Monitoring system, monitoring device, monitoring method, and program

Abstract

To provide a technique for appropriately determining a level of safety of a person at a platform.SOLUTION: A monitoring system 1 includes: an imaging section 10 for imaging a platform PF of a station ST; a tracking section 20 for detecting a person at the platform PF based on an image acquired by the imaging section 10; and a determination section 30 for determining a level of safety of the person in response to a position of the person detected by the tracking section 20. The determination section 30 determines that the safety of the person is high when the person is detected in an area R1 which is most separated from one end part of the platform PF to which a railway vehicle TRN enters among areas R1 to R3 of the platform PF, determines that the safety of the person is low when the person is detected in the area R3 closest to the one end part, determines that the safety of the person is high when the person is detected in the area R2 and the person exists in the area R2 after moving from the area R1, and determines that the safety of the person is low when the person exists in the area R2 after moving from the area R3.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present disclosure relates to surveillance systems and the like. [Background technology]

[0002] 2. Description of the Related Art Conventionally, there is known a technique for using an imaging unit (camera) to capture an image of a station platform and monitor people on the station platform (see, for example, Patent Document 1).

[0003] In Patent Document 1, camera images are used to monitor the movements of people on the platform, and when a person enters a specified area within the platform, it is determined that there is a danger, and a voice announcement is made of a message to the person determined to be in danger, or the station staff room is notified of the danger. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2003-224844 A Summary of the Invention [Problem to be solved by the invention]

[0005] However, when a person is near the boundary of a predetermined area, the judgment result regarding the safety of the person may change frequently, and therefore, for example, station staff may not be able to take appropriate action based on the judgment result.

[0006] In view of the above-mentioned problems, an object of the present invention is to provide a technology that can more appropriately determine the level of safety of people on the platform. [Means for solving the problem]

[0007] In order to achieve the above object, in one embodiment of the present disclosure, An imaging unit that captures images of the station platform; a first detection unit that detects a person present on the platform based on an image acquired by the imaging unit; a determination unit that determines whether the safety of the person is in a first state in which the safety of the person is relatively high or in a second state in which the safety of the person is relatively low, depending on the position of the person detected by the first detection unit; the determination unit determines that the safety of a person is in the first state when the first detection unit detects a person in the first area, which is the farthest from one end of the platform where railway vehicles enter, among a first area, a second area, and a third area defined on the platform; determines that the safety of a person is in the second state when the first detection unit detects a person in the third area, which is the closest to the one end; determines that the safety of a person is in the second state when the first detection unit detects a person in the second area between the first area and the third area, and determines that the safety of a person is in the first state when the person moves from the first area to be in the second area; and determines that the safety of a person is in the second state when the person moves from the third area to be in the second area. A monitoring system is provided.

[0008] In another embodiment of the present disclosure, a detection unit that detects a person present on the platform based on an image acquired by an imaging unit that images the platform of the station; a determination unit that determines whether the safety of the person is in a first state in which the safety of the person is relatively high or in a second state in which the safety of the person is relatively low, depending on the position of the person detected by the detection unit; the determination unit determines that the safety of a person is in the first state when the detection unit detects a person in the first region, which is the farthest from one end of the platform where railway vehicles enter, among a first region, a second region, and a third region defined on the platform; determines that the safety of a person is in the second state when the detection unit detects a person in the third region, which is the closest to the one end; determines that the safety of a person is in the first state when the detection unit detects a person in the second region between the first region and the third region, and determines that the safety of a person is in the first state when the person moves from the first region and is in the second region; and determines that the safety of a person is in the second state when the person moves from the third region and is in the second region. A monitoring device is provided.

[0009] In still another embodiment of the present disclosure, A detection step in which the information processing device detects a person present on the platform based on an image acquired by an imaging unit that images the platform of the station; a determination step of determining, by the information processing device, whether the safety of the person is in a first state in which the safety of the person is relatively high or in a second state in which the safety of the person is relatively low, depending on the position of the person detected by the detection step; In the determination step, if a person is detected by the detection step in the first area, which is the farthest from one end of the platform where railway vehicles enter, among a first area, a second area, and a third area defined on the platform, the safety of the person is determined to be in the first state; if a person is detected by the detection step in the third area, which is the closest to the one end, the safety of the person is determined to be in the second state; if a person is detected by the detection step in the second area which is between the first area and the third area, the safety of the person is determined to be in the first state when the person moves from the first area to be in the second area; and if the person moves from the third area to be in the second area, the safety of the person is determined to be in the second state. A monitoring method is provided.

[0010] In still another embodiment of the present disclosure, In the information processing device, a detection step of detecting a person present on the platform based on an image acquired by an imaging unit that images the platform of the station; a determining step of determining whether the safety of the person is in a first state in which the safety of the person is relatively high or in a second state in which the safety of the person is relatively low, depending on the position of the person detected by the detecting step; In the determination step, if a person is detected by the detection step in the first area, which is the farthest from one end of the platform where railway vehicles enter, among a first area, a second area, and a third area defined on the platform, the safety of the person is determined to be in the first state; if a person is detected by the detection step in the third area, which is the closest to the one end, the safety of the person is determined to be in the second state; if a person is detected by the detection step in the second area which is between the first area and the third area, the safety of the person is determined to be in the first state when the person moves from the first area to be in the second area; and if the person moves from the third area to be in the second area, the safety of the person is determined to be in the second state. The program is provided. Effect of the Invention

[0011] According to the above-described embodiment, the safety of people on the platform can be more appropriately determined. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 illustrates an example of a monitoring system. [Diagram 2] FIG. 2 is a diagram illustrating an example of a main flow chart of a process relating to monitoring of people in a home by a monitoring system. [Diagram 3] FIG. 2 is a diagram illustrating a first example of a method for monitoring people in a home by the monitoring system. [Figure 4]FIG. 2 is a diagram showing a first example of a sub-flowchart of a process relating to monitoring of occupants in the home by the monitoring system. [Diagram 5] FIG. 11 is a diagram illustrating a second example of a method for monitoring people in a home by the monitoring system. [Figure 6] FIG. 13 is a diagram illustrating a second example of a sub-flowchart of a process relating to monitoring of occupants in the home by the monitoring system. [Figure 7] FIG. 11 is a diagram illustrating a third example of a method for monitoring people in a home by the monitoring system. [Figure 8] FIG. 13 is a diagram illustrating a third example (part 1) of a sub-flowchart of a process related to monitoring of people in the home by the monitoring system. [Figure 9] FIG. 13 is a diagram illustrating a third example (part 2) of a sub-flowchart of the process related to monitoring of people in the home by the monitoring system. [Figure 10] FIG. 11 is a diagram for explaining a fourth example of a method for monitoring people in a home by the monitoring system. [Figure 11] FIG. 11 is a diagram for explaining a fifth example of a method for monitoring people in a home by the monitoring system. [Figure 12] FIG. 13 is a diagram for explaining a sixth example of a method for monitoring people in a home by the monitoring system. [Figure 13] FIG. 13 is a diagram illustrating a sixth example of a sub-flowchart of a process relating to monitoring of people in the home by the monitoring system. [Figure 14] FIG. 13 is a diagram illustrating a sixth example of a sub-flowchart of a process relating to monitoring of people in the home by the monitoring system. [Figure 15] FIG. 13 is a diagram illustrating another example of a monitoring system. [Figure 16] 13 is a flowchart generally illustrating a seventh example of a sub-flowchart of a process related to monitoring of people in the home by the monitoring system. [Figure 17] 13 is a flowchart outlining an eighth example of a sub-flowchart of a process relating to monitoring of people in the home by the monitoring system. [Figure 18] FIG. 4 is a diagram showing a first example of a screen displayed on the display unit. [Figure 19] FIG. 11 is a diagram showing a second example of a screen displayed on the display unit. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Hereinafter, an embodiment will be described with reference to the drawings.

[0014] [Example of a surveillance system] An example of a monitoring system 1 will be described with reference to FIG.

[0015] FIG. 1 is a diagram showing an example of the configuration of a monitoring system 1. As shown in FIG.

[0016] The monitoring system 1 is installed at a station ST and monitors the status of a platform PF where a railway vehicle TRN enters and stops.

[0017] The surveillance system 1 includes an imaging unit 10, a tracking unit 20, a determination unit 30, and a display unit 40.

[0018] The imaging unit 10 captures an image of the platform PF and obtains an image showing the state of the platform PF. The imaging unit 10 may also be capable of obtaining an image including the state of the inside of a railway vehicle TRN entering the platform PF, in addition to the state of the platform PF. The imaging range of the imaging unit 10 is set, for example, so as to include the entire range in the length direction (longitudinal direction) of the platform PF in which the entering railway vehicles TRN are adjacent to each other. The length direction (longitudinal direction) of the platform PF is the direction along the entering railway vehicle TRN.

[0019] The imaging unit 10 is, for example, a monocular surveillance camera permanently installed in the platform PF and attached to the ceiling of the platform PF. For example, the imaging unit 10 captures an image showing the latest situation of the platform PF every 1 / 30 seconds during the business hours of the station ST and outputs the data.

[0020] The monitoring system 1 may be capable of capturing an image of the entire monitoring range of the home PF using one imaging unit 10, or may be capable of covering the entire monitoring range of the home PF using a plurality of imaging units 10.

[0021] The imaging unit 10 is communicatively connected to the tracking unit 20 via a one-to-one communication line, a local network (LAN: Local Area Network), or the like, and the output (image data) of the imaging unit 10 is taken into the tracking unit 20.

[0022] The tracking unit 20 performs image analysis based on the output (image data) of the imaging unit 10, detects a person at the platform PF, and tracks the movement of the person within the imaging range corresponding to the image. In addition, when the imaging range of the imaging unit 10 includes the inside of the railway vehicle TRN, the tracking unit 20 may detect a person inside the railway vehicle TRN and track the movement of the person from the platform PF to inside the railway vehicle TRN.

[0023] The tracking unit 20 uses known image processing techniques to recognize people from within the image, thereby detecting people in the home PF and estimating their positions.

[0024] The function of the tracking unit 20 may be realized by any hardware or a combination of any hardware and software. For example, the tracking unit 20 is mainly configured with a computer (information processing device) including a central processing unit (CPU), a memory device, an auxiliary storage device, a high-speed arithmetic device, and an interface device. The memory device is, for example, a static random access memory (SRAM) or a dynamic random access memory (DRAM). The auxiliary storage device is, for example, a hard disc drive (HDD), a solid state drive (SSD), a flash memory, or an electrically erasable programmable read-only memory (EEPROM). The high-speed arithmetic device includes, for example, a graphics processing unit (GPU), a field-programmable gate array (FPGA), or an application specific integrated circuit (ASIC). The interface device includes, for example, a communication interface for communicating with other components of the monitoring system 1 such as the imaging unit 10 and the determination unit 30. The interface device also includes, for example, an external interface for connecting to a recording medium. This allows the tracking unit 20 to, for example, retrieve various programs from a recording medium via an external interface and install them in the auxiliary storage device.

