Information processing apparatus, information processing method, and program

A single-camera system for train door monitoring reduces equipment and processing costs by analyzing images from a diagonal overhead view to detect people and foreign objects in danger zones, addressing the inefficiencies of multiple-camera systems.

JP2026032684APending Publication Date: 2026-02-27ODAKYU ELECTRIC RAILWAY +1
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
JP2024135453
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2024-08-14
Publication Date
2026-02-27

AI Technical Summary

Technical Problem

Conventional vehicle door monitoring systems require multiple cameras and extensive image processing, leading to large equipment, high costs, and heavy processing loads.

Method used

An information processing device that uses a single camera positioned diagonally above a train to capture images of the train and platform, performing image analysis to detect people and foreign objects in a danger area, reducing the need for multiple cameras and processing resources.

Benefits of technology

Effectively detects foreign objects and people using a smaller number of cameras, enabling a more compact and cost-effective monitoring system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2026032684000001_ABST
    Figure 2026032684000001_ABST
Patent Text Reader

Abstract

To reduce the number of cameras.SOLUTION: The information processing device includes a control unit configured to acquire captured screen data captured at an angle of view at which a train stopped at a platform is looked down from obliquely above and a side surface of at least one car of the train on the platform side and a part of the platform on a side of the side surface are captured, and detect at least one of a person present in a dangerous region including the part of the platform in a captured screen and a foreign object sandwiched between any of a plurality of doors provided on the side surface by image analysis of the captured screen.SELECTED DRAWING: Figure 6
Need to check novelty before this filing date? Find Prior Art

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] There is a vehicle door monitoring system that acquires camera images taken by an external camera around a vehicle door, detects the door from the camera image, sets a judgment area based on the detected position of the door, acquires an image of the judgment area from the camera image, judges whether or not a foreign object is caught in the door based on the image of the judgment area, and issues an alarm output based on the judgment result (for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2018-167594 Summary of the Invention [Problem to be solved by the invention]

[0004] The above-mentioned conventional technology requires a camera for each door, which requires multiple cameras and processing of the images captured by each camera, which results in problems such as large equipment and processing scale, high costs, and heavy loads.

[0005] An object of the present disclosure is to provide an information processing device, an information processing method, and a program that can suitably detect people and foreign objects using a small number of cameras. [Means for solving the problem]

[0006] One aspect of the present disclosure is an information processing device that includes a control unit that acquires image data captured from an angle of view that looks down diagonally from above at a train stopped at a platform and captures the side of at least one car of the train facing the platform and a portion of the platform to the side of that side, and that performs image analysis of the image data to detect at least one of a person present in a danger area that includes a portion of the platform within the image data and a foreign object caught between one of multiple doors provided on the side.

[0007] One aspect of the present disclosure is an information processing method in which an information processing device acquires image data captured from an angle of view that looks down diagonally from above at a train stopped at a platform and captures the side of at least one car of the train facing the platform and a portion of the platform to the side of that side, and detects at least one of a person present in a danger area including a portion of the platform within the image data and a foreign object caught between one of multiple doors provided on the side, through image analysis of the image data.

[0008] One aspect of the present disclosure is a program that causes a computer of an information processing device to execute the steps of: acquiring image data captured from an angle of view that looks down diagonally from above at a train stopped at a platform and captures the side of at least one car of the train facing the platform and a portion of the platform to the side of that side; and detecting, through image analysis of the image data, at least one of a person present in a danger area including a portion of the platform within the image data, and a foreign object caught between one of multiple doors provided on the side.

