 38 detect the height of a person. The height sensors 38 are provided at a number of locations on the housing 30 in order to detect the height of a person at a number of locations within the passage. In the example shown in FIG. 2, the height sensors 38 are provided within a sensor cover provided on the top surface of the housing. The height sensors 38 are, for example, configured with reflective detection sensors. Any height sensor 38 may be used as long as it detects whether a person is above a predetermined height at which the person is determined to be an adult.

FIG. 3 is a diagram showing an example of setting the detection position by the height sensor 38 installed in the ticket gate 13. As shown in FIG.
In the example shown in Fig. 3, the height sensor 38 detects whether or not a person is present at the detection position indicated by the arrow b in Fig. 3. That is, the height sensor 38 outputs a signal indicating that a person has been detected if the person passing through the passage is taller than the arrow a. On the other hand, the height sensor 38 does not detect a person passing through the passage if the person is shorter than the arrow b.

The arrow b in FIG. 3 is set to a height that determines whether or not a person passing through the passage formed by the ticket gate 13 is an adult. Therefore, when the height sensor 38 detects a person, the ticket gate 13 determines that the person passing through the passage is an adult. When the height sensor 38 does not detect a person in the passage, the ticket gate 13 determines that the person is not an adult.

In the configuration example shown in FIG. 2, the ticket gate 13 is equipped with a camera 39. The camera 39 captures images of people passing through the passage of the ticket gate 13. The camera 39 is installed so that the whole body of a single user using the ticket gate 13 can be captured at multiple points. The camera 39 may be installed in the housing 30, or may be installed at a location separate from the housing 30. The camera 39 does not have to be a component provided in the ticket gate 13. For example, the camera 39 may be a surveillance camera installed at a station to monitor users at the ticket gate or the ticket gate 13.

Next, the configuration of a control system in the ticket gate 13 serving as the passage control device according to this embodiment will be described.
FIG. 4 is a block diagram showing an example of the configuration of a control system in the ticket gate 13 serving as a passage control device according to this embodiment.
In the example configuration shown in Figure 4, the ticket gate 13 has a processor 41, a ROM 42, a RAM 43, a memory unit 44, a communication unit 45, a clock 46, a door control unit 47, a speaker 48, a ticket processing unit 32, a display unit 33, a passage detection sensor 36, a proximity sensor 37, a height sensor 38, a camera 39, a camera interface 49, etc.

The processor 41, ROM 42, and RAM 43 function as a control unit that controls the entire ticket gate 13. The processor 41 is, for example, a calculation unit such as a CPU. The processor 41 realizes various processing functions by executing programs stored in the ROM 42 or the memory unit 44. For example, the processor 41 executes a ticket gate processing program stored in the ROM 42 or the memory unit 44 for performing the ticket gate processing described below.

The ROM 42 is a non-volatile memory. The ROM 42 functions as a program memory that stores programs executed by the processor 41. The ROM 42 stores programs executed by the processor 41, control data, etc.

RAM 43 is a memory that temporarily stores data. RAM 43 functions as a working memory. RAM 43 loads programs and stores data that is being processed by processor 41. RAM 43 also functions as a buffer memory that temporarily stores communication data, etc.

The storage unit 44 is a memory that stores data. The storage unit 44 includes a rewritable non-volatile memory such as an HDD or SSD. The storage unit 44 stores information such as fare information used in ticket gate processing. The storage unit 44 may also store programs executed by the processor 41, control information, etc.

The communication unit 45 is a communication interface for communicating with the monitoring panel 12. The communication unit 45 may be an interface for wired communication with the monitoring panel 12, or an interface for wireless communication. The communication unit 45 transmits information indicating the operating status to the monitoring panel 12, and receives control instructions such as the operating mode from the monitoring panel 12.

The clock 46 measures time. The processor 41 obtains the elapsed time from the time measured by the clock 46. The clock 46 may be a timer that measures the elapsed time from the time instructed by the processor 41.
The speaker 48 outputs a sound. For example, the speaker 48 outputs a sound such as an alarm when the user is not allowed to pass through.

The ticket processing unit 32 processes the ticket medium presented by the user. In this embodiment, the ticket processing unit 32 is configured with a card reader/writer that performs contactless communication with a contactless IC card (or a mobile terminal device with the same function as a contactless IC card) as the ticket medium. The ticket processing unit 32 may also include a mechanism for processing a magnetic ticket as the ticket medium.

The ticket processing unit 32 communicates with a contactless IC card, which serves as a ticket medium, held over the reading unit (predetermined reading range), and acquires information recorded on the IC card. When the ticket processing unit 32 allows a user who has held an IC card over the reading unit to pass, it writes information indicating the results of ticket gate processing, etc., to the held IC card.

The display unit 33 is a display that displays guidance to users passing through the ticket gate. The display unit 33 is installed, for example, near the exit side or near the center of the main body of the ticket gate 13 so that it is easy for users passing through the ticket gate to see. The display content, such as guidance, displayed on the display unit 33 is controlled by the processor 41.

The door control unit 47 includes a door opening/closing mechanism that opens and closes the doors 34 and 35. The door control unit 47 controls the passage of users by opening and closing the doors 34 and 35. For example, the door control unit 47 closes the door 34 to prevent passage through the passage that the ticket gate 13 forms at the ticket gate.

The passage detection sensor (human detection sensor) 36 is a sensor that detects the presence or absence of a person (or object). As shown in FIG. 2, the passage detection sensor 36 is installed on the side of the ticket gate 13 so that a number of predetermined positions in the passage are detection positions, and outputs a detection signal indicating the presence or absence of a person at each position to the processor 41. Based on the detection results of the passage detection sensor 36, the processor 41 can detect the position of a person in the passage and whether or not a person is passing through the passage.