[0025] For example, the tracking unit 20 applies a trained model based on local features obtained from the image data or the image data itself to recognize a person in the image and estimate the person's position in the home PF. The trained model is obtained by, for example, using a teacher data set to perform machine learning (supervised learning) on ​​a base learning model. For example, each piece of teacher data included in the teacher data set is composed of a combination of image features obtained from an image as input data or the image data itself, and data specifying an image portion including a person on the image as output data. As a result, the trained model can recognize a person on the image using image features obtained from the image data or the image data itself as input, and output data of a range (for example, a rectangular range) including the person on the image. For example, the trained model is mainly composed of a deep neural network (DNN), and the efficiency of machine learning of the DNN is improved by applying backpropagation (error backpropagation method) based on the teacher data. The trained model may also include a U-Net that can recognize a person from image data using image data as input.

[0026] In addition, the tracking unit 20 may apply a rule-based algorithm such as pattern matching based on feature points and local feature amounts of image data to recognize people in the image.

[0027] The tracking unit 20 applies a known object tracking method to track the movement path (trajectory) of a person detected on an image, thereby enabling the tracking unit 20 to output not only the current (latest) position of the person on the home PF in the image, but also the history (trajectory) of the person's past positions on the home PF as tracking results.

[0028] For example, the tracking unit 20 applies MOT (Multi Object Tracking) to track the movement of a person detected on an image. The tracking unit 20 may also apply an optical flow method to track the movement of a person detected on an image.

[0029] The tracking unit 20 is communicatively connected to the determination unit 30 via a one-to-one communication line or a local network, and the output of the tracking unit 20 (the position of a detected object and the tracking result) is input to the determination unit 30.

[0030] Based on the output of the tracking unit 20, the judgment unit 30 judges (evaluates) the safety of the person detected by the tracking unit 20 on the current (latest) image.

[0031] The function of the determination unit 30 may be realized by any hardware or a combination of any hardware and software. For example, the determination unit 30 is mainly configured with a computer (information processing device) including a CPU, a memory device, an auxiliary storage device, and an interface device. The memory device is, for example, an SRAM or a DRAM. The auxiliary storage device is, for example, an HDD, an SSD, a flash memory, an EEPROM, and the like. The interface device includes, for example, a communication interface for communicating with other components of the monitoring system 1 such as the tracking unit 20 and the display unit 40. The interface device also includes, for example, an external interface for connecting to a recording medium. Thereby, the determination unit 30 can, for example, retrieve various programs from a recording medium through the external interface and install them in the auxiliary storage device.

[0032] For example, the determination unit 30 determines, for each person detected on the current (latest) image by the tracking unit 20, whether the safety of that person is in a relatively high or low state.

[0033] Furthermore, the determination unit 30 may perform an evaluation of the overall safety of people in a predetermined range of the home PF when the predetermined range of the home PF is viewed as a whole, based on the evaluation result of the safety of each person detected on the current (latest) image by the tracking unit 20. The predetermined range is, for example, the entire range of the home PF (i.e., the monitoring range of the surveillance system 1) included in the imaging range of the imaging unit 10. Furthermore, when the surveillance system 1 includes multiple imaging units 10, the predetermined range may be defined for each of multiple imaging ranges corresponding to the multiple imaging units 10, and may be the range of the home PF included in each imaging range.

[0034] For example, when the determination unit 30 determines that the safety of each of all people in a predetermined range detected by the tracking unit 20 based on the current (latest) image is relatively high, the determination unit 30 determines that the overall safety of the people in the predetermined range is relatively high. On the other hand, when the determination unit 30 determines that the safety of at least one person among all people in a predetermined range detected by the tracking unit 20 based on the current (latest) image is relatively low, the determination unit 30 determines that the overall safety of the people in the predetermined range is relatively low.

[0035] The judgment unit 30 is communicatively connected to the display unit 40 via a one-to-one communication line or a local network, and the output of the judgment unit 30 (judgment results for each detected person and judgment results for all people within a specified range) is input to the display unit 40.

[0036] The functions of the tracking unit 20 and the determination unit 30 may be realized by one information processing device, or may be realized in a distributed manner by three or more information processing devices.

[0037] The display unit 40 is installed on the platform PF and displays information to station staff at the station ST and crew members of the stopped railway car TRN. The display unit 40 is, for example, a liquid crystal display or an organic EL (Electroluminescence) display.

[0038] The display unit 40 displays, for example, the judgment result of the judgment unit 30. As a result, the station staff of the station ST and the crew of the railway vehicle TRN can confirm the judgment result of the safety of each person at the platform PF and the judgment result of the overall safety of the people at the platform PF through the display content of the display unit 40. The display unit 40 may also display the judgment result of the judgment unit 30 together with the output image of the imaging unit 10. As a result, the station staff of the station ST and the crew of the railway vehicle TRN can confirm the judgment result of the safety of each person at the platform PF and the judgment result of the overall safety of the people at the platform PF together with the actual situation of the platform PF through the display content of the display unit 40. Furthermore, when displaying the judgment result of the safety of each person at the platform PF by the judgment unit 30, the display unit 40 may display an image (for example, a frame surrounding the detected person) that identifies the person detected by the tracking unit 20 by superimposing it on the output image of the imaging unit 10. As a result, the station staff of the station ST and the crew of the railway vehicle TRN can confirm the detected person and the judgment result corresponding to that person in association with each other.

[0039] The process of generating an image to be displayed on the display unit 40 may be performed by an information processing device (eg, a microcomputer) mounted on the display unit 40, or may be performed by the determination unit 30.

[0040] The display unit 40 may acquire image data from the imaging unit 10 by being directly connected to the imaging unit 10 so as to be able to communicate with it via a one-to-one communication line or a local network, or may acquire image data from the imaging unit 10 via the tracking unit 20 and the judgment unit 30.

[0041] The monitoring system 1 may be capable of displaying information about the entire monitoring range of the home PF using a single display unit 40, or may cover the entire monitoring range of the home PF using multiple display units 40 so as to be capable of displaying information about the entire monitoring range of the home PF.

[0042] [Surveillance system processing] Next, an overview of the processing of the monitoring system 1 will be described with reference to FIG.

[0043] FIG. 2 is a diagram illustrating an example of a main flowchart of a process related to monitoring people in the home PF by the monitoring system 1. As shown in FIG.

[0044] The flowchart in FIG. 2 is repeatedly executed at predetermined processing intervals during a monitored time period including the operating time period of vehicles at the home PF, for example.

[0045] As shown in FIG. 2, the tracking unit 20 acquires the latest image from the imaging unit 10 in step S10.

[0046] When the process of step S10 is completed, the tracking unit 20 proceeds to step S20.

[0047] In step S20, the tracking unit 20 performs a process of detecting a person in the home PF (detection process) and a process of tracking the movement route (history) of the detected person (tracking process) based on the image data acquired in step S10.

[0048] When the process of step S20 is completed, the tracking unit 20 outputs the results of the detection process and tracking process to the determination unit 30.

[0049] In step S30, the determination unit 30 performs a process (individual determination process) to determine the level of safety of each person detected in the current (latest) image based on the results of the detection process and tracking process of the tracking unit 20.

[0050] When the process of step S30 is completed, the decision unit 30 proceeds to step S40.

[0051] In step S40, the determination unit 30 performs a process (comprehensive determination process) of determining the overall safety of people within a predetermined range of the home PF when the predetermined range is viewed as a whole, based on the determination result in step S30.

[0052] When the process of step S40 is completed, the judgment unit 30 outputs the results of the individual judgment process and the comprehensive judgment process to the display unit 40.

[0053] The process of step S40 may be omitted. In this case, when the process of step S30 is completed, the determination unit 30 outputs the result of the individual determination process to the display unit 40.

[0054] In step S50, the display unit 40 generates and outputs an image based on the results of the determinations in steps S30 and S40.

[0055] This allows the display unit 40 to display an image including the determination results of steps S30 and S40.

[0056] When the process of step S40 is omitted, the display unit 40 generates and outputs an image based on the determination result of step S30.

[0057] When the process of step S50 is completed, the process of this main flowchart ends.

[0058] Thus, in this example, the monitoring system 1 can judge (evaluate) the safety of people at the platform PF based on images from the imaging unit 10, and display the judgment results to station staff at the platform PF, crew members of the railway vehicle TRN, etc.

[0059] [First example of how to monitor people in the nursing home] Next, a first example of a method for monitoring people in the home PF by the monitoring system 1 will be described with reference to Fig. 3 and Fig. 4. Specifically, a first example of a method for determining the safety of people in the home PF by the determination unit 30 will be described.

[0060] <Summary> Fig. 3 is a diagram illustrating a first example of a method for monitoring people in a home PF by the monitoring system 1. Specifically, Fig. 3 is a diagram illustrating transitions in the results of judgments about the level of safety of people M11 to M14 moving around the home PF.

[0061] As shown in FIG. 3, in this example, regions R1 to R3 for determining the safety of people are provided in the home PF.

[0062] Of the regions R1 to R3, the region R1 is the furthest in the width direction of the platform PF from one end where the railway cars TRN enter the track adjacent to each other (hereinafter, for convenience, "one end on the railway car TRN side").

[0063] Of the regions R1 to R3, the region R3 is closest to one end of the platform PF on the side of the railway vehicle TRN in the width direction.

[0064] Of the regions R1 to R3, the region R2 is located between the regions R1 and R3 in the width direction of the home PF.

[0065] In this example, the boundary line B1 between the regions R1 and R2 and the boundary line B2 between the regions R2 and R3 are set parallel to a straight line or curve corresponding to one end of the platform PF on the side of the railway vehicle TRN in a top view.

[0066] Boundary line B1 is set, for example, on the inner side (the farther side as seen from the end of the platform PF on the railroad vehicle TRN side) of the braille block laid along one end of the platform PF on the railroad vehicle TRN side. Boundary line B2 is set, for example, on the outer side (the closer side as seen from the end of the platform PF on the railroad vehicle TRN side) of the braille block.

[0067] When a person is detected in area R1 by tracking unit 20 based on the current (latest) image from imaging unit 10, determination unit 30 determines that the person is highly safe (see the determination results of person M11 and person M12).

[0068] On the other hand, when a person is detected in area R3 by tracking unit 20 based on the current (latest) image from imaging unit 10, determination unit 30 determines that the safety of that person is low (see the determination results of persons M11 and M12).

[0069] Furthermore, when a person is detected in area R2 by the tracking unit 20 based on the current (latest) image of the imaging unit 10, the determination unit 30 determines whether the safety of the person is high or low based on the tracking result (movement path) of the person by the tracking unit 20. Specifically, when the person is moving from area R1 to area R2, the determination unit 30 maintains the determination result of the person's safety in area R1 and determines that the person's safety is high (see the determination result of person M13). On the other hand, when the person is moving from area R3 to area R2, the determination unit 30 maintains the determination result of the person's safety in area R3 and determines that the person's safety is low (see the determination result of person M14).

[0070] In this way, in this example, between the area R1 where the safety of a person is high and the area R3 where the safety of a person is low, the area R2 is provided as a buffer area for switching between high and low safety of a person. Therefore, the surveillance system 1 can suppress a situation in which the safety of a person frequently switches between a high and low state due to small movements of the person who is present in an unstable position between the areas R1 and R3.