[0009] Aspects of the present disclosure include a non-transitory storage medium on which the above-described program is recorded, and a It may include an information processing system including a processing device. [Effects of the Invention]

[0010] According to the present disclosure, foreign matter can be suitably detected using a small number of cameras. [Brief explanation of the drawings]

[0011] [Figure 1] 1A and 1B are diagrams illustrating the installation position and angle of view of the camera relative to the train. [Figure 2] 2A and 2B are diagrams illustrating the danger area. [Figure 3] FIG. 3 is a diagram illustrating an example of the configuration of the information processing device according to the embodiment. [Figure 4] FIG. 4 is a table showing objects and conditions detected by image analysis. [Figure 5] FIG. 5 is a diagram illustrating train detection, door status, train status, person status, and object status, as well as abnormal events resulting from combinations of these statuses. [Figure 6] FIG. 6 is a flowchart illustrating an example of processing by the information processing device. [Figure 7] FIG. 7 is a flowchart illustrating an example of processing by the information processing device. [Figure 8] 8A and 8B are diagrams showing an example of a screen in which the results of image analysis are superimposed on a captured image. [Figure 9] FIG. 9 is a diagram showing an example of a screen in which the result of image analysis is superimposed on a captured image. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, an embodiment of an information processing device will be described with reference to the drawings. The configuration of the embodiment is an example, and the configuration of the present disclosure is not limited to the configuration of the embodiment. The information processing device according to the embodiment is an information processing device that can detect a person in a danger area (danger zone) or a foreign object caught in a vehicle door by analyzing an image captured by a camera installed outside the vehicle, such as on a platform at a railway station.

[0013] 1A and 1B are diagrams illustrating the installation position and angle of view of a camera relative to a train. FIGS. 2A and 2B are diagrams illustrating danger areas. FIG. 3 is a diagram illustrating an example of the configuration of an information processing device. FIG. 1A is a diagram schematically illustrating a train (formation) 1 stopped at a station platform 20, viewed from the right side. FIG. 1B is a diagram schematically illustrating the platform 20 and train 1 shown in FIG. 1A, viewed from above. FIG. 2A is a diagram schematically illustrating the platform 20 and train 1 shown in FIG. 1A, viewed from the front. FIG. 2B is a diagram schematically illustrating a portion of the platform 20 and train 1 shown in FIG. 2A, viewed from above. In the example shown in FIGS. 1A and 1B, the direction indicated by the arrow (from bottom to top in FIGS. 1A and 1B) is the traveling direction of the train 1, and the end of the train 1 in the traveling direction is the front. However, the traveling direction of the train 1 may be opposite to the arrow.

[0014] A train 1 is made up of one or more cars (railroad cars) 2. In the examples shown in Figures 1A and 1B and Figures 2A and 2B, the train 1 is illustrated as including a car 2a and a car 2b coupled to the car 2a. However, the number of cars 2 that make up the train is arbitrary.

[0015] Doors 11 for people (passengers, etc.) to get on and off are provided on the side 3 of the vehicle 2. In the example shown in Figures 1A and 1B and Figures 2A and 2B, four doors 11 are provided at predetermined intervals on one side 3. Similar doors 11 are also provided on the opposite side, but are not shown in the figure. However, the number of doors 11 provided on the side 3 of the vehicle 2 is arbitrary.

[0016] The door 11 opens and closes an opening used as an entrance and exit for passengers, etc. The door 11 includes two rectangular door pieces that move in opposite directions in response to the opening and closing of the door 11. The door pieces move (slide) in the direction of approach in response to the opening operation, and abut at approximately the center of the opening to close the opening. In contrast, during the closing operation, each door piece moves (slides) in the opposite direction toward the door pocket and is stored in the door pocket. This opens the opening and makes it usable as a boarding / alighting entrance. However, the door 11 may have a form other than the above example, such as a single door, and the door may be opened or closed automatically or manually.

[0017] In addition, a side light 12 is provided for each car on the side 3 of the car 2. In the examples shown in FIGS. 1A and 1B and 2A and 2B, one side light 12 is installed in the upper center of the side 3. However, the number and installation locations of the side lights 12 can be set as appropriate. The side lights 12 are turned off when all doors of one car of the car 2 are closed and turned on if at least one door is open. For example, the side lights 12 are turned on while passengers are getting on and off when the train 1 is stopped (parked) and turned off while the train is moving. Meanwhile, the side lights 12 are turned on when an object with a predetermined thickness (e.g., a thickness exceeding 10 mm) is caught in the door, and are turned off otherwise. Thus, by detecting that the side lights 12 are on when the train 1 is stopped, it is possible to detect a foreign object being caught in the door.