The approach sensor 37 detects a person approaching the ticket gate 13. The approach sensor 37 is provided on the entrance side of the housing of the ticket gate 13, and outputs a detection signal to the processor 41 indicating the presence or absence of a person near the entrance of the ticket gate 13.

The passage detection sensor 36 and the approach sensor 37 are sensors that detect the position of a person (user) using the ticket gate 13. In the example shown in FIG. 2, the processor 41 monitors whether the user is in area A, area B, area C, or the passage control area.

Area A is an area in front of the passage formed by the ticket gate 13. When the approach sensor 37 is turned on (when the approach sensor 37 detects a person), the processor 41 determines that the user is present in area A.
Area B is an area on the entrance side of the passage. When the passage detection sensor 36a is turned on, the processor 41 determines that the user has entered the passage (area B).

Area C is the area following area B, and is the area through which users moving in the direction of arrow a pass after area B. When passage detection sensor 36b turns on, processor 41 determines that the user has moved from area B to area C. The passage control area is the area following area C, and is the area through which users moving in the direction of arrow a pass after area C. When passage detection sensor 36c turns on, processor 41 determines that the user has moved from area C to the passage control area.

The height sensor 38 is an example of a sensor that measures the height of a person. The height sensor 38 detects whether the height of a person is equal to or greater than a predetermined height (height considered to be an adult). For example, the height sensor 38 is a reflective sensor with a detection position set as shown in FIG. 3, and detects the presence or absence of a person at the predetermined height. The height sensor 38 outputs an ON signal when it detects a person at the detection position set to the predetermined height that is determined to be an adult. When the processor 41 receives an ON signal from the height sensor 38, it determines that a person who is equal to or greater than the predetermined height and is determined to be an adult is present.

The height measurement sensor is not limited to the height sensor 38 configured as shown in Figures 2 and 3. The height sensor 38 as a height measurement sensor may be any sensor that acquires a signal that can determine whether or not the height (body height) of a user at a predetermined position in the passage is equal to or greater than a predetermined height. For example, the height sensor 38 may be a height measurement sensor that detects the height of a user from one side, or may be a sensor that detects the height of a user using a measuring device other than a reflective sensor. The height sensor 38 may also be replaced with a height measurement sensor that uses a three-dimensional sensor or the like, or may be replaced with a sensor that measures the height from an image captured by a camera (height measurement sensor).

The camera 39 captures images at a predetermined cycle (frame rate). For example, the camera 39 is set to capture an area including areas A, B, and C shown in FIG. 2. The camera 39 may also be driven to capture areas A, B, and C as the capture areas in response to the movement of the user. For example, the processor 41 may be set to capture area A while waiting and while the approach sensor 37 is detecting a user, capture area B while the passage detection sensors 36a and 36b detect that the user is in area B, and capture area C while the passage detection sensors 36b and 36c detect that the user is in area C.

The camera 39 is also installed so as to capture the entire body or a part of the user who uses the ticket gate 13. The camera 39 is to capture the part of the user that is the target of posture determination in the posture determination process described below. For example, when performing posture determination processing to determine posture based on the movement of the entire body of the user, the camera 39 is set to capture an image of a range including the entire body of the user in front of the ticket gate 13 and in the passage. When performing posture determination processing to determine posture based on the movement of the upper body of the user, the camera 39 is set to capture an image of a range including the upper body of the user in front of the ticket gate 13 and in the passage. When performing posture determination processing to determine posture based on the movement of the part above the knees of the user, the camera 39 captures an image of a range above the knees of the user in front of the ticket gate 13 and in the passage.

The camera 39 is connected to the camera interface 49. The camera 39 may be installed separately from the ticket gate 13. When the camera 39 is installed separately from the ticket gate 13, the processor 41 is configured to acquire images captured by the camera 39 via the camera interface 49. Furthermore, the camera 39 does not have to be one for each ticket gate 13, and one camera 39 may be provided for multiple ticket gates 13. In this case, the camera interface 49 may acquire images captured by the camera 39 of an area including the passage of the ticket gate 13 via a network.

Next, a ticket gate process (passage control process) performed by the ticket gate 13 as the passage control device according to the embodiment will be described.
FIG. 5 is a flowchart for explaining the ticket examination process performed by the ticket examiner 13 as the passage control device according to the embodiment.
During standby, the processor 41 of the ticket gate 13 detects the approach of a user to the passageway based on the detection result of the approach sensor 37. The processor 41 also detects the entry of a user into the passageway based on the detection result of the passage detection sensor .

When the processor 41 detects a user approaching (or entering) the aisle (ST1, YES), it executes posture determination processing, height determination processing, and ticket reading processing (ST2, 3, 4). Note that the posture determination processing, height determination processing, and ticket reading processing are not limited to the order shown in FIG. 5, and may be executed in any order. Also, some or all of the posture determination processing, height determination processing, and ticket reading processing may be executed in parallel.

The posture determination process is a process for determining the posture of a user when passing through a passage. The processor 41 executes the posture determination process for determining the posture of the user based on the image captured by the camera 39 (ST2). The posture determination process in the ticket gate 13 as a passage control device is a process for determining whether the user is intentionally changing their posture (whether they are passing through the passage in the correct posture) when the height sensor 38 detects the height. The posture determination process will be explained in detail later.