[0071] In this example, the railcar TRN enters only one end of the platform PF in the width direction, but may enter both ends. In this case, the determination unit 30 may perform both a person safety determination based on one end and a person safety determination based on the other end in the width direction of the platform PF. The same applies to the second to fifth examples described below.

[0072] <Processing flow> Fig. 4 is a diagram illustrating a first example of a sub-flowchart of a process related to monitoring of a home PF by the monitoring system 1. Specifically, Fig. 4 is a diagram illustrating a first example of a sub-flowchart of the process (individual determination process) of step S30 in Fig. 2.

[0073] The flowchart in Fig. 4 is executed for each of all people detected in the detection process in step S20 in Fig. 2. The same applies to the flowcharts in Figs. 6, 8, and 9.

[0074] 4, in step S102, the determination unit 30 determines which of the regions R1 to R3 the position (coordinates) of the target person detected by the tracking unit 20 corresponds to. If the position (coordinates) of the target person detected by the tracking unit 20 corresponds to region R1, the determination unit 30 proceeds to step S106, if the position (coordinates) of the target person detected by the tracking unit 20 corresponds to region R1, the determination unit 30 proceeds to step S104, if the position (coordinates) of the target person detected by the tracking unit 20 corresponds to region R2, and the determination unit 30 proceeds to step S108, if the position (coordinates) of the target person detected by the tracking unit 20 corresponds to region R3.

[0075] In step S104, the judgment unit 30 judges whether or not the safety of the target person was judged to be high in the previous flowchart. If the judgment unit 30 judges that the safety of the person is high in the previous flowchart, the judgment unit 30 proceeds to step S106, otherwise proceeds to step S108.

[0076] In step S106, the determination unit 30 determines that the safety of the subject person is high.

[0077] On the other hand, in step S108, the determination unit 30 determines that the safety of the subject person is low.

[0078] When the process of step S106 or step S108 is completed, the determination unit 30 proceeds to step S110.

[0079] In step S110, the determination unit 30 outputs the determination result of the safety of the target person.

[0080] When the process of step S110 is completed, the determination unit 30 ends the process of this flowchart.

[0081] In this example, when a person is detected in region R2 by the tracking unit 20, the determination unit 30 determines that the person's safety is high if the person's safety was determined to be high in the previous flowchart, and otherwise determines that the person's safety is low. This allows the determination unit 30 to determine that the person's safety is high if the person has moved from region R1 to region R2, and to determine that the person's safety is low if the person has moved from region R3 to region R2.

[0082] [Second example of how to monitor people in the nursing home] Next, a second example of the method for monitoring people in the home PF by the monitoring system 1 will be described with reference to Fig. 5 and Fig. 6. Specifically, a second example of the method for determining the safety of people in the home PF by the determination unit 30 will be described.

[0083] The following description will focus on the differences from the first example described above, and descriptions of the same or corresponding content as in the first example described above may be omitted or simplified.

[0084] <Summary> Fig. 5 is a diagram illustrating a second example of the method for monitoring people in the home PF by the monitoring system 1. Specifically, Fig. 5 is a diagram illustrating transitions in the determination results of the level of safety for people M21 to M23 moving around the home PF.

[0085] As shown in FIG. 5, in this example, regions R1 to R4 for determining the safety of people are provided in the home PF.

[0086] In this example, unlike the first example described above, the region R3 is divided into regions R3 and R4.

[0087] The region R4 is closest to one end of the platform PF on the side of the railway vehicle TRN in the width direction.

[0088] The region R3 is located between the region R2 and the region R4 in the width direction of the home PF.

[0089] In this example, the boundary line B3 between regions R3 and R4, like the boundary line B1 between regions R1 and R2 and the boundary line B2 between regions R2 and R3, is set parallel to a straight line or curve that corresponds to one end of the platform PF on the side of the railway vehicle TRN. For example, when platform doors are installed on the platform PF, the boundary line B3 corresponds to the end face of the inside of the platform doors (the side opposite the railway vehicle TRN in the width direction of the platform PF) in a top view.

[0090] As in the first example described above, when a person is detected in area R1 by the tracking unit 20 based on the current (latest) image from the imaging unit 10, the judgment unit 30 judges that the person is highly safe (see the judgment results for person M21 and person M22).

[0091] On the other hand, when a person is detected in area R4 by the tracking unit 20 based on the current (latest) image from the imaging unit 10, the determination unit 30 determines that the safety of that person is low (see the determination result of M22).

[0092] Furthermore, when a person is detected in the area R3 by the tracking unit 20 based on the current (latest) image of the imaging unit 10, the determination unit 30 determines whether the person is safe or not depending on the time the person continues to stay in the area R3 (see the determination results of the persons M21 to M23). Specifically, when the duration of the person's stay in the area R3 is relatively short with respect to the predetermined time T1, the determination unit 30 determines that the person's safety is low (see the determination results of the persons M21 and M22). On the other hand, when the duration of the person's stay in the area R3 is relatively long with respect to T1, the determination unit 30 determines that the person's safety is high (see the determination result of the person M23). The duration of the person's stay being short and long relative to the predetermined time T1 means, for example, that the duration of the person's stay is shorter than the predetermined time T1 and that the duration of the person's stay is equal to or longer than the predetermined time T1. Furthermore, the term "a person's duration of stay being relatively short and long with respect to the predetermined time T1" may refer to the person's duration of stay being equal to or shorter than the predetermined time T1 and longer than the predetermined time T1. This allows the determination unit 30 to determine that even if a person is present in the area R3 relatively close to one end of the platform PF on the railcar TRN side, the safety of the person is high in a situation where the person is likely to stay in the area R3 for a certain period of time and approach the railcar TRN side (area R4). Therefore, the determination unit 30 can more appropriately determine the safety of the person present in the area R3.

[0093] The predetermined time T1 is set in the range of, for example, 3 to 5 seconds. 3 to 5 seconds corresponds to, for example, the time it takes for a person to walk between adjacent doors at the front and rear of a stopped railcar TRN. As a result, when a person stays in the area R3 for a time longer than or equal to the time it takes to walk between adjacent doors at the front and rear of the railcar TRN, the determination unit 30 can determine that the person has no intention of boarding the railcar TRN and is unlikely to approach the railcar TRN side (area R4). In addition, the predetermined time T1 may be fixed or may be variable according to some condition.

[0094] Even if the duration of a person's stay in area R3 becomes longer than the predetermined time T1 and the judgment unit 30 judges the person to be highly safe, the person may later move to area R2 and return to area R3. In this case, the judgment unit 30 judges the person to be low in safety according to the above rule (see the judgment result of person M23).

[0095] Furthermore, when a person is detected in the area R2 by the tracking unit 20 based on the current (latest) image of the imaging unit 10, the determination unit 30 determines whether the safety of the person is high or low based on the tracking result (movement route) of the person by the tracking unit 20. Specifically, when the person is moving from the area R1 to the area R2, the determination unit 30 maintains the determination result of the person's safety in the area R1 and determines that the person's safety is high (see the determination result of person M13). On the other hand, when the person is moving from the area R3 to the area R2, the determination unit 30 maintains the determination result of the person's safety in the area R3 (see the determination results of persons M22 and M23). More specifically, when the determination unit 30 has determined that the person's safety is low in the area R3, the determination unit 30 determines that the person's safety is low (see the determination result of person M22). On the other hand, when the determination unit 30 has determined that the person's safety is high in the area R3, the determination unit 30 determines that the person's safety is high (see the determination result of person M23). As a result, even if a person has moved from area R3 to area R2, if that person continues to stay in area R3 for a certain period of time and is unlikely to approach area R4, the determination unit 30 can determine that the person's safety continues to be low. Therefore, the determination unit 30 can more appropriately determine the safety of a person who has moved from area R3 to area R2.

[0096] In this way, in this example, the monitoring system 1 judges the safety of a person according to the duration of the person's stay in area R3. This allows the monitoring system 1 to more appropriately judge the safety of a person depending on the duration of the person's stay in area R3.

[0097] Note that, between the regions R3 and R4, there is no buffer region for the transition between high and low levels of human safety, like the region R2 between the regions R1 and R3. This is because it is unlikely that people will wait near the boundary line B3 between the regions R3 and R4, which is very close to one end of the platform PF on the side of the railway vehicle TRN, and frequent changes in the judgment results of the high and low levels of human safety due to people passing through the regions R3 and R4 are unlikely to occur. In this example, as in the first example described above, the region that combines the regions R3 and R4 may be set as the region R3. The same may be true for the third example described below.

[0098] <Processing flow> Fig. 6 is a diagram illustrating a second example of a sub-flowchart of the process related to monitoring of the home PF by the monitoring system 1. Specifically, Fig. 6 is a diagram illustrating a second example of a sub-flowchart of the process (individual determination process) of step S30 in Fig. 2.

[0099] 6, in step S202, the determination unit 30 determines whether the position (coordinates) of the target person detected by the tracking unit 20 corresponds to area R1, area R2, area R3, or area R4. If the position (coordinates) of the target person detected by the tracking unit 20 corresponds to area R1, the determination unit 30 proceeds to step S210, if it corresponds to area R2, the determination unit 30 proceeds to step S204, and if it corresponds to area R3 or area R4, the determination unit 30 proceeds to step S206.

[0100] In step S204, the judgment unit 30 judges whether or not the safety of the target person was determined to be high in the previous flowchart. If the judgment unit 30 judges that the safety of the person is high in the previous flowchart, the judgment unit 30 proceeds to step S210, otherwise proceeds to step S212.

[0101] On the other hand, in step S206, the determination unit 30 determines whether the position (coordinates) of the target person detected by the tracking unit 20 corresponds to area R3 or R4. If the position (coordinates) of the target person detected by the tracking unit 20 corresponds to area R3, the determination unit 30 proceeds to step S208, and if it corresponds to area R4, the determination unit 30 proceeds to step S212.

[0102] In step S208, the determination unit 30 determines whether or not the duration of the target person's stay in the area R3 is equal to or longer than a certain time (predetermined time T1) based on the tracking result of the tracking unit 20. If the duration of the target person's stay in the area R3 is equal to or longer than the predetermined time T1, the determination unit 30 proceeds to step S210, and otherwise proceeds to step S212.

[0103] Steps S210, S212, and S214 are the same as the processes in steps S106, S108, and S110 in FIG. 4, and therefore description thereof will be omitted.

[0104] When the process of step S214 is completed, the determination unit 30 ends the process of this flowchart.

[0105] Thus, in this example, when a target person is detected in area R3 by the tracking unit 20, the determination unit 30 determines whether the target person is safe or not depending on whether the person's continuous stay time in area R3 is equal to or longer than the predetermined time T1. As a result, when the target person stays in area R3 for a certain period of time, the determination unit 30 can determine that the target person is unlikely to approach area R4 and determine that the person is safe.

[0106] Furthermore, as described above, when the combined area of ​​areas R3 and R4 is set as area R3, step S206 is omitted, and when it is determined in step S202 that the position (coordinates) of the target person detected by the tracking unit 20 is in area R3, the process proceeds to step S208.