[0018] As shown in FIGS. 1A and 1B, a camera 36 is installed on an extension of the platform 20 (to the right of the train 1 as viewed from above, or on the platform 20). The camera 36 has a field of view (angle of view) 25 that enables the camera to capture an image of the train 1 stopped beside the platform 20 or traveling alongside the platform 20 from an oblique downward angle. This angle of view 25 allows the side 3 of the train 1 facing the platform 20 to be included within the screen, across multiple cars 2, as shown in FIGS. 8A and 8B and 9. The angle of view 25 is set so that the entire side of at least one car 2 of the train 1 (excluding the portion hidden by the platform 20) is included within the screen. In this embodiment, the angle of view 25 is set so that the side surfaces 3 of two cars 2 of the train 1 are included within the screen.

[0019] As shown in Figures 2A and 2B, the platform 20 has an edge 21 that is approximately parallel to the side surface 3 of the train 1, with a small gap between the edge 21 and the side surface 3. A line that is approximately parallel to the edge 21 is provided a certain distance inward from the edge 21 into the platform 20. The line may be, for example, a straight line or a curved line. The line may be, for example, the edge of a white line that prohibits people from entering the side of the edge 21 when the train 1 is moving (arriving or departing), or the edge of a guiding block for visually impaired people that is arranged in a straight line at a certain distance from the edge 21. The line may also be other than these. Figures 2A and 2B illustrate, as an example, a guiding block band 23 on the platform 20, where multiple guiding blocks for visually impaired people are arranged in a straight line. An edge 23a of the guiding block band 23 on the edge 21 side forms a straight line that is a certain distance from the edge 21 of the platform 20. The area including the top surface of the platform 20 between the straight line formed by the edge 23a and the straight line formed by the edge 21 is defined (prescribed) as a danger area (danger zone) 40. The line approximately parallel to the edge 23a may be a line forming the edge of a wide line (strip-like) such as the guide block band 23 or a white line, or may be an invisible line such as the center line of the guide block band 23 or a white line. Furthermore, the guide block band 23 or the white line may be a continuous line (strip-like) or may be an intermittently drawn line (strip-like).

[0020] The danger area 40 can be defined to include a space whose cross section, with the top surface of the platform 20 as its base, is a triangular prism or a square prism. For example, in the example shown in Figures 2A and 2B, the danger area 40 is defined as a space whose cross section is a right-angled triangular prism. The right-angled triangle has a base a, which is a point on the line forming the edge 23a and a line extending from this point in a direction perpendicular to the edge 21 (side surface 3), and a side b that rises vertically along the end point of this base on the edge 21 side (an end point on the platform 20 side, i.e., a point near the edge 21; for example, a point on the edge 21, a point slightly inside the edge 21, or a point provided on the side surface 3), and extends to a predetermined height on the side surface 3, and a line b that connects the upper end point of this vertical line to a point on the edge 23a. It consists of hypotenuse c.

[0021] FIG. 3 is a diagram showing an example of the configuration of an information processing device. In FIG. 3, the information processing device 30 can be configured using a dedicated or general-purpose information processing device (computer) such as a personal computer (PC), a workstation (WS), or a server machine. However, the information processing device 30 may also be a collection (cloud) of one or more computers. The information processing device 30 may be mounted on the train 1 or installed outside the train 1. The information processing device 30 may be a fixed terminal or a portable terminal.

[0022] The information processing device 30 includes a control device 31 (controller: an example of a control unit), a communication interface 33 (communication IF 33), an input device 34, an output device 35, and a camera 36, ​​which are interconnected via a bus B.

[0023] The control device 31 includes a processor and a storage device. The storage device includes a main storage device and an auxiliary storage device. The main storage device is used as at least one of a storage area for programs and data, a program development area, a program work area, and a buffer area for communication data. The main storage device is configured with RAM (Random Access Memory) or a combination of RAM and ROM (Read Only Memory). The auxiliary storage device is used as a storage area for data and programs. A non-volatile storage medium is used as the auxiliary storage device. Examples of non-volatile storage media include a hard disk, a solid state drive (SSD), and a flash memory. or EEPROM (Electrically Erasable Programmable Read-Only Memory) The storage device can include a drive device for a disk recording medium.