The height determination process is a process for determining the height of a user passing through a passage. The processor 41 executes a height determination process for determining the height of a user present at a predetermined position in the passage based on the detection result of the height sensor 38 (ST3). The height determination process in the ticket gate 13 as a passage control device is a process for determining whether the height of a user is equal to or greater than the height of an adult based on whether the user is detected by the height sensor 38, whose detection position is set to a height that determines an adult.

The ticket reading process is a process of reading ticket information from a ticket medium presented (held) by a user. The processor 41 executes a ticket reading process in which the ticket processing unit 32 reads ticket information from the ticket medium presented by the user (ST4). For example, the ticket processing unit 32 reads ticket information recorded on an IC card serving as a ticket medium that the user holds over the reading unit. The ticket processing unit 32 may also be configured to read ticket information recorded on a magnetic ticket serving as a ticket medium presented by the user. The ticket processing unit 32 may also be configured to read code information in which the ticket information displayed on the ticket medium presented by the user is encoded.

When the processor 41 reads the ticket information from the ticket medium (ticket) presented by the user, it determines whether the ticket presented by the user is a child's ticket (ST5) based on the read ticket information.

If the ticket presented by the user is not a child's ticket (ST5, NO), processor 41 determines that the user is an adult and determines whether the adult is allowed to pass based on the ticket information of the ticket (ST13). If the ticket information of the ticket presented by the user allows adults to pass (ST14, YES), processor 41 notifies the user that they are allowed to pass by displaying a notice on display unit 33 indicating that they are allowed to pass (ST15), and opens door 34 (ST16). Processor 41 then allows the user to pass.

If the ticket presented by the user is a child's ticket (ST5, YES), the processor 41 determines whether the height (stature) of the user is determined to be non-adult (child) by the height determination (ST6). For example, the processor 41 determines whether the height (stature) of the user is equal to or greater than a predetermined height based on whether the user is detected by the height sensor 38.

If the height is determined to be that of an adult, i.e., if the height sensor 38 detects the user (ST6, NO), the processor 41 proceeds to ST13 and performs a passage determination as an adult based on the ticket information read from the ticket medium presented by the user.

Also, if the height is not that of an adult, i.e., if the user is not detected by the height sensor 38 (ST6, YES), the processor 41 determines whether the user was in a posture that prevented the height from being measured correctly (poor posture) when passing through a specified position in the passage (when passing the detection position of the height sensor) (ST7). Here, poor posture is a posture that prevents the height from being measured by the height sensor 38 as a height measurement sensor (a posture that prevents the height from being measured correctly). In other words, the processor 41 detects, as poor posture, that the user is passing through the passage in a posture that prevents the height (height) from being detected correctly by the height sensor 38.

6 and 7 are diagrams for explaining the posture of a user when passing through the passage of the ticket gate 13. FIG.
The height sensor 38 is installed so as to detect a person who is at least a certain height that is determined to be an adult, as shown in Fig. 4. Therefore, when an adult user of at least the certain height passes through the passage while standing upright and facing forward, the height sensor 38 detects the user, as shown in Fig. 6. However, if an adult user of at least the certain height passes through the passage in a lowered posture (lowered height), the height sensor 38 cannot detect the user, as shown in Fig. 7. Therefore, when a user who presents a child's ticket is not detected by the height sensor 38, the processor 41 detects a user who is trying to pass through in a posture that will not be detected as an adult by determining whether or not the user has poor posture.

If it is determined that the user has passed through the passage in a posture that allows the height to be measured correctly (normal posture) (ST7, NO), the processor 41 proceeds to ST13 and performs a child passage determination based on the ticket information (ticket information of a child's ticket) read from the ticket medium presented by the user.

For example, if a user determined to be in a normal posture is allowed to pass, the processor 41 encourages the user to pass by opening the door 34 and displaying a notice on the display unit 33 indicating that the user is allowed to pass. However, if the user determined to be in a normal posture is a child, the processor 41 may leave the door 34 open to protect the user even if the user is not allowed to pass. This allows a user (a user determined to be a child) who is confirmed to be in a normal posture (no intentional changes in posture) to pass through based on the ticket information.

If it is determined that the user is not in a position that allows height to be measured correctly (ST7, YES), the processor 41 saves an image including an image of the user captured by the camera 39 (ST8). For example, the processor 41 stores images captured by the camera 39 as the user passes through an aisle in a memory such as the RAM 43 or the storage unit 44. When the processor 41 saves an image of the user, it transmits the image stored in the memory to the server 11 via the monitoring panel 12 together with information indicating that the user was in a position that allows height to be measured. In this way, the processor 41 saves an image of the user who has been determined not to be in a position that allows height to be measured correctly in the server 11, which is a higher-level device.

The processor 41 may also store in the memory unit 44 an image of a user who is determined to be in an unsuitable posture for measuring height correctly. The image stored in the server 11 or the memory unit 44 can be used as evidence that the posture has been determined to be unsuitable. For example, the image stored in the server 11 or the memory unit 44 may be disclosed to the user.

When the server 11 or the monitoring panel 12 receives information from the ticket gate 13 that the posture has been determined to be unacceptable, the server 11 or the monitoring panel 12 may display information indicating the ticket gate 13 for which the user has determined that the posture is unacceptable on the display unit of the monitoring panel 12. This allows the staff member monitoring the operation of the ticket gate 13 on the monitoring panel 12 to be notified that there is a user who is unable to pass due to their posture.

Also, if the user is in a posture that does not allow the height to be measured correctly (ST7, YES), the processor 41 outputs an alarm sound from the speaker 48 (ST9) and guides the user to pass through again in the correct posture (ST10). For example, the processor 41 displays, as guidance, a guidance screen on the display unit 33 instructing the user to pass through again in the correct posture. The processor 41 may also display, as guidance, a guidance screen on the display unit 33 instructing the user to go to the attendant's window. Furthermore, the processor 41 may guide the user to pass through a specific ticket gate 13.