[0107] [Third example of how to monitor people in the nursing home] Next, a third example of the method for monitoring people in the home PF by the monitoring system 1 will be described with reference to Fig. 7 to Fig. 9. Specifically, a third example of the method for determining the safety of people in the home PF by the determination unit 30 will be described.

[0108] The following description will focus on the differences from the first and second examples described above, and descriptions of the same or corresponding content as the first and second examples described above may be omitted or simplified.

[0109] <Summary> Fig. 7 is a diagram illustrating a third example of the method for monitoring people in the home PF by the monitoring system 1. Specifically, Fig. 7 is a diagram illustrating transitions in the determination results of the level of safety for people M31 to M33 moving around the home PF.

[0110] As shown in FIG. 7, in this example, similar to the above-described second example, regions R1 to R4 for determining the safety of people are provided in the home PF.

[0111] As in the first and second examples described above, when a person is detected in area R1 by the tracking unit 20 based on the current (latest) image from the imaging unit 10, the judgment unit 30 judges that the person is highly safe (see the judgment results of person M31 and person M32).

[0112] On the other hand, as in the second example described above, if a person is detected in area R4 by the tracking unit 20 based on the current (latest) image from the imaging unit 10, the judgment unit 30 judges that the person's safety is low (see the judgment result of M32).

[0113] In addition, when a person is detected in area R3 by the tracking unit 20 based on the current (latest) image from the imaging unit 10, the judgment unit 30 judges the safety of that person depending on the amount of time that the person continues to stay in area R3 (see the judgment results of persons M31 to M33).

[0114] Specifically, the determination unit 30 determines that the safety of a person is high when a condition regarding the duration of stay of the person in the area R3 (hereinafter, "stay duration condition") is satisfied, and determines that the safety of the person is low when the condition is not satisfied. The stay duration condition in the area R2 is that the person enters the area R3 from the area R1 via the area R2 or from the area R4, and then continues to stay in the area that combines the areas R2 and R3, and the duration of the person's stay in the area R3 has become or is becoming relatively long relative to the predetermined time T1. As a result, when the person moves to the area R2 and returns to the area R3 after the duration of the person's stay in the area R3 becomes relatively long relative to the predetermined time T1, the determination unit 30 can maintain the determination that the safety of the person is high (see the determination result of person M33). Therefore, the determination unit 30 can determine that the safety of a person who has been staying in the areas R2 and R3 for a relatively long time and is unlikely to suddenly move toward the area R4 is high. Therefore, the determination unit 30 can more appropriately determine whether the person detected by the tracking unit 20 in the area R3 is safe or not.

[0115] In addition, the determination unit 30 may determine that the safety of the person is high when a condition (stay duration condition) regarding the person's stay duration in the area consisting of the combined areas R2 and R3 is satisfied, and may determine that the safety of the person is low when the condition is not satisfied. The stay duration condition in the area consisting of the combined areas R2 and R3 is that the person continues to stay in the area consisting of the combined areas R2 and R3 after entering the area R3 from the area R1 via the area R2 or from the area R4, and the person's stay duration in the area is relatively long compared to the predetermined time T1. As a result, the determination unit 30 can maintain the determination that the safety of the person is high when the stay duration condition in the area consisting of the combined areas R2 and R3 is satisfied during the process of moving from the area R3 to the area R2 and returning to the area R3 again (see the determination result of person M33). In addition, the determination unit 30 can determine that the safety of the person is high if the stay duration condition in the area consisting of the combined areas R2 and R3 is satisfied even if the stay duration condition in the area R3 is not satisfied. Therefore, for example, in a situation where a person is present near the boundary line B2 between areas R2 and R3 and the person's position is unstable between areas R2 and R3, even if the condition for the duration of stay in area R3 is not satisfied, the judgment unit 30 can determine that the person is highly safe.

[0116] Furthermore, when a person is detected in the area R2 by the tracking unit 20 based on the current (latest) image of the imaging unit 10, the determination unit 30 determines whether the safety of the person is high or low based on the tracking result (movement route) of the person by the tracking unit 20. Specifically, when the person is moving from the area R1 to the area R2, the determination unit 30 maintains the determination result of the person's safety in the area R1 and determines that the person's safety is high (see the determination result of person M31). On the other hand, when the person is moving from the area R3 to the area R2, the determination unit 30 maintains the determination result of the person's safety in the area R3 (see the determination results of people M32 and M33). Specifically, when the determination unit 30 has determined that the person's safety is high in the area R3, it maintains the determination that the person's safety is high (see the determination result of person M33), and when the determination unit 30 has determined that the person's safety is low in the area R3, it maintains the determination that the person's safety is low (see the determination result of person M32).

[0117] Furthermore, when a person is detected in area R2 by the tracking unit 20 based on the current (latest) image of the imaging unit 10, the determination unit 30 may determine whether the person is highly safe depending on whether the stay duration condition in the combined area of ​​areas R2 and R3 described above is met. That is, based on the tracking result of the person by the tracking unit 20, the determination unit 30 may determine that the person's safety in area R2 is high if the stay duration condition in the combined area of ​​areas R2 and R3 is met, and may determine that the person's safety in area R2 is low if the condition is not met.

[0118] In this way, in this example, the determination unit 30 determines the safety of a person detected in areas R2 and R3 by the tracking unit 20, taking into consideration the person's continued stay in the combined area of ​​areas R2 and R3. This allows the monitoring system 1 to more appropriately determine the safety of a person moving across areas R2 and R3.

[0119] <Processing flow> Fig. 8 is a diagram that shows a third example (part 1) of a sub-flowchart of the process related to monitoring people in the home PF by the monitoring system 1. Specifically, Fig. 8 is a diagram that shows a third example (part 1) of a sub-flowchart of the process (individual judgment process) of step S30 in Fig. 2. Fig. 9 is a diagram that shows a third example (part 2) of a sub-flowchart of the process related to monitoring people in the home by the monitoring system 1. Specifically, Fig. 9 is a diagram that shows a third example (part 2) of a sub-flowchart of the process (individual judgment process) of step S30 in Fig. 2.

[0120] In this example, the stay flag F1 of the region R3 is used. The stay flag F1 indicates whether the stay duration condition of the target person in the above-mentioned region R3 is met, and when it is "on", it indicates that the stay duration condition of the target person is met, and when it is "off", it indicates that the stay duration condition of the target person is not met. The initial state of the stay flag F1 is "off", and the determination unit 30 successively updates the stay flag F1 based on the tracking result (movement route history) of the tracking unit 20.

[0121] As shown in FIG. 8, steps S302, S304, and S306 are the same as the processes in steps S202, S204, and S206 in FIG. 6, and therefore a description thereof will be omitted.

[0122] Step S308 corresponds to step S208 in FIG.

[0123] In step S308, the determination unit 30 determines whether the current (latest) visit flag F1 of the target person is "on". If the current (latest) visit flag F1 is "on", the determination unit 30 proceeds to step S310, and if the visit flag F1 is "off", the determination unit 30 proceeds to step S312.

[0124] Steps S310, S312, and S314 are the same as steps S210, S212, and S214 in FIG. 6, and therefore description thereof will be omitted.

[0125] When the process of step S314 is completed, the determination unit 30 ends the process of this flowchart.

[0126] In this way, in this example, when a target person is detected in area R3 by the tracking unit 20, the determination unit 30 determines whether the target person is safe or not depending on whether the stay flag F1 of area R3 corresponding to the target person is on or off. This allows the determination unit 30 to determine whether the target person is safe or not depending on whether the stay duration condition of the target person in area R3 is met.

[0127] Also, instead of the flowchart in FIG. 8, the flowchart in FIG. 9 may be implemented.

[0128] In this example, the stay flag F2 of the combined area of ​​areas R2 and R3 is used. The stay flag F2 indicates whether the stay duration condition of the target person in the combined area of ​​areas R2 and R3 is met, and when it is "on", it indicates that the stay duration condition of the target person is met, and when it is "off", it indicates that the stay duration condition of the target person is not met. The initial state of the stay flag F2 is "off", and the determination unit 30 successively updates the stay flag F2 based on the tracking result (movement route history) of the tracking unit 20.

[0129] As shown in FIG. 9, step S402 is the same as the process of step S302 in FIG. 8, and therefore a description thereof will be omitted.

[0130] Step S404 corresponds to step S304 in FIG.

[0131] In step S404, the judgment unit 30 judges whether or not the safety of the target person was determined to be high in the previous flowchart. If the judgment unit 30 judges that the safety of the person is high in the previous flowchart, the judgment unit 30 proceeds to step S410, otherwise proceeds to step S408.

[0132] Step S406 is the same as the process in step S306 in FIG. 8, and therefore a description thereof will be omitted.

[0133] Step S408 corresponds to step S308 in FIG.

[0134] In step S408, the determination unit 30 determines whether the current (latest) visit flag F2 of the target person is "on". If the current (latest) visit flag F2 is "on", the determination unit 30 proceeds to step S410, and if it is "off", the determination unit 30 proceeds to step S412.

[0135] Steps S410, S412, and S414 are the same as steps S310, S312, and S314 in FIG. 8, and therefore description thereof will be omitted.

[0136] When the process of step S414 is completed, the determination unit 30 ends the process of this flowchart.

[0137] Thus, in this example, when the tracking unit 20 detects a target person in areas R2 and R3, the determination unit 30 determines whether the target person is safe or not based on whether the stay flag F2 of the area R3 corresponding to the target person is on or off. This allows the determination unit 30 to determine whether the target person is safe or not based on whether the stay duration condition of the target person in the combined area of ​​areas R2 and R3 is met.

[0138] As described above, when the combined area of ​​areas R3 and R4 is set as area R3, step S306 is omitted in Fig. 8, and if it is determined in step S302 that the position (coordinates) of the target person detected by the tracking unit 20 is area R3, the process proceeds to step S308. Similarly, in this case, step S406 is omitted in Fig. 9, and if it is determined in step S402 that the position (coordinates) of the target person detected by the tracking unit 20 is area R3, the process proceeds to step S408.

[0139] [Example 4 of how to monitor people in the nursing home] Next, a fourth example of the method for monitoring people in the home PF by the monitoring system 1 will be described with reference to Fig. 10. Specifically, a fourth example of the method for determining the safety of people in the home PF by the determination unit 30 will be described.

[0140] The following description will focus on the differences from the above-mentioned first to third examples, and description of the same or corresponding content as the above-mentioned first to third examples may be omitted or simplified.

[0141] <Summary> Fig. 10 is a diagram illustrating a fourth example of the method for monitoring people in the home PF by the monitoring system 1. Specifically, Fig. 10 is a diagram showing changes in the determination results of the level of safety for people M41 to M43 moving around the home PF.

[0142] As shown in FIG. 10, in this example, similar to the second and third examples described above, regions R1 to R4 for determining the safety of people are provided in the home PF.

[0143] As in the first to third examples described above, when a person is detected in area R1 by the tracking unit 20 based on the current (latest) image from the imaging unit 10, the judgment unit 30 judges that the person is highly safe (see the judgment results for person M41 and person M42).

[0144] On the other hand, as in the second and third examples described above, if a person is detected in area R4 by the tracking unit 20 based on the current (latest) image from the imaging unit 10, the judgment unit 30 judges that the person's safety is low (see the judgment result of M42).