[0024] The communication IF 33 is a circuit that performs communication processing. For example, the communication IF 33 is a network interface card (NIC). The communication IF 33 may also be a wireless communication circuit that performs wireless communication (5G, wireless LAN (Wi-Fi), BLE, etc.). F23 may be a combination of a circuit that performs wired communication processing and a wireless communication circuit.

[0025] The input device 34 includes keys, buttons, pointing devices, touch panels, etc., and is used to input information. The output device 35 is, for example, a display, etc., which displays information and data. The output device 35 may also include an audio output device such as a speaker. The camera 36 is a video camera, which captures video (images) including the side 3 of the train 1 and a portion of the platform 20 on that side 3 side at a predetermined frame rate.

[0026] The processor constituting the control device 31 performs various processes by executing various programs stored in the storage device. The processor is, for example, a Central Processing Unit (CPU). The CPU is also called a Microprocessor Unit (MPU). The processor may have a single processor configuration or a multi-processor configuration. Furthermore, a single physical CPU connected via a single socket may have a multi-core configuration. The processor may include an arithmetic unit with various circuit configurations, such as a Digital Signal Processor (DSP) or a Graphics Processing Unit (GPU). Furthermore, the processor may have a configuration that cooperates with at least one of an integrated circuit (IC), other digital circuit, and analog circuit. The integrated circuit includes an LSI, an application specific integrated circuit (ASIC), and a programmable logic device (PLD). A PLD may be, for example, a CP LD, Field-Programmable Gate Array (FPGA). This also includes what are called microcontrollers (MCUs), SoCs (System-on-a-chips), system LSIs, chipsets, etc. The entities that execute the processing of the above-mentioned processors, ASICs, PLDs, MCUs, SoCs, chipsets, etc. are examples of "circuitry."

[0027] The processor executes the program stored in the storage device, thereby controlling the control device 31. can operate as a device including a discriminator 32. The discriminator 32 is, for example, a discriminator 32 that has undergone machine learning using image data as training data. The control device 31 uses data of captured images (video) captured by the camera 36 to perform image analysis (feature extraction, comparison, etc.) of the captured images, and can detect the presence and status of the train 1, the door 11, the danger area 40, people, and objects. From the detection results, the discriminator 32 can determine whether a person has entered the danger area 40 under specified conditions, and whether a foreign object has been caught in the door 11 (door entrapment).

[0028] FIG. 4 is a table showing the objects and states detected by image analysis. The control device 31 of the information processing device 30 uses a discriminator 32 to detect the train 1, the door 11, people, and objects. With regard to the train 1, the presence or absence of the train 1 on the platform 20 and the state of the train 1 (stopped or running (moving)) are detected. With regard to the door 11, the presence or absence of the door 11 and the state of the door 11 (open or closed) are detected. With regard to people, it is detected whether they are inside or outside the danger area. With regard to objects, the presence or absence of objects on the side of the vehicle is detected.

[0029] FIG. 5 is a diagram illustrating train detection, door status, train status, person status, and object status, as well as abnormal events resulting from combinations of these statuses. As shown in FIG. 4, the control device 31 (discriminator 32) can detect each of the states of train detection, door status, train status, person status, and object status shown in FIG. 5, and each state has two or more statuses. The control device 31 (discriminator 32) detects the occurrence of an abnormal event under predetermined conditions (a specific combination of statuses). The abnormal events consist of an abnormal person event and an abnormal object event. The abnormal person event indicates the presence of a person in a danger area, and the abnormal object event indicates an object caught in the door 11.

[0030] 5, the control device 31 detects the occurrence of an abnormal person event when the train state is a "running (arrival)" state and the person state is a "danger area" (a state in which a person is in the danger area 40). The control device 31 also detects the occurrence of an abnormal person event when the train state is a "running (departure)" state and the person state is a "danger area" (a state in which a person is in the danger area 40).