Furthermore, if the user's posture does not allow for correct height measurement (ST7, YES), the processor 41 closes the door 34 to prevent the user from passing through (ST11). However, if the processor 41 determines that the ticket is for a child, it may assume that the user is actually a child and leave the door 34 open to ensure the safety of the user (child). Whether or not to close the door when posture is poor is set in each ticket gate 13. For example, it may be possible to set the door opening/closing control for poor posture in each ticket gate 13 from the monitoring panel 12, etc.

As described above, the ticket gate as a passage control device according to the embodiment controls the passage of a user based on the results of the posture determination process, the height determination process, and the ticket reading process. If the ticket medium presented by the user is a child's ticket, the ticket gate will not allow the user to pass through the passage if the posture determination process determines that the user is in a posture that makes it impossible to measure the height using the height sensor.
As a result, the passage control device of the embodiment can determine whether a user is passing through an aisle in a posture that allows the user to measure whether they are the same height (stature) as an adult, and can prevent fraudulent use of a child's ticket by passing through an aisle in a posture that intentionally makes the user appear shorter.

Next, a posture determination process in the ticket gate 13 as the passage control device according to the embodiment will be described.
The posture determination process is a process for determining whether or not a user is intentionally assuming an unnatural posture at a predetermined position in the passageway. The posture determination process estimates the posture of the user and determines whether or not the posture is normal based on changes in the estimated posture.

For posture estimation, the ticket gate 13 can apply a known process that uses deep learning to estimate a person's skeleton points from an image. For example, the processor 41 of the ticket gate 13 estimates the posture of a user from an image of the user captured by the camera 39 using a method such as OpenPose (see https://arxiv.org/abs/1812.08008). The ticket gate 13 may also estimate the posture using other methods for determining the skeleton.

The processor 41 of the ticket gate 13 as a passage control device according to the embodiment executes a posture determination process using posture estimation for each user who uses the ticket gate 13 as a passage control device. This allows the processor 41 to execute a ticket gate process (passage control process) using the results of the posture determination process as described above. The processor 41 can also execute several example operations as posture determination processes. In this embodiment, three example operations will be described as examples of posture determination processes.

First, a first operation example of the attitude determination process in the ticket gate 13 as the passage control device according to the embodiment will be described.
In a first operation example, the ticket gate 13 determines whether the posture of a user passing through an area on the entrance side of a passage (for example, area B shown in FIG. 2) is such that the height can be correctly determined.

For example, the ticket gate 13 detects that a user has entered the passageway using the passage detection sensor 36a, and starts a posture determination process for that user. When the ticket gate 13 starts the posture determination process, it extracts an image of the user from the image captured by the camera 39, and calculates a feature amount that indicates the user's posture. The ticket gate 13 also stores a reference value that serves as the basis for posture determination in the memory unit 44. The ticket gate 13 compares the feature amount that indicates the user's posture, calculated from the image captured by the camera 39, with the reference value, to determine whether the user's posture is one that allows for correct height determination.

FIG. 8 is a flowchart for explaining a processing example as a first operation example of the attitude determination processing in the ticket gate 13 as the passage control device according to the embodiment.
The processor 41 of the ticket gate 13 detects the approach of a user using the approach sensor 37. The processor 41 detects the user detected by the approach sensor 37 using the passage detection sensor 36a. When the passage detection sensor 36a detects the user, the processor 41 determines that the user has passed through a predetermined position (ST21). According to the example shown in Fig. 2, the processor 41 determines that the user has passed through a predetermined position when the user enters area B.

The processor 41 may detect the timing when the passenger ticket medium is presented to the ticket processing unit 32 as the timing when the passenger passes the predetermined position. For example, the processor 41 may detect that the user has passed the predetermined position when the ticket processing unit 32 detects an IC card presented by the user, or when the user inserts a ticket such as a magnetic ticket into the ticket processing unit. Furthermore, the processor 41 may detect that the user has passed the predetermined position from an image captured by the camera 39.

When the processor 41 detects that a user has passed through a predetermined position, the processor 41 acquires an image (passing image) including the user taken by the camera 39 through the camera interface 49. The processor 41 extracts an image of the user from the image taken by the camera 39, and calculates features (posture features) for determining the posture of the user from the image of the user (ST22). For example, the processor 41 performs posture estimation to estimate the feature points (skeleton points) of a person's skeleton for the image of the user.

When estimating skeletal points, the processor 41 calculates the width of both shoulders, the relative angle between the shoulders and waist, the ratio between the shoulders and waist, the waist joint during movement, the swaying range of the shoulder joint and waist, the angles of the elbow joint and knee joint, etc. as features included in the posture features of the user. The processor 41 may also calculate the feature space distribution of the skeletal feature points as the features. Furthermore, the features for determining the posture may be any that can detect poor posture, and may be a combination of multiple index values.

The processor 41 also calculates a feature amount indicating the posture of the user from the skeleton points obtained by posture estimation. For example, the processor 41 calculates the angle of one or more joints from the skeleton points as the feature amount. As a specific example, the processor 41 may identify both shoulders of the upper body from the captured image, extract shoulder joints from the skeleton points at both shoulders of the upper body, and calculate the angle of the shoulder joint as the feature amount. The processor 41 may also identify the waist from the captured image, extract the waist joint from the skeleton points at the waist, and calculate the angle of the waist joint as the feature amount.