[0145] Furthermore, as in the third example described above, when a person is detected in area R3 by the tracking unit 20 based on the current (latest) image from the imaging unit 10, the judgment unit 30 judges the safety of that person based on the amount of time that the person continues to stay in area R3 (see the judgment results for persons M41 to M43).

[0146] Specifically, if the condition for the duration of stay of the person in area R3 is satisfied, the determination unit 30 determines that the safety of the person is high, and if the condition is not satisfied, the determination unit 30 determines that the safety of the person is low.

[0147] In this example, if the person moves from area R3 to area R4 while the stay duration condition of the person in area R3 is satisfied, a grace period (predetermined time T2) is provided. That is, the stay duration condition of the person in area R3 is maintained in a state of being satisfied for a predetermined time T2 after the person moves to area R4. The predetermined time T2 may be fixed or may be variable according to some condition. Specifically, at the time point when the person moves from area R3 to area R4, the clocking of the stay duration in area R3 is paused at that time point, and the state is maintained for a predetermined time T2 starting from that time point. Then, if the person returns to area R3 during the predetermined time T2 starting from that time point, the clocking of the paused stay duration is resumed, and the state in which the stay duration in area R3 is relatively long compared to the predetermined time T1 is maintained. On the other hand, if the person does not return to area R3 during the predetermined time T2 starting from that time point, the stay duration is reset to zero. As a result, when a person detected by the tracking unit 20 moves to the area R4 in a state where the stay duration condition in the area R3 is satisfied, if the person immediately returns to the area R3, the determination unit 30 can immediately determine that the person's safety is high (see the determination result of the person M43). For example, a person who moves from the area R3 to the area R4 to board a railway car TRN entering the platform PF may realize that the train is not the intended train and immediately return to the area R3. In such a case, since the person is unlikely to move to the area R4 immediately, the determination unit 30 can determine that the safety of the person who has returned to the area R3 is high. Therefore, the determination unit 30 can more appropriately determine the level of safety of the person present in the area R3.

[0148] In addition, as in the third example described above, the judgment unit 30 may judge that a person's safety is high if the condition for the duration of stay in the area consisting of the combined areas R2 and R3 of that person is met, and may judge that the person's safety is low if the condition is not met.

[0149] In this example, if the person moves from area R3 to area R4 while the stay duration condition in the combined area R2 and R3 of the person is satisfied, a grace period (predetermined time T2) is provided, as in the case of the stay duration condition in area R3 described above. As a result, even if the person detected by the tracking unit 20 moves to area R4 while the stay duration condition in the combined area R2 and R3 is satisfied, if the person returns to area R3 immediately, the determination unit 30 can immediately determine that the person's safety is high (see the determination result of person M43). Therefore, the determination unit 30 can more appropriately determine the safety of the person present in area R3.

[0150] In this example, even if a person detected by the tracking unit 20 moves from area R3 to area R4 while the stay duration condition is satisfied, a grace period (predetermined time T2) is provided until the stay duration condition is no longer satisfied. Therefore, if a person who has been determined to be highly safe in area R3 moves from area R3 to area R4 and immediately returns to area R3, the monitoring system 1 can immediately determine that the person is highly safe. Therefore, the monitoring system 1 can more appropriately determine the safety of the person in area R3.

[0151] <Processing flow> In this example, the sub-flowchart of the processing (individual judgment processing) of step S30 in Figure 2 (fourth example of a sub-flowchart of processing related to monitoring of people in the home PF by the monitoring system 1) can be realized by a flowchart similar to the third example described above (Figures 8 and 9).

[0152] In this case, the stay flags F1 and F2 are successively updated by the judgment unit 30, taking into account the grace period (predetermined time T2) set for the stay duration condition when the target person moves from area R3 to area R4, as described above.

[0153] Thus, in this example, when a target person is detected in area R3 by the tracking unit 20, the determination unit 30 determines whether the person is safe or not based on the on / off state of the stay flags F1 and F2, taking into consideration the grace period (predetermined time T2) set for the stay duration condition. In this way, for example, in a situation where a person who moves to area R4 while the stay duration condition is satisfied immediately returns to area R3, the determination unit 30 can determine that the person is unlikely to head back to area R4, and immediately determine that the person is safe.

[0154] [5th ​​example of how to monitor people in the nursing home] Next, a fifth example of the method for monitoring people in the home PF by the monitoring system 1 will be described with reference to Fig. 11. Specifically, a fifth example of the method for determining the safety of people in the home PF by the determination unit 30 will be described.

[0155] The following description will focus on the differences from the above-mentioned first to fourth examples, and description of the same or corresponding content as the above-mentioned first to fourth examples may be omitted or simplified.

[0156] <Summary> Fig. 11 is a diagram illustrating a fifth example of the method for monitoring people in the home PF by the monitoring system 1. Specifically, Fig. 11 is a diagram showing transitions in the determination results of the level of safety for people M51 to M53 moving around the home PF.

[0157] As shown in FIG. 11, in this example, similar to the above-described second to fourth examples, regions R1 to R4 for determining the safety of people are provided in the home PF.

[0158] As in the first to fourth examples described above, when a person is detected in area R1 by the tracking unit 20 based on the current (latest) image from the imaging unit 10, the judgment unit 30 judges that the person is highly safe (see the judgment results for person M51 and person M52).

[0159] On the other hand, as in the second to fourth examples described above, if a person is detected in area R4 by the tracking unit 20 based on the current (latest) image from the imaging unit 10, the judgment unit 30 judges that the person's safety is low (see the judgment result of M52).

[0160] Furthermore, similar to the third and fourth examples described above, when a person is detected in area R3 by the tracking unit 20 based on the current (latest) image from the imaging unit 10, the judgment unit 30 judges the safety of that person based on the amount of time that the person continues to stay in area R3 (see the judgment results for persons M51 to M53).

[0161] Specifically, if the condition for the duration of stay of the person in area R3 is satisfied, the determination unit 30 determines that the safety of the person is high, and if the condition is not satisfied, the determination unit 30 determines that the safety of the person is low.

[0162] In this example, if the person moves from area R3 to area R4 while the stay duration condition in area R3 is satisfied, and then boards a railway vehicle TRN from area R4, a grace period (predetermined time T3) is provided. That is, the stay duration condition in area R3 is maintained as being satisfied for a predetermined time T3 after the person moves to area R4. The predetermined time T3 is set to be, for example, larger (longer) than the above-mentioned predetermined time T2. The predetermined time T3 may be fixed or may be variable according to some conditions. Specifically, at the time point when the person moves from area R3 to area R4, the timing of the stay duration in area R3 is temporarily stopped at that time point, and the state is maintained for the predetermined time T3 starting from that time point. Then, if the person gets on the railcar TRN from the area R4, gets off in the area R4, and returns to the area R3 during the predetermined time T3 starting from that time, the measurement of the suspended stay duration is resumed, and the stay duration in the area R3 is maintained relatively long relative to the predetermined time T1. On the other hand, if the person does not return to the area R3 after getting on the railcar TRN during the predetermined time T3 starting from that time, the stay duration is reset to zero. As a result, when the stay duration condition in the area R3 is satisfied, even if the person detected by the tracking unit 20 moves to the railcar TRN via the area R4, if the person immediately returns to the area R3 via the area R4, the determination unit 30 can immediately determine that the safety of the person is high (see the determination result of the person M43). For example, there may be a case where a person who gets on the railcar TRN entering the platform PF from the area R3 via the area R4 realizes that it is not the intended train, gets off in the area R4, and immediately returns to the area R3. In such a case, since the person is unlikely to move to area R4 immediately, the determination unit 30 can determine that the safety of the person who has returned to area R3 is high. Therefore, the determination unit 30 can more appropriately determine the safety of the person in area R3.

[0163] In addition, as in the third example described above, the judgment unit 30 may judge that a person's safety is high if the condition for the duration of stay in the area consisting of the combined areas R2 and R3 of that person is met, and may judge that the person's safety is low if the condition is not met.

[0164] In this example, as in the case of the stay duration condition in the region R3 described above, if the person moves from region R3 to region R4 while the stay duration condition in the region R2 and R3 is satisfied, and then boards the railroad car TRN from region R4, a grace period (predetermined time T3) is provided. As a result, even if the person detected by the tracking unit 20 moves to the railroad car TRN via region R4 while the stay duration condition in the region R2 and R3 is satisfied, if the person immediately returns to region R3 via region R4, the determination unit 30 can immediately determine that the person's safety is high (see the determination result of person M43). Therefore, the determination unit 30 can more appropriately determine the level of safety of the person present in region R3.

[0165] In this example, even if a person detected by the tracking unit 20 moves from area R3 to area R4 while the stay duration condition is satisfied, a grace period (predetermined time T3) is provided until the stay duration condition is no longer satisfied. Therefore, if a person who has been determined to be highly safe in area R3 boards a railway vehicle TRN from area R3 via area R4, immediately disembarks in area R4, and returns to area R3, the monitoring system 1 can immediately determine that the person is highly safe. Therefore, the monitoring system 1 can more appropriately determine the safety of people in area R3.

[0166] The determination unit 30 may use both the grace period (predetermined time T2) of the fourth example described above and the grace period (predetermined time T3) of this example. In this case, the determination unit 30 may determine whether the user has returned to the area R3 without boarding the railroad vehicle TRN or has boarded the railroad vehicle TRN and returned to the area R3 based on the tracking result of the tracking unit 20, and may determine whether to use the predetermined time T2 or the predetermined time T3. In this example, the area R3 may be set to be the area that combines the areas R3 and R4, as in the first example described above. In this case, when the user boards the railroad vehicle TRN from the area R3, a grace period (predetermined time T3) is set until the stay duration condition is not satisfied.

[0167] <Processing flow> In this example, the sub-flowchart of the processing (individual judgment processing) of step S30 in Figure 2 (a fifth example of a sub-flowchart of processing related to monitoring of people in the home PF by the monitoring system 1) can be realized by a flowchart similar to the third example described above (Figures 8 and 9).

[0168] In this case, the stay flags F1 and F2 are successively updated by the judgment unit 30, taking into account the grace period (predetermined time T3) set for the stay duration condition when the target person moves from area R3 to area R4, as described above.

[0169] In this example, when a target person is detected in area R3 by the tracking unit 20, the determination unit 30 determines whether the person is safe or not based on the on / off state of the stay flags F1 and F2, taking into consideration the grace period (predetermined time T3) set for the stay duration condition. In this way, for example, in a situation where a person who boards a railway car TRN via area R4 while the stay duration condition is satisfied gets off in area R4 and immediately returns to area R3, the determination unit 30 can determine that the person is unlikely to head to area R4 again and immediately determine that the person is safe.

[0170] [Example 6 of how to monitor people in nursing homes] Next, a sixth example of the method for monitoring people in the home PF by the monitoring system 1 will be described with reference to Fig. 12 to Fig. 14. Specifically, the sixth example of the method for determining the safety of people in the home PF by the determination unit 30 will be described.

[0171] The following description will focus on the differences from the above-mentioned first to fifth examples, and description of the same or corresponding content as the above-mentioned first to fifth examples may be omitted or simplified.