[0031] Furthermore, the control device 31 detects the occurrence of an abnormal object event when the train state is "stopped," the door state is "doors closed by moving in the closing direction," and the object state is "object trapped in the door" (a state in which an object is present on the side surface 3 of the vehicle 2). Furthermore, the control device 31 detects the occurrence of an abnormal object event when the train state is "stopped," the door state is "doors stopped and closed," and the object state is "object trapped in the door" (a state in which an object is present on the side surface 3 of the vehicle 2). When the control device 31 detects an abnormal event (an abnormal person event or an abnormal object event), it may output (issue) an alarm. The alarm may be issued by displaying information indicating the abnormal event on the output device 35 (display), or by outputting a sound indicating the abnormal event from the output device 35. Furthermore, information or a signal indicating the occurrence of the abnormal event may be transmitted to another device using the communication IF 33, and the other device may issue an alarm.

[0032] 6 and 7 are explanatory diagrams of an example of the operation of the control device 31 (including the discriminator 32) of the information processing device 30. In FIG. 6, when a captured image (video frame) captured by the camera 36 is input, the control device 31 detects a person from the captured image through image analysis and tracks the person (step S01). At this time, the control device 31 defines a rectangle (person box 50) surrounding the detected person (see FIGS. 8A and 8B and 9). The coordinates of the person box 50 on the captured screen change in accordance with the movement of the person.

[0033] In step S02, the control device 31 receives the coordinate information of the person box 50 and the The occurrence of an abnormal person event is determined based on information indicating the coordinates (pixels) of the danger area 40 in the captured image obtained by the detector 32 and information indicating the door state (open or closed) obtained by the detector 32.

[0034] In step S03, when the doors are closed (= the train is moving) and there is an overlap between the coordinates of the person box and the coordinates of the danger area 40, the control device 31 determines that there is a person in the danger area 40 and detects the occurrence of an abnormal person event (step S03). At this time, the control device 31 outputs an alarm, etc., as necessary.

[0035] In the process shown in Fig. 6, the control device 31 controls the display of the image (video RAM), displays the captured image on the screen of the display included in the output device 35, and can display the above-mentioned person box 50 surrounding the person in the captured image (see Figs. 8A and 8B and 9). The person box can be displayed, for example, in red (shown by a two-dot chain line in Figs. 8A and 8B and 9) when the person is inside the danger area 40, and can be displayed, for example, in green (shown by a dashed line in Figs. 8A and 8B and 9) when the person is outside the danger area 40 (safe area). By visually recognizing the person box 50 displayed in red, a crew member or manager can recognize the occurrence of an abnormal person event.

[0036] In step S001 in FIG. 7, when a captured image (video frame) captured by the camera 36 is input, the control device 31 detects one or more doors 11 from the captured image through image analysis and detects the door state (open or closed). The door state obtained here can be used in the processing of step S02 shown in FIG. 6. In addition, a door box 51 surrounding the door 11 detected in the captured image is defined using pixel coordinates of the captured image. The control device 31 can display a rectangle representing the door box 51 in a specific color (e.g., white) on the screen displaying the captured image (in FIGS. 8A and 8B and FIG. 9, the door box 51 is displayed with a solid line).

[0037] In step S002, the control device 31 checks the train state. In step S003, if the train state is a stopped state, the control device 31 crops a door box 51 (specifies one of the multiple door boxes 51).

[0038] In step S004, the control device 31 identifies the door state and confirms that the door is closed, and in step S005, it segments all objects within the door box 51 (generates object masks).

[0039] In step S006, the control device 31 performs a process to remove object masks that are not on the center line of the door 11 (where the door pieces abut) in the captured image. As a result, it is determined that the object corresponding to the object mask to be removed is not an object caught in the door 11. In contrast, the control device 31 determines that the object mask on the center line of the door 11 is an object caught in the door 11 (object caught in the door), and detects the object (step S007). In other words, the occurrence of an abnormal object event (object caught in the door) is detected. The control device 31 can output an alarm, etc., as necessary. The processes of steps S004 to S007 are performed for each door box 51.