The processor 41 may also identify the shoulders and waist from the captured image, extract the shoulder joints from the skeleton points of the shoulders, extract the waist joints from the skeleton points of the waist, and calculate the angle between the shoulder joints and the waist joints as a feature amount. The processor 41 may also identify the user's knees and waist from the captured image, extract the knee joints of both knees from the skeleton points of the knees, extract the waist joints of the waist from the skeleton points of the waist, and calculate the angle between the knee joints and the waist joints as a feature amount.
The processor 41 may also calculate a ratio calculated from the distance between the joints of the skeleton as the posture feature amount. The processor 41 may also calculate a feature space distribution calculated from the coordinate values of the skeleton points as the feature amount.

The feature amount calculated by the processor 41 as the posture feature amount is set in advance from the feature amounts described above. The feature amount calculated as the posture feature amount may be one or more. The feature amount calculated as the posture feature amount may be set according to the installation environment of the camera 39 or the shooting range that the camera 39 can shoot. For example, in an environment where it is difficult for the camera 39 to shoot the lower body of a user passing through an aisle, a feature amount that can be obtained from an image of the upper body of the user may be set as the posture feature amount.

When the processor 41 calculates a posture feature indicating the posture of the user from the passing image, it compares the calculated posture feature of the user with a posture reference value (judgment reference) for posture judgment assuming a posture in which the height sensor 38 can correctly measure height (normal posture) (ST23). In the first operation example, the ticket gate 13 stores a posture reference value for judging that the posture is normal assuming a normal posture in the memory unit 44. The posture reference value is set according to the feature used as the posture feature. In addition, when multiple feature values are used as the posture feature value, the posture reference value may be set to multiple reference values corresponding to each feature value.

The posture reference value may also be calculated by calculating posture feature values from image data of people with normal postures collected from a public database, etc., and creating a posture reference value for a normal posture and an acceptable range for the posture reference value from the distribution of these posture feature values. The posture reference value may also be calculated from multiple posture feature values calculated from images captured by the camera 39 when multiple subjects pass through the passage at the ticket gate 13 with normal postures.

If the posture feature amount of the user calculated from the passing image captured by the camera 39 is within an allowable range (reference range) based on the posture reference value (ST24, YES), the processor 41 determines that the user has a normal posture ("no problem") (ST25). If the posture feature amount of the user exceeds the reference range for the posture reference value (ST24, NO), the processor 41 determines that the user has an abnormal posture ("cannot determine whether the user is an adult or not") (ST26). The processor 41 stores the posture determination result for the user in the RAM 43 or the like, and ends the posture determination process.

As described above, according to the first operation example, the ticket gate as a passage control device compares a preset posture reference value for determining posture with posture feature values calculated from an image of a user passing through a passage captured by a camera. If the calculated posture feature values are within a predetermined tolerance range for the posture reference value, the ticket gate determines that the user has a normal posture, and if the calculated posture feature values are not within a predetermined reference range for the posture reference value, the ticket gate determines that the user has poor posture (a posture that makes it impossible to measure the height).

As a result, a ticket gate as a passage control device that applies the posture determination process of the first operation example can determine whether a user has passed through an aisle in a normal posture. The ticket gate can determine whether a user passing through an aisle is in a state where their height can be measured normally by a height sensor, and can detect when a user who has presented a child's ticket or the like attempts to pass through an aisle in an unnatural posture.

Next, a second operation example of the attitude determination process in the ticket gate 13 as the passage control device according to the embodiment will be described.
9, the ticket gate 13 according to the second operation example tracks a user in an area just before the user enters the passageway, and creates a posture reference value for the user based on posture feature values calculated from an image of the user in the area just before the passageway (e.g., area A). When the user enters the passageway, the ticket gate 13 according to the second operation example uses the posture reference value for the user to determine the user's posture through posture determination processing as described in the first operation example.

FIG. 10 is a flowchart for explaining a processing example as a second operation example of the attitude determination process in the ticket gate 13 as the passage control device according to the embodiment.
In the ticket gate 13, the approach sensor 37 detects a user who has entered area A shown in Fig. 2, which is set in front of the passage. When the approach sensor 37 detects a user, the processor 41 detects the user as a new passerby (user) (ST31). When the processor 41 detects a new passerby, it starts tracking the user (ST32).

For example, the processor 41 detects that a user has moved from area A to area B based on detection signals from the approach sensor 37 and the passage detection sensor 36a. That is, the processor 41 determines that the user is in the area before the aisle (area A) when the approach sensor 37 is on, and detects that the user has entered the aisle (area B) from the area before the aisle (area A) when the approach sensor 37 is off and the passage detection sensor 36a is on. The processor 41 may also detect that the user has moved from area A to area B based on an image of the user extracted from an image captured by the camera 39.

The processor 41 generates a posture reference value for determining posture changes for the user while the user is in the area in front of the aisle. After starting tracking of the user, the processor 41 monitors the user's entry into the aisle to detect that the user has moved from the area in front of the aisle into the aisle (ST33).

If the user has not entered the passage (ST33, NO), the processor 41 calculates posture feature values indicating the posture of the user from the image of the user extracted from the image captured by the camera 39 (ST34). For example, the processor 41 calculates posture feature values indicating the posture of the user from the image of the user extracted from the image captured by the camera 39 by posture estimation as described above.

When the processor 41 calculates the posture characteristic amount of the user standing in front of the aisle, it updates the posture reference value for that user based on the calculated posture characteristic amount (ST35). For example, the processor 41 secures a memory area in the memory unit 44 for storing the posture reference value for the user standing in front of the aisle, and updates the posture reference value for the user stored in the memory unit 44 based on the posture characteristic amount of the user calculated from the images captured by the camera 39 at a predetermined frame rate.