[0172] <Summary> Fig. 12 is a diagram illustrating a sixth example of the method for monitoring people in a home PF by the monitoring system 1. Specifically, Fig. 12 is a diagram illustrating the transition of the judgment result of the level of safety for people M61-M63 moving through the home PF when the congestion level of the home PF is relatively high with respect to a predetermined standard.

[0173] In this example, the determination unit 30 switches the method of determining whether the safety of a person detected by the tracking unit 20 is high or low depending on the level of congestion in the home PF.

[0174] The determination unit 30 estimates the degree of congestion at the platform PF based on, for example, the output (captured image) of the imaging unit 10. Specifically, the determination unit 30 may estimate the degree of congestion at the platform PF based on the number of people detected by the tracking unit 20. Furthermore, instead of or in addition to the output of the imaging unit 10, the determination unit 30 may estimate the degree of congestion at the platform PF based on other information related to the number of people at the platform PF. The other information related to the number of people at the platform PF is, for example, information related to the number of people who passed through a ticket gate at a station where the target platform PF is located per unit time, information related to the occupancy rate of a railway vehicle TRN that stopped at the platform PF within a recent specified time, etc.

[0175] When the degree of congestion in the home PF is not relatively high with respect to a predetermined standard (hereinafter, "first standard"), the determination unit 30 determines the level of safety of the person based on the position of the person detected by the tracking unit 20 and the movement history in the home PF. For example, the determination unit 30 determines the level of safety of the person detected by the tracking unit 20 by any one of the determination methods of the first to fourth examples described above.

[0176] On the other hand, when the congestion level at the platform PF is relatively high compared to a predetermined standard (first standard), the determination unit 30 determines whether the person is safe or not based only on the position of the person detected by the tracking unit 20 and the position of the person in the movement history at the platform PF. Specifically, when the person is present within a predetermined area including one end of the platform PF where the railcar TRN enters, the determination unit 30 determines that the person's safety is low, and when the person is present outside the predetermined area, the determination unit 30 determines that the person's safety is high.

[0177] The degree of congestion of a home PF being relatively high with respect to a predetermined standard may mean that the degree of congestion is equal to or greater than the predetermined standard, or may mean that the degree of congestion exceeds the predetermined standard.

[0178] For example, as shown in FIG. 12, unlike the second to fourth examples described above, the home PF is provided with an area R5 in addition to the areas R1 to R4.

[0179] In this example, unlike the second to fourth examples described above, the region R4 is divided into regions R4 and R5.

[0180] Area R5 is closest to one end of the platform PF on the side of the railway car TRN in the width direction.

[0181] The region R4 is located between the region R3 and the region R5 in the width direction of the home PF.

[0182] In this example, the boundary line B4 of the regions R4 and R5, like the boundary line B1 of the regions R1 and R2, the boundary line B2 of the regions R2 and R3, and the boundary line B3 of the regions R3 and R4, is set parallel to a straight line or curve corresponding to one end of the platform PF on the side of the railway vehicle TRNS. For example, when platform doors are installed on the platform PF, the boundary line B4 corresponds to the end face of the outer side of the platform doors (the side of the railway vehicle TRN in the width direction of the platform PF) in a top view.

[0183] When a person detected by the tracking unit 20 is outside area R5, i.e., when the person is in areas R1 to R4, the determination unit 30 determines that the person's safety is high. On the other hand, when a person detected by the tracking unit 20 is in area R5, the determination unit 30 determines that the person's safety is low. This allows the determination unit 30 to determine whether a person detected by the tracking unit 20 is safe or not, depending on whether the person's position is within area R5 or not, regardless of the movement history (movement route) of the person on the home PF (see the determination results of persons M61 to M63).

[0184] For example, when the degree of congestion in the home PF becomes relatively high, the space in which a person can move is restricted, and the person may not be able to move as he or she wishes. In addition, when the degree of congestion in the home PF becomes relatively high, tracking of the person may become difficult due to overlapping of people on the image of the imaging unit 10. As a result, the determination of the level of safety of a person based on the movement history of the person in the home PF may not function properly. In contrast, in this example, when the degree of congestion in the home PF is relatively high compared to the first criterion, the determination unit 30 determines the level of safety of the person based only on the position of the person on the home PF and the position of the person in the movement history. This allows the determination unit 30 to more appropriately determine the safety of the person on the home PF in accordance with the situation in which the degree of congestion in the home PF is high.

[0185] Also, for example, when the degree of congestion of the platform PF becomes relatively high, many people moving or waiting are present in an area relatively close to one end of the platform PF on the side of the railway vehicle TRN in the width direction. Therefore, if it is determined that the safety of these people is low, the state in which the safety of the platform PF is determined to be low continues, and it is possible that the doors of the railway vehicle TRN and the platform doors cannot be closed. As a result, it is possible that the operation of the railway vehicle TRN is hindered. In contrast, in this example, when the degree of congestion of the platform PF is relatively high compared to the first standard, the determination unit 30 limits the area in which the safety of people is determined to be low to an area R5 at one end of the platform PF on the side of the railway vehicle TRN in the width direction. This allows the determination unit 30 to determine that only people who actually board the railway vehicle TRN or people who actually disembark from the railway vehicle TRN at one end of the platform PF in the width direction are in a low safety state. Therefore, the determination unit 30 can determine that the safety of people who do not intend to approach and board the railway vehicle TRN is high, even if they are present in an area relatively close to the railway vehicle TRN, such as the area R4. Therefore, determination unit 30 can suppress delays in the departure of railway vehicle TRN while ensuring the safety of the platform PF.

[0186] In this example, when the congestion level of the home PF is relatively high with respect to the first criterion, the determination unit 30 may further switch the method of determining whether the safety of the person detected by the tracking unit 20 is high or low, depending on the congestion level of the home PF. That is, the determination unit 30 may switch the method of determining whether the safety of the person detected by the tracking unit 20 is high or low, between three or more stages, depending on the congestion level of the home PF. The same may be applied to the seventh example described below.

[0187] For example, when the congestion level of the platform PF is not relatively high with respect to the second standard, the determination unit 30 determines whether the safety of the person detected by the tracking unit 20 is high or low by the monitoring method of the first example described above. The second standard is set lower than the first standard. On the other hand, when the congestion level of the platform PF is not relatively high with respect to the first standard and is relatively high with respect to the second standard, the determination unit 30 determines whether the safety of the person detected by the tracking unit 20 is high or low by the monitoring method of any one of the second to fourth examples described above. This is because the higher the congestion level of the platform PF, the higher the possibility that the person will move or stay in an area closer to the railroad vehicle PF in the width direction of the platform PF (for example, area R3 in the second to fourth examples described above).

[0188] Thus, in this example, when the degree of congestion in the home PF becomes relatively high with respect to the first criterion, the determination unit 30 switches the determination method to determine the level of safety of a person based only on the position and movement history of the person detected by the tracking unit 20. This allows the determination unit 30 to more appropriately determine the level of safety of a person in the home PF in accordance with a situation in which the degree of congestion in the home PF is extremely high.

[0189] <Processing flow> 13 and 14 are diagrams each showing a schematic example of a sub-flowchart of processing related to a home PF by the monitoring system 1. Specifically, Fig. 13 is a diagram showing a schematic example of a sub-flowchart of the processing (individual determination processing) of step S30 in Fig. 2, and Fig. 14 is a sub-flowchart showing an example of the processing of step S508 in Fig. 13.

[0190] As shown in FIG. 13, in step S502, the determination unit 30 estimates the congestion degree of the home PF.

[0191] When the process of step S502 is completed, the determination unit 30 proceeds to step S504.

[0192] In step S504, the determination unit 30 determines whether the congestion level of the home PF is relatively high with respect to a predetermined standard. If the congestion level of the home PF is not relatively high with respect to the predetermined standard, the determination unit 30 proceeds to step S506, and if the congestion level is relatively high, the determination unit 30 proceeds to step S508.

[0193] In step S506, the determination unit 30 performs an individual determination process taking into consideration the movement history (movement route) on the home PF of the person detected by the tracking unit 20. For example, the determination unit 30 executes the process of any one of the sub-flowcharts of Fig. 4, Fig. 6, Fig. 8, and Fig. 9 described above as the individual determination process taking into consideration the movement history (movement route) on the home PF of the person detected by the tracking unit 20.

[0194] On the other hand, in step S508, the determination unit 30 performs individual determination processing without considering the movement history (movement route) on the home PF of the person detected by the tracking unit 20. Specifically, the processing of the sub-flowchart in FIG.

[0195] 14, in step S602, the determination unit 30 sets an area for determining whether the person is safe or not depending on the person's position. Specifically, the determination unit 30 sets an area with low safety (for example, area R5 in FIG. 12) at one end of the platform PF on the side of the railway vehicle TRN in the width direction.

[0196] When the process of step S602 is completed, the decision unit 30 proceeds to step S604.

[0197] In step S604, the determination unit 30 determines whether or not the position (coordinates) of the target person detected by the tracking unit 20 is within the set area of ​​step S602. If the position of the target person is outside the set area, the determination unit 30 proceeds to step S606, and if it is within the set area, the determination unit 30 proceeds to step S608.

[0198] Steps S606, S608, and S610 are the same as steps S106, S108, and S110 in FIG. 4, and therefore description thereof will be omitted.

[0199] When the process of step S610 is completed, the determination unit 30 ends the process of this sub-flowchart.

[0200] Returning to FIG. 13, when the process of step S506 or step S508 is completed, the determination unit 30 ends the process of this sub-flowchart.

[0201] In this way, in this example, when the congestion level of the home PF is not relatively high, the determination unit 30 determines whether the person is safe based on the person's position and movement history, and when the congestion level of the home PF is relatively high, the determination unit 30 can determine whether the person is safe based only on the person's position. Specifically, when the congestion level of the home PF is relatively high, the determination unit 30 can determine whether the person is safe based on whether the position of the person detected by the tracking unit 20 is within a set area.

[0202] [Other examples of surveillance systems] Next, another example of the monitoring system 1 will be described with reference to FIG.

[0203] In this example, the same symbols are used for configurations that are the same as or correspond to the above-mentioned example (Figure 1), and the description will focus on the parts that are different from the above-mentioned example, and descriptions of the same or corresponding content as the above-mentioned example may be omitted.

[0204] FIG. 15 is a diagram showing another example of the monitoring system 1. As shown in FIG.

[0205] As shown in FIG. 15, in this example, the surveillance system 1 differs from the above-mentioned example (FIG. 1) in that it includes a sensor 50 and a detection unit 60 in addition to an imaging unit 10, a tracking unit 20, a judgment unit 30, and a display unit 40, but may be the same as the above-mentioned example in other respects.

[0206] In this example, the platform screen door PSD is installed at one end of the platform PF on the side of the railway vehicle TRN in the width direction.

[0207] The sensor 50 is provided at the platform door PF, and acquires and outputs information regarding obstacles (people) present at the opening of the platform door PSD and obstacles (people) in the space between the platform door PF and the railway vehicle TRN. For example, the sensor 50 is provided at each opening of the platform door PSD.

[0208] The sensor 50 is communicatively connected to the detection unit 60 via a one-to-one communication line, a local network, or the like, and the output of the sensor 50 (hereinafter, “sensor information”) is input to the detection unit 60.