[0040] 8A and 8B, and FIG. 9 show examples of screen displays in which the results of image analysis are superimposed on a captured image. In the examples shown in FIGS. 8A and 8B and 9, a screen (e.g., On the screen displayed on the display, the side 3 of the train 1 and the A part of the platform 20 is displayed. On the screen, the area between the edge 21 of the platform 20 and the white line 24 is defined as a danger area 40, and a frame line indicating the danger area 40 is displayed in a predetermined color (for example, red) (shown as a dashed line in Figs. 8A and 8B and Fig. 9). In addition, a person box 50 that does not overlap with the danger area 40 is displayed in green (dashed line), and the danger area A person box 50 that overlaps with the person 40 is displayed in red (two-dot chain line). The multiple doors 11 provided on the side surface 3 are displayed surrounded by door boxes 51. Door boxes 51 are displayed for all doors 11 detected from the captured image. In Figures 8A and B and Figure 9, the door 11 on the foremost vehicle 2a and the four doors 11 on the vehicle 2b following vehicle 2a are detected, and door boxes 51 are displayed.

[0041] In the example shown in FIG. 8A, a person 60 is standing in the danger area 40, holding an umbrella (foreign object 61) caught in the door 11. In this case, the person box 50 of the person 60 is displayed in red (two-dot chain line), allowing a viewer to recognize (detect) the presence of a person in the danger area 40 and that the person is caught in the door. Note that the door box 51 can display text indicating the status of the door 11, such as "door close." Furthermore, the screen can display text indicating the status of the train 1, such as "train_stop." In the examples shown in FIGS. 8A and 8B and 9, a white line 24 is drawn between the edge 21 of the platform 20 and a guiding block strip 23 for visually impaired persons (FIG. 8A). The inside of the white line 24 (the side of the edge 21) is used as a no-go area while the train 1 is moving. The danger area 40 is, for example, defined between the center line of the white line 24 and the edge 21 or the side surface 3 of the vehicle 2. In FIG. 8B, a person 60 is displayed on all fours in the danger area 40. In this case, the person box 50 of the person 60 is displayed in red (two-dot chain line), so that a viewer of the screen can recognize (detect) the presence of the person 60 in the danger area 40. Also, in FIG. 9, a bag is displayed that is caught in the door 11 and sticking out. The object caught in the door (foreign object 61: umbrella in FIG. 8A, bag in FIG. 9) can be displayed in a manner that makes it easy for a viewer of the screen to recognize, for example, by highlighting. Image processing (image display control) such as highlighting the person box 50, door box 51, and object caught in the door can be performed by the control device 31 or a device other than the control device 31.

[0042] According to the information processing device (method for detecting a person in a danger zone and a person trapped in a door) according to the embodiment, a control device 31 (controller) obtains captured image data by looking diagonally down at a train 1 moving or stopped beside a platform 20 from above, with an angle of view that captures the side 3 of at least one car of the train 1 facing the platform 20 and a portion of the platform 20 to the side of the side 3. By analyzing the captured image, the control device 31 can detect at least one of a person present in a danger zone 40 including a portion of the platform 20 within the captured image, and a foreign object trapped between one of multiple doors 11 provided on the side 3. This allows a single camera to determine whether or not a person is trapped between multiple doors. This means that the number of cameras can be reduced, enabling a smaller-scale monitoring system than conventional technology to be constructed. Furthermore, a person present in a danger zone can be detected on the side of at least one car.

[0043] The control device 31 (control unit) can use image analysis to detect a person in the danger area 40 when the train 1 is moving or when the train doors 11 are closed. The control device 31 can also use image analysis to detect that an object is caught in the doors 11 when the train 1 is stopped.

[0044] In the captured image, the danger area 40 can include the top surface of the platform 20 sandwiched between the side 3 of the train 1 or the edge 21 of the platform 20 and a straight line 23a that is approximately parallel to the edge 21 and is located a certain distance inward from the edge 21 of the platform 20.

[0045] The danger area 40 is a vertical line extending from the end point of the base a on the platform 20 side to a predetermined height on the side 3 of the train 1. The base a extends horizontally from the straight line 23a to the edge 21 of the platform 20 or the side 3 of the train 1. It can include a triangular prism-shaped space having a cross section of a right triangle abc formed by a side b and a hypotenuse c connecting the upper end point of side b and the end point of the base a on the straight line 23a side.