In other words, while the user is in the area in front of the passage, the processor 41 repeatedly calculates the posture characteristic values of the user from the image captured by the camera 39 and updates the posture reference value for the user using the calculated posture characteristic values.

When the user enters the passage (ST33, YES), the processor 41 ends tracking of the user (ST36) and ends creation of the posture reference value for the user who has entered the passage. Note that when the user exits the area in front of the passage without moving into the passage (for example, when the person being tracked is no longer detected by the approach sensor 37 and the passage detection sensor 36a), the processor 41 may end tracking of the user and discard the posture reference value for the user.

When the processor 41 ends tracking of the user as the user enters the passageway, it executes posture determination processing for the user using the posture reference value for the user stored in the memory unit 44 (ST37). For example, the processor 41 executes posture determination processing as described in the first operation example. As a result, in the second operation example, it is possible to execute posture determination processing that determines whether the posture of the user who entered the passageway is normal or poor, using a posture reference value created based on posture feature amounts calculated from an image of the user himself/herself.

11 and 12 are diagrams for explaining an example of the posture determination process according to the second operation example.
Fig. 11 shows a state where a user P1 enters the passage from the area in front of the passage without changing his/her posture, and Fig. 12 shows a state where a user P2 enters the passage from the area in front of the passage while changing his/her posture.

In the example shown in FIG. 11, user P1 enters the passageway in a posture similar to that in the area just before the passageway. Therefore, there is little change between the posture reference value for user P1 based on the posture feature amount calculated from the image of user P1 in the area just before the passageway and the posture feature amount calculated from the image of user P1 when he enters the passageway.

In contrast, in the example shown in FIG. 12, user P2 has a different posture in the passage than in the area before the passage. For this reason, the ticket gate 13 has a large amount of change between the posture reference value for user P2 based on the posture feature amount calculated from the image of user P2 in the area before the passage and the posture feature amount calculated from the image of user P2 who has entered the passage.

That is, in the second operation example, the ticket gate 13 creates a posture reference value for the user each time the user approaches the passage. Therefore, a ticket gate to which the posture determination process of the second operation example is applied determines that the posture of the user is normal if there is no change in posture as shown in FIG. 11 before and after entering the passage, and determines that the posture of the user is poor if there is a change in posture as shown in FIG. 12.

As described above, a ticket gate as a passage control device that applies the posture determination process of the second operation example can create a posture reference value for each user based on posture feature amounts calculated from images taken before the user passes through a specified position (e.g., the entrance to an aisle), and can create a posture reference value that reflects the individual characteristics of each user, such as their body type and walking posture, and can accurately detect posture changes for each user.

Next, a third operation example of the attitude determination process in the ticket gate 13 as the passage control device according to the embodiment will be described.
The ticket gate 13 according to the third operation example sets a plurality of passing positions through which a user passes when passing through a passage, and judges the user's posture based on the amount of change in posture feature of the user at each passing position. That is, in the third operation example, without using a preset posture reference value, it is judged based on the posture change of the user at a specific passing position whether the user passed through the passage with a normal posture (posture change within an acceptable range) or with a large change in posture (posture change exceeding the acceptable range).

FIG. 13 is a flowchart for explaining a processing example as a third operation example of the attitude determination process in the ticket gate 13 as the passage control device according to the embodiment.
The ticket gate 13 performing the posture determination process of the third operation example is assumed to have a plurality of passing positions (predetermined positions). For example, the ticket gate 13 calculates posture feature values indicating the posture of one user in each of areas A, B, and C shown in Fig. 2. In this case, the processor 41 calculates the posture feature values of the user in each area each time the approach sensor 37, the passage detection sensor 36a, and the passage detection sensor 36b detect the user.

That is, when the processor 41 of the ticket gate 13 detects a user at a predetermined passing position (detection position) (ST51), it acquires a captured image from the camera 39 that includes an image of the user who is in an area including the detection position. The processor 41 calculates posture feature values that indicate the posture of the user from an image of the user extracted from the captured image acquired from the camera 39 (ST52). Here, the processor 41 stores the posture feature values calculated from the images captured by the camera 39 at each detection position in a memory such as the RAM 52 or the storage unit 44.

Furthermore, processor 41 repeatedly executes the processes of ST51 to ST52 until it detects that the user has passed all the passing positions. If the user has not passed all the passing positions (ST53, NO), processor 41 calculates posture feature values indicating the posture of the user from the image captured by camera 39 each time it detects that the user has passed the next passing position, and stores the calculated posture feature values in a memory such as RAM 42 in association with the detected position. For example, processor 41 detects the up-down relationship, left-right relationship, and forward/backward swaying range of specific parts of the user (e.g., head, shoulders, hands, etc.) and the angles of specific joints (e.g., shoulders, hips, both shoulders, elbows, etc.) as posture feature values at each detected position. These posture feature values at each detected position may be feature values that can be used to determine whether a user passing through the passage has intentionally changed their posture.

When the user has passed through all the passage positions (ST53, YES), the processor 41 calculates the amount of change between the posture feature values at each detection position from the posture feature values at each detection position stored in memory (ST54). For example, the processor 41 calculates the amount of change between the posture feature values at adjacent detection positions to detect the amount of change in the posture feature values, such as the up-down relationship, left-right relationship, and forward/backward sway of specific parts of the user (e.g., head, shoulders, hands, etc.), and the angle of specific joints (e.g., shoulders and hips, both shoulders, elbows, etc.). With such amounts of change, it can be detected that the user has intentionally changed his/her posture while passing through the passageway.