[0209] The monitoring system 1 may include a camera attached to the railway vehicle TRN instead of or in addition to the sensor 50. In this case, the camera is mounted on the railway vehicle TRN so as to be able to capture an image of the opening of the platform door PSD and the space between the platform door PSD and the railway vehicle TRN. For example, a camera is provided for each door of the railway vehicle TRN and is attached to the outer surface of the railway vehicle TRN. The camera may also be an interior monitoring camera provided for each door of the railway vehicle TRN. In this example, the platform door PSD may be omitted. In this case, the monitoring system 1 includes a camera attached to the railway vehicle TRN instead of the sensor 50. In this case, the camera is mounted on the railway vehicle TRN so as to be able to capture an image of the vicinity of one end of the platform PF that is very close to the railway vehicle TRN in the width direction.

[0210] The detection unit 60 detects a person present at the opening of the platform door PSD or in the space between the platform door PSD and the railway vehicle TRN based on the output of the sensor 50. The detection unit 60 may also detect a person present at the opening of the platform door PSD or in the space between the platform door PSD and the railway vehicle TRN based on the output of a camera mounted on the railway vehicle TRN entering the platform PF.

[0211] The function of the detection unit 60 may be realized by any hardware or a combination of any hardware and software. For example, the detection unit 60 is mainly configured with a computer (information processing device) including a CPU, a memory device, an auxiliary storage device, and an interface device. The memory device is, for example, an SRAM or a DRAM. The auxiliary storage device is, for example, an HDD, an SSD, a flash memory, an EEPROM, and the like. The interface device includes, for example, a communication interface for communicating with other components of the monitoring system 1 such as the determination unit 30 and the sensor 50. The interface device also includes, for example, an external interface for connecting to a recording medium. This allows the detection unit 60 to, for example, retrieve various programs from a recording medium through the external interface and install them in the auxiliary storage device.

[0212] The detection unit 60 is communicably connected to the determination unit 30 via a one-to-one communication line, a local network, or the like, and the output of the detection unit 60 is taken into the determination unit 30.

[0213] The functions of the tracking unit 20, the determination unit 30, and the detection unit 40 may be integrated and realized in one or two information processing devices. Also, the functions of the tracking unit 20, the determination unit 30, and the detection unit 40 may be distributed and realized by four or more information processing devices.

[0214] The determination unit 30 determines the level of safety of people in the home PF based on the output of the tracking unit 20 and the output of the detection unit 60.

[0215] [Example 7 of how to monitor people in nursing homes] Next, a seventh example of the method for monitoring people in the home PF by the monitoring system 1 will be described with reference to Fig. 16. Specifically, a seventh example of the method for determining the safety of people in the home PF by the determination unit 30 will be described.

[0216] The following description will focus on the differences from the above-mentioned first to sixth examples, and may omit or simplify the description of the same or corresponding content as the above-mentioned first to sixth examples. In addition, the description of this example will be given on the premise of the monitoring system 1 in FIG.

[0217] <Summary> In this example, similar to the sixth example described above, the determination unit 30 switches the method of determining whether the safety of the person detected by the tracking unit 20 is high or low depending on the level of congestion in the home PF.

[0218] When the congestion level of the home PF is not relatively high with respect to a predetermined standard (first standard), the determination unit 30 determines whether the person is safe or not based on the position of the person detected by the tracking unit 20 and the movement history of the person in the home PF, similar to the sixth example described above. For example, the determination unit 30 determines whether the person detected by the tracking unit 20 is safe or not by any one of the determination methods of the first to fourth examples described above.

[0219] On the other hand, when the congestion level of the platform PF is relatively high compared to a predetermined standard (first standard), the determination unit 30 determines whether the safety of the person detected by the tracking unit 20 is high or low, based on the output of the sensor 50 or the camera mounted on the railway vehicle TRN. For example, the determination unit 30 determines that the safety of the person detected by the tracking unit 20 who is also detected by the detection unit 60 is low, and determines that the safety of the person not detected by the detection unit 60 is high.

[0220] As a result, when the degree of congestion in the home PF is relatively high with respect to the first criterion, the determination unit 30 can determine the safety of a person based only on the person's position on the home PF and the person's position in the movement history. Specifically, the determination unit 30 can determine the level of safety of a person based on whether or not the person detected by the tracking unit 20 is located within a range detectable by the detection unit 60 based on the sensor 50 or the like. Therefore, the determination unit 30 can more appropriately determine the safety of a person on the home PF in accordance with a situation in which the degree of congestion in the home PF is high.

[0221] In addition, in this example, when the degree of congestion of the platform PF is relatively high with respect to the first standard, the determination unit 30 can limit the subjects for which the safety is determined to be low to only those people located within a range detectable by the detection unit 60 based on the sensor 50 or the like. This allows the determination unit 30 to determine that only people who actually board the railway vehicle TRN or who actually disembark from the railway vehicle TRN at one end of the width direction of the platform PF are in a low safety state. Therefore, the determination unit 30 can determine that the safety of a person who does not intend to approach and board the railway vehicle TRN is high, even if the person is located in an area relatively close to the railway vehicle TRN, such as the area R4 in FIG. 12 described above. Therefore, the determination unit 30 can suppress delays in the departure of the railway vehicle TRN while ensuring the safety of the platform PF.

[0222] Thus, in this example, similar to the sixth example described above, when the degree of congestion in the home PF becomes relatively high with respect to the first criterion, the determination unit 30 switches the determination method to determine the level of safety of a person based only on the position and movement history of the person detected by the tracking unit 20. This allows the determination unit 30 to more appropriately determine the level of safety of a person in the home PF in accordance with a situation in which the degree of congestion in the home PF is extremely high.

[0223] <Processing flow> In this example, the process of switching the individual determination process depending on the congestion level of the home PF is the same as the sub-flowchart of the sixth example in Fig. 13. Therefore, in this example, Fig. 13 will be used and the description will focus on the parts that are different from the sixth example.

[0224] 15 is a diagram illustrating roughly a seventh example of the sub-flowchart of the processing related to the home PF by the monitoring system 1. Specifically, it is a sub-flowchart showing another example of the processing in step S508 in FIG.

[0225] At the start of the sub-flowchart of FIG. 15, a human detection flag, which will be described later, is initialized to "OFF."

[0226] As shown in FIG. 15, the determination unit 30 acquires the latest output of the detection unit 60 in step S702.

[0227] When the process of step S702 is completed, the process proceeds to step S704.

[0228] In step S704, the determination unit 30 determines whether or not a person has been detected based on the detection of a person by the detection unit 60. If a person has been detected by the detection unit 60, the determination unit 30 proceeds to step S706, and if a person has not been detected by the detection unit 60, the determination unit 30 proceeds to step S712 for all people detected by the tracking unit 20.

[0229] In step S706, the determination unit 30 sets a human detection flag, which indicates that a human has been detected by the detection unit 60, to "ON."

[0230] As a result, even if the detection unit 60 detects only people who are not captured in the image captured by the imaging unit 10, in the overall judgment process of step S40 in Figure 2, the judgment unit 30 can judge that the overall safety is low based on the setting state of the human detection flag to be "ON".

[0231] When the process of step S706 is completed, the decision unit 30 proceeds to step S708.

[0232] In step S708, the determination unit 30 extracts a person to be detected by the detection unit 60 from among the people detected by the tracking unit 20. For example, the determination unit 30 extracts the person to be detected by performing a matching process for each person detected by the tracking unit 20 between the position of that person and the position of the person detected by the detection unit 60 or the installation position of the corresponding sensor 50 or the like.

[0233] When the process of step S708 is completed, the decision unit 30 proceeds to step S710.

[0234] The process of step S710 is performed for each person detected by the tracking unit 20.

[0235] In step S710, the determination unit 30 determines whether or not the target person detected by the tracking unit 20 is a person extracted in step S708 to be detected by the detection unit 30. If the target person is not a person to be detected by the detection unit 30, the determination unit 30 proceeds to step S712, and if the target person is a subject, the determination unit 30 proceeds to step S714.

[0236] Steps S712, S714, and S716 are the same as steps S106, S108, and S110 in FIG. 4, and therefore description thereof will be omitted.

[0237] When the process of step S716 is completed, the determination unit 30 ends the process of this sub-flowchart.

[0238] In this way, in this example, when the congestion level of the home PF is not relatively high, the determination unit 30 determines whether the person is safe based on the person's position and movement history, and when the congestion level of the home PF is relatively high, the determination unit 30 can determine whether the person is safe based only on the person's position. Specifically, when the congestion level of the home PF is relatively high, the determination unit 30 can determine whether the person is safe based on whether the person detected by the tracking unit 20 is a person detected by the detection unit 60 based on the output of the sensor 50, etc.

[0239] [Example 8 of how to monitor people in nursing homes] Next, an eighth example of the method for monitoring people in the home PF by the monitoring system 1 will be described with reference to Fig. 17. Specifically, an eighth example of the method for determining the safety of people in the home PF by the determination unit 30 will be described.

[0240] The following description will focus on differences from the above-mentioned first to seventh examples, and may omit or simplify descriptions of contents that are the same as or correspond to the above-mentioned first to seventh examples. Also, in this example, similar to the above-mentioned seventh example, the description will be given on the premise of the monitoring system 1 in FIG. 15.

[0241] <Summary> In this example, the determination unit 30 determines the safety of people in the home PF based on the output (captured image) of the imaging unit 10, and determines the safety of people in the home PF based on the output of the sensor 50.

[0242] For example, the determination unit 30 determines whether the safety of the person detected by the tracking unit 20 is high or low based on the output of the imaging unit 10 by using any one of the monitoring methods of the first to fifth examples described above.

[0243] Additionally, when a person is detected by the detection unit 60 based on the output of the sensor 50 or the like, the determination unit 30 determines that the safety of the person detected by the detection unit 60 is low.

[0244] For example, in the case of short people such as children or people using wheelchairs, there is a possibility that they are hidden by the platform door PSD or people in the vicinity and are not captured in the captured image of the imaging unit 10. Therefore, the tracking unit 20 cannot detect these people who are relatively close to the railway vehicle TRN, and as a result, the determination unit 30 may determine that the overall safety of people on the platform PF is high without grasping the presence of these people. In contrast, in this example, the detection unit 60 can detect short people such as children or people using wheelchairs who are present at the opening of the platform door PSD or between the platform door PSD and the railway vehicle TRN based on the output of the sensor 50, etc. Then, the determination unit 30 can determine that the safety of the person detected by the detection unit 60 is low. Therefore, the determination unit 30 can determine that the overall safety of people on the platform PF is low based on the presence of short people such as children or people using wheelchairs who are present at the opening of the platform door PSD or between the platform door PSD and the railway vehicle TRN.

[0245] In addition, the determination unit 30 may perform matching between the person detected by the tracking unit 20 and the person detected by the detection unit 50.

[0246] In this manner, in this example, the determination unit 30 can determine the safety of people in the home PF based on the output of the imaging unit 10, that is, the image of the home PF, as well as the output of the sensor 50 and the like.