[0046] The control device 31 can output an alarm when a person or foreign object is detected, i.e., display information indicating the alarm (such as characters, figures, or symbols) on the display or sound an alarm from the speaker. By outputting an alarm, the driver or manager can be notified of an abnormality. In image analysis, the control device 31 can determine that the train 1 is stopped or that a foreign object is caught in one of the multiple doors 11 when the side lights 12 of the train 1 are on, and can determine that the train 1 is running when the side lights 12 are off.

[0047] The control device 31 displays, on the captured image, at least one of a first display indicating the danger area 40, a second display (door box) indicating the door 11, and a third display (person box) indicating a person present on the platform 20, and can display a person present in the danger area 40 and a foreign object caught between the door 11 in a manner different from the first to third displays (FIGS. 8A and B, FIG. 9). This makes it possible to display a person or object that is the subject of an abnormal event with good visibility. The configuration of the above-described embodiment can be modified as appropriate without departing from the scope of the invention. [Explanation of symbols]

[0048] 1 train, 3 side, 11 door, 12 side light, 20 platform, 21 edge, 23 guide block strip, 23a edge, 25 field of view (angle of view), 30 information processing device, 31 control device, 36 camera, 40 danger area

Claims

1. Obtaining image data obtained by looking down from diagonally above a train moving or stopped beside a platform, and capturing an image of at least one carriage of the train on the platform side at an angle of view that captures an image of the side of the train for at least one carriage and a part of the platform to the side of that side; By analyzing the image of the captured image, at least one of a person present in a dangerous area including a part of the platform in the captured image and a foreign object caught between any of a plurality of doors provided on the side is detected. A control unit that executes An information processing device comprising:

2. The control unit detects a person in the danger area by the image analysis when the train is running or the doors of the train are closed. The information processing device according to claim 1.

3. The control unit detects, by the image analysis, that an object is caught in the door while the train is stopped. The information processing device according to claim 1 .

4. The danger area includes, in the captured image, the upper surface of the platform sandwiched between the side of the train or the edge of the platform and a line approximately parallel to the edge, which is located a certain distance inward from the edge of the platform. The information processing device according to claim 1 .

5. The danger area includes a triangular prism-shaped space with a cross section of a right triangle consisting of a base that extends horizontally from the straight line toward the edge of the platform or the side of the train, a perpendicular line that rises vertically from the end point of the base on the platform side to a predetermined height on the side of the train, and a hypotenuse that connects the upper end point of the perpendicular line and the end point of the base on the straight line side, when viewed from the front of the train. The information processing device according to claim 4 .

6. The information processing apparatus according to claim 1 , wherein the control unit performs processing to output an alarm when the person or the foreign object is detected.

7. When the image analysis shows that the side lights of the train are on, the control unit determines that the train is stopped or that a foreign object is caught in one of the plurality of doors, and when the side lights of the train are off, the control unit determines that the train is running.

4. The information processing device according to claim 2 or 3.

8. The control unit displays, on the captured image, at least one of a first display indicating the danger area, a second display indicating the door, and a third display indicating a person present on the platform, and displays the person present in the danger area and a foreign object caught in the door in a manner different from the first to third displays. The information processing device according to claim 1 .

9. The information processing device Obtaining image data obtained by looking down diagonally from above at a train stopped at a platform and capturing an image of at least one car of the train on the platform side and a part of the platform to the side of the side; By analyzing the image of the captured image, at least one of a person present in a dangerous area including a part of the platform in the captured image and a foreign object caught between any of a plurality of doors provided on the side is detected. An information processing method that performs the above.

10. The computer of the information processing device Obtaining image data obtained by looking down diagonally from above at a train stopped at a platform and capturing an image of at least one car of the train on the platform side and a part of the platform to the side of the side; detecting at least one of a person present in a dangerous area including a part of the platform in the captured image and a foreign object caught between any of a plurality of doors provided on the side by analyzing the image of the captured image; A program that executes the following.

Citation Information

Patent Citations

  • Vehicle door monitoring system and vehicle door monitoring method

    JP2018167594A