After calculating the amount of change in posture feature at each detection position, the processor 41 determines whether the calculated amount of change in posture feature at each detection position is within an acceptable range (criterion range) that is used as a criterion for determining whether the posture is normal (ST55). For example, the judgment criteria range is stored in advance in the storage unit 44. The judgment criteria range may be created from publicly available data that indicates an index for detecting posture changes. For example, when the user passes through all specified detection positions (for example, the center of the ticket gate), the processor 41 compares the amount of change in posture feature extracted from an image of the user passing through each detection position with the judgment criteria range to determine whether the posture of the user who passed each detection position has changed significantly beyond the acceptable range.

If the calculated change in posture feature amount is within the judgment criteria range (ST55, YES), processor 41 judges that the user passed through the aisle in a normal posture ("no problem") (ST56). If the calculated change in posture feature amount exceeds the judgment criteria range (ST55, NO), processor 41 judges that the user passed through the aisle in a posture that does not allow for normal height judgment ("cannot determine whether the user is an adult or not") (ST57). Processor 41 stores the posture judgment result for the user in a memory such as RAM 43, and ends the posture judgment process.

14 and 15 are diagrams for explaining the posture determination process of the third operation example by the ticket gate 13. In FIG.
Fig. 14 shows a user P3 passing through the passage from the area before the passage without changing his/her posture, and Fig. 15 shows a user P4 entering the passage from the area before the passage and passing through the passage while changing his/her posture.

In the example shown in FIG. 14, user P3 passes through the passage in a posture similar to that in the area in front of the passage. Therefore, there is little change in posture feature values calculated from images of user P3 in the area in front of the passage, the area near the entrance to the passage (area B), and the area near the center of the passage (area C).

In contrast, in the example shown in FIG. 15, user P4 enters the aisle in a different posture than the area in front of the aisle, and then passes near the center of the aisle after entering the aisle in a posture similar to the area in front of the aisle. For this reason, there is a large change in posture feature values calculated from images of user P4 in the area in front of the aisle, the area near the entrance to the aisle (area B), and the area near the center of the aisle (area C). In the example shown in FIG. 15, user P4's knee joint angle has changed significantly compared to before entering the aisle. These posture changes are calculated as changes in posture feature values.

As shown in FIG. 14, user P3, who maintains a constant posture from in front of the aisle through the passage and after passing through the passage, can be determined to have passed through the passage in a normal posture, because the change in posture feature falls within the judgment criteria range. Also, as shown in FIG. 15, user P4, who changes his posture significantly while passing through the passage, can be determined to have not passed through the passage in a normal posture (passed through the passage in a posture that does not allow for correct height measurement), because the change in posture feature exceeds the judgment criteria range.

As described above, the ticket gate as a passage control device that applies the posture determination process of the third operation example can calculate posture feature amounts at each of a number of predetermined detection positions from images of a user passing through each of the detection positions, and can detect changes in the user's posture while passing through an aisle based on the amount of change in the posture feature amounts at each detection position. This makes it possible to accurately detect users who intentionally change their posture to pass through an aisle, and to detect users whose posture prevents their height (body height) from being measured correctly.

Although some embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and spirit of the invention, and are included in the scope of the invention and its equivalents described in the claims.
The following is an appended summary of the contents of the claims as originally filed.
[1]
In a traffic control device for controlling passage through a passage,
an interface for acquiring a photographed image from a camera that captures an image of a photographed area including the passage;
a processor for determining a posture of a pedestrian passing through the passage based on an image of the pedestrian extracted from an image captured by the camera;
A traffic control device having the above configuration.
[2]
In a traffic control device for controlling passage through a passage,
a height measurement sensor that detects the height of a person passing through the passage;
an interface for acquiring a photographed image from a camera that captures an image of a photographed area including the passage;
a processor that determines a posture of a pedestrian passing through the passage based on an image of the pedestrian extracted from a captured image acquired by the interface, and controls the passage of the pedestrian based on a height detection result by the height measurement sensor and a determination result of the posture of the pedestrian.
[3]
The processor,
calculating a posture feature amount indicating a posture of the pedestrian from an image captured by the camera when the pedestrian passes a predetermined position in the passage;
If the posture feature amount is within an allowable range with respect to the posture reference value, it is determined that the pedestrian has passed through the passage in a posture that allows the height measurement sensor to measure the pedestrian's height,
If the posture feature value exceeds an allowable range with respect to a posture reference value, it is determined that the pedestrian has passed through the passage in a posture that makes it impossible to measure the pedestrian's height by the height measurement sensor.
The traffic control device according to [2].
[4]
the processor creates a posture reference value for the pedestrian based on an image of the pedestrian in an area in front of the passage extracted from an image captured by the camera;
The traffic control device according to [3].
[5]
The processor,
calculating posture feature amounts of the pedestrian from images captured by a camera each time the pedestrian passes through a plurality of predetermined detection positions;
When a change amount of the posture feature amount of the pedestrian who has passed through a plurality of detection positions is within an allowable range, it is determined that the pedestrian has passed through the passage in a constant posture;
If a change in posture feature amount of the pedestrian who passed through a plurality of detection positions exceeds an allowable range, it is determined that the pedestrian passed through the passage while changing his/her posture.
The traffic control device according to [2].
[6]
The processor estimates skeleton points of the passerby from the captured image, and calculates feature information obtained from the skeleton points of the passerby as posture feature amounts.
A traffic control device according to any one of [2] to [5].
[7]
The processor estimates the joints of the pedestrian as skeleton points of the pedestrian, and calculates a ratio calculated from a distance between the joints of the pedestrian as a posture feature.
The traffic control device according to [6].
[8]
The processor calculates a feature space distribution calculated from coordinate values of skeleton points of the passerby as a posture feature.
The traffic control device according to [6].
[9]
Further, a ticket processing unit is provided for reading information recorded on the ticket medium,
the processor, when the ticket medium read by the ticket processing unit and presented by the pedestrian is a child's ticket, does not allow the pedestrian to pass through the passage if the pedestrian's posture makes it impossible to measure the pedestrian's height by the height measurement sensor;
The traffic control device according to [2].
[10]
The height measurement sensor is a sensor that detects whether or not the height of the person passing through the passage is equal to or greater than a predetermined height.
The traffic control device according to [2].
[11]
Further, a door provided on the exit side of the passage;
A door control unit that controls opening and closing of the door,
the processor closes the door by the door control unit when the posture of the passerby is such that the height measurement sensor cannot accurately measure the height of the passerby.
The traffic control device according to [2].
[12]
Further, the image sensor has a memory for storing the image.
the processor stores in the memory an image of the posture of the passerby when the posture of the passerby is such that the height of the passerby cannot be accurately measured by the height measurement sensor;
The traffic control device according to [2].
[13]
Further, a display for displaying guidance is provided,
the processor displays on the display a guidance screen that guides the pedestrian to pass through another passage when the pedestrian's posture is such that the height measurement sensor cannot accurately measure the pedestrian's height.
The traffic control device according to [2].
[14]
Further, a display for displaying guidance is provided,
the processor displays, on the display, a guide for correcting a posture when passing through the passage when the posture of the passerby is such that the height measurement sensor cannot accurately measure the height of the passerby.
The traffic control device according to [2].
[15]
Further, the device has a speaker for outputting sound,
the processor outputs an alarm through the speaker when the posture of the passerby is such that the height of the passerby cannot be accurately measured by the height measurement sensor.
The traffic control device according to [2].

1...Ticket gate system, 11...Server, 12...Monitoring panel, 13...Ticket gate (passage control device), 32...Ticket processing unit, 33...Display unit, 34, 35...Door, 36 (36a, 36b, 36c, 36d, 36e, 36f)...Passing detection sensor, 37...Approach sensor, 38 (38a, 38b, 38c, 38d)...Height sensor (height measurement sensor), 39...Camera, 41...Processor, 42...ROM, 43...RAM (memory), 44...Storage unit (memory), 45...Communication unit, 46...Clock, 47...Door control unit, 48...Speaker, 49...Camera interface (interface).

Claims (11)

In a traffic control device for controlling passage through a passage,
a height measurement sensor that detects the height of a person passing through the passage;
an interface for acquiring a photographed image from a camera that captures an image of a photographed area including the passage;
a processor that determines a posture of a pedestrian passing through the passage based on an image of the pedestrian extracted from the photographed image acquired by the interface, and controls the passage of the pedestrian based on a height detection result by the height measurement sensor and a determination result of the posture of the pedestrian ,
The processor,
calculating a posture feature amount indicating a posture of the pedestrian from an image captured by a camera each time the pedestrian passes through a plurality of predetermined detection positions;
When a change amount of the posture feature amount of the pedestrian who has passed through a plurality of detection positions is within an allowable range, it is determined that the pedestrian has passed through the passage in a constant posture;
If a change in posture feature amount of the pedestrian who passed through a plurality of detection positions exceeds an allowable range, it is determined that the pedestrian passed through the passage while changing his/her posture.
Traffic control device. The processor estimates skeleton points of the passerby from the captured image, and calculates feature information obtained from the skeleton points of the passerby as posture feature amounts.
The traffic control device according to claim 1 . The processor estimates the joints of the pedestrian as skeleton points of the pedestrian, and calculates a ratio calculated from a distance between the joints of the pedestrian as a posture feature.
The traffic control device according to claim 2 . The processor calculates a feature space distribution calculated from coordinate values of skeleton points of the passerby as a posture feature.
The traffic control device according to claim 2 . Further, a ticket processing unit is provided for reading information recorded on the ticket medium,
the processor, when the ticket medium read by the ticket processing unit and presented by the pedestrian is a child's ticket, does not allow the pedestrian to pass through the passage if the pedestrian's posture makes it impossible to measure the pedestrian's height by the height measurement sensor;
The traffic control device according to claim 1 . The height measurement sensor is a sensor that detects whether or not the height of the person passing through the passage is equal to or greater than a predetermined height.
The traffic control device according to claim 1 . Further, a door provided on the exit side of the passage;
A door control unit that controls opening and closing of the door,
the processor closes the door by the door control unit when the posture of the passerby is such that the height measurement sensor cannot accurately measure the height of the passerby.
The traffic control device according to claim 1 . Further, the image sensor has a memory for storing the image.
the processor stores in the memory an image of the posture of the passerby when the posture of the passerby is such that the height of the passerby cannot be accurately measured by the height measurement sensor;
The traffic control device according to claim 1 . Further, a display for displaying guidance is provided,
the processor displays on the display a guidance screen that guides the pedestrian to pass through another passage when the pedestrian's posture is such that the height measurement sensor cannot accurately measure the pedestrian's height.
The traffic control device according to claim 1 . Further, a display for displaying guidance is provided,
the processor displays, on the display, a guide for correcting posture when passing through the passage when the posture of the passerby is such that the height measurement sensor cannot accurately measure the height of the passerby.
The traffic control device according to claim 1 . Further, the device has a speaker for outputting sound,
the processor outputs an alarm through the speaker when the posture of the passerby is such that the height of the passerby cannot be accurately measured by the height measurement sensor.
The traffic control device according to claim 1 .