[0247] <Processing flow> Fig. 17 is a diagram illustrating an eighth example of a sub-flowchart of the processing related to the home PF by the monitoring system 1. Specifically, Fig. 17 is a diagram illustrating an eighth example of a sub-flowchart of the processing (individual determination processing) of step S30 in Fig. 2.

[0248] 17, in step S802, the determination unit 30 performs individual determination processing based on the captured image of the imaging unit 10. Specifically, the determination unit 30 determines the level of safety of each person detected by the tracking unit 20 based on the output (captured image) of the imaging unit 10. For example, the determination unit 30 executes processing of any one of the sub-flowcharts of FIG. 4, FIG. 6, FIG. 8, and FIG. 9 as the individual determination processing based on the captured image of the imaging unit 10.

[0249] When the process of step S802 is completed, the process proceeds to step S804.

[0250] In step S804, the determination unit 30 performs individual determination processing based on the output of the sensor 50, etc. For example, the determination unit 30 executes the processing of FIG.

[0251] When the process of step S804 is completed, the determination unit 30 ends the process of this sub-flowchart.

[0252] In this manner, the determination unit 30 can determine the safety of the home PF based on the output of the imaging unit 10, and can also determine the safety of people in the home PF based on the output of the sensor 50 and the like.

[0253] [Examples of what is displayed on the display] Next, specific examples of the display contents of the display unit 40 will be described with reference to FIG. 18 and FIG.

[0254] <Example 1> Fig. 18 is a diagram showing a first example of a screen displayed on display unit 40. Specifically, Fig. 18 is a diagram showing an example of a screen on display unit 40 when any one of the first to fifth examples of the above-mentioned home PF occupant monitoring method is applied, or when any one of the sixth or seventh example of the above-mentioned home PF occupant monitoring method is applied, and the congestion degree of the home PF is not relatively high with respect to a first criterion.

[0255] As shown in FIG. 18, an image 180 captured by the imaging section 10 is displayed on the screen of the display section 40.

[0256] The image 180 shows a railcar TRN, a platform PF, and people M11 and M12 on the platform PF.

[0257] Moreover, on the screen of the display unit 40, images 181 to 186 are displayed superimposed on the image 180.

[0258] Image 181 represents a boundary line relating to the level of safety of people on platform PF. Image 181 includes images 181a and 181b. Specifically, image 181a represents the above-mentioned boundary line B1, and image 181b represents the above-mentioned boundary line B2. This allows station staff at station ST and crew members of railway car TRN to visually distinguish and confirm regions R1 to R3 for determining the safety of people M11 and M12.

[0259] Image 182 is a rectangular frame indicating that person M12 has been detected by the tracking unit 20.

[0260] In the width direction of the platform PF, the person M11 steps toward the door DR of the railcar TRN into an area R3 closer to the railcar TRN than the image 181b corresponding to the boundary line B2. Therefore, the person M11 is surrounded by a solid-line rectangular frame, which indicates that the person M11 is less safe.

[0261] The image 183 is a rectangular frame indicating that the person M2 has been detected by the tracking unit 20.

[0262] Person M12 is standing in area R1 on the opposite side of image 181a, which corresponds to boundary line B1, from railway vehicle TRN in the width direction of the platform PF, and is therefore surrounded by a dotted frame, indicating that person M12 is highly safe.

[0263] This allows station staff at station ST and train crew on the railway vehicle TRN to visually confirm the safety assessment results for each of people M11 and M12.

[0264] The rectangular frames of the images 182 and 183 may be distinguished by color, thickness, or the like, instead of line type.

[0265] Image 184 is drawn on top of image 182 (rectangular frame) and is a unique identification number (ID: Identification) representing person M1 corresponding to image 182.

[0266] Image 185 is drawn on top of image 183 (rectangular frame) and is a unique identification number (ID) representing person M2 corresponding to image 183.

[0267] Basically, the same identification numbers are assigned to the persons M11 and M12 for the images 184 and 185 as long as the persons M11 and M12 continue to be detected and tracked by the tracking unit 20. This allows station staff at the station ST and crew members of the railway vehicle TRN to easily identify the persons M11 and M12 from among the multiple persons in the image 120 by the identification numbers of the images 184 and 185 and confirm their movements.

[0268] An image 186 shows the overall safety assessment result for all people on the home PF detected from the image 180, that is, the result of the above-mentioned overall assessment process.

[0269] In this example, the determination unit 30 determines that the safety of the person M11 is low, and as a result of the determination, the overall safety of the person on the platform PF corresponding to the image 180 is low. Therefore, text information "Decision: Danger" is drawn on the image 186, which indicates that the overall safety of the person on the platform PF corresponding to the image 180 is determined to be low by the determination unit 30. This allows station staff at the station ST and crew members of the railway vehicle TRN to easily grasp the overall safety of the person on the platform PF corresponding to the image 180 by checking the image 186. Therefore, for example, in a situation where the railway vehicle TRN has entered the platform PF and is stopped, the station staff at the station ST and crew members of the railway vehicle TRN can determine whether or not it is acceptable to close the open door DR by checking the image 186.

[0270] It should be noted that the image 186 may be omitted. For example, as described above, when the comprehensive judgment process (step S40 in FIG. 2) is omitted, the image 186 is omitted.

[0271] Thus, in this example, on the display unit 40, images 181-186 corresponding to information on the person detection result by the tracking unit 20 and the judgment result of the person's safety by the judgment unit 30 are displayed superimposed on an image 180 showing the platform PF and the railway vehicle TRN. This allows station staff at the station ST and crew members of the railway vehicle TRN to easily grasp the person detection situation by the tracking unit 20 and the judgment result of the person's safety by the judgment unit 30, in addition to the situation of the person on the platform PF.

[0272] <Second Example> Fig. 19 is a diagram showing a second example of a screen displayed on the display unit 40. Specifically, Fig. 19 is a diagram showing an example of a screen on the display unit 40 when the sixth or seventh example of the method for monitoring people in the home PF described above is applied and the degree of congestion in the home PF is relatively high with respect to the first criterion.

[0273] As shown in FIG. 19, an image 190 captured by the imaging section 10 is displayed on the screen of the display section 40.

[0274] The image 190 shows a railcar TRN, a platform PF, and people M21 to M2 on the platform PF.

[0275] Moreover, on the screen of the display unit 40, images 191 to 197 are displayed superimposed on the image 190.

[0276] Image 191 represents a boundary line relating to the high and low safety of people on platform PF. Image 191 includes images 191a to 191d. Specifically, images 191a to 191d represent the above-mentioned boundary lines B1 to B4, respectively. This allows station staff at station ST and crew members of railway car TRN to visually distinguish and confirm regions R1 to R5 for determining the safety of people M21 to M25.

[0277] Images 192 to 196 are rectangular frames indicating that people M21 to M25 have been detected by the tracking unit 20, respectively.

[0278] When the degree of congestion at the platform PF is relatively high with respect to the first criterion, the person M21 is present in an area R5 on the railway vehicle TRN side of the image 191d corresponding to the boundary line B4. Therefore, the determination unit 30 can determine that the safety of the person M21 is low.

[0279] Furthermore, when the degree of congestion at the platform PF is relatively high with respect to the first criterion, the person M21 is present at the opening of the platform door PSD. Therefore, the person M21 is detected by the detection unit 50 based on the output of the sensor 50, etc., and the determination unit 30 can determine that the safety of the person M21 is low.

[0280] Therefore, person M21 is surrounded by a solid-line rectangular frame, which indicates that his / her safety is low.

[0281] On the other hand, when the congestion degree of the home PF is relatively high with respect to the first criterion, the people M22 to M25 are present in any of the areas R1 to R4, so the determination unit 30 can determine that the safety of the people M22 to M25 is high.

[0282] Furthermore, when the degree of congestion on the platform PF is relatively high with respect to the first criterion, people M22-M25 are present inside the platform PF relative to the platform doors PSD. Therefore, people M22-M25 are not detected based on the output of the sensor 50, etc., and the determination unit 30 can determine that the safety of people M22-M25 is high.

[0283] Therefore, persons M22 to M25 are surrounded by a dotted frame, which indicates that they are highly safe.

[0284] Note that the rectangular frames of image 192 and images 193 to 196 may be distinguished by color, thickness, etc., rather than by line type, as in the case of the first example (images 182, 183) described above. Also, in this example, as in the case of the first example described above, images corresponding to unique identification numbers representing each of persons M21 to M25 may be added to images 192 to 196.

[0285] An image 197 shows the overall safety assessment result for the person on the home PF corresponding to the image 190, that is, the result of the above-mentioned overall assessment process.

[0286] In this example, the determination unit 30 determines that the safety of the person M21 is low, and as a result of the determination, the overall safety of the person on the platform PF corresponding to the image 190 is low. Therefore, text information "Decision: Danger" is drawn on the image 197, which indicates that the overall safety of the person on the platform PF corresponding to the image 190 is determined to be low by the determination unit 30. This allows the station staff of the station ST and the crew of the railway vehicle TRN to easily grasp the overall safety of the person on the platform PF corresponding to the image 190 by checking the image 197. Therefore, for example, in a situation where the railway vehicle TRN has entered the platform PF and is stopped, the station staff of the station ST and the crew of the railway vehicle TRN can determine whether or not the open door DR may be closed by checking the image 197.

[0287] Note that image 197 may be omitted. For example, as described above, when the comprehensive judgment process (step S40 in FIG. 2) is omitted, image 197 is omitted. Also, as described above, even if the tracking unit 20 cannot detect a specific person from image 190 because the person is hidden by the platform door PSD or surrounding people, the person may be detected by detection unit 60 based on the output of sensor 50, etc. In this case, an image representing information that a person not shown in image 190 is in a low safety state may be displayed superimposed on image 190.

[0288] Thus, in this example, on the display unit 40, images 191-197 corresponding to information on the person detection result by the tracking unit 20 and the judgment result of the person's safety by the judgment unit 30 are displayed superimposed on an image 190 showing the platform PF and the railway vehicle TRN. This allows station staff at the station ST and crew members of the railway vehicle TRN to easily grasp the person detection situation by the tracking unit 20 and the judgment result of the person's safety by the judgment unit 30, in addition to the situation of the person on the platform PF.

[0289] [Effect] Next, the operations of the monitoring system, the monitoring device, the monitoring method, and the program according to the present embodiment will be described.

[0290] In this embodiment, the monitoring system includes an imaging unit, a first detection unit, and a determination unit. The monitoring system is, for example, the above-mentioned monitoring system 1. The imaging unit is, for example, the above-mentioned imaging unit 10. The first detection unit is, for example, the above-mentioned tracking unit 20. The determination unit is, for example, the above-mentioned determination unit 30. Specifically, the imaging unit captures an image of a station platform. The station platform is, for example, the platform PF of the above-mentioned station ST. The first detection unit detects a person present on the platform based on an image acquired by the imaging unit. The determination unit determines whether the person's safety is in a first state in which the person's safety is relatively high or in a second state in which the person's safety is relatively low, depending on the position of the person detected by the first detection unit. More specifically, the determination unit determines whether the person's safety is in a first state or a second state based on the position of the person detected by the first detection unit and the movement history on the platform. For example, when the first detection unit detects a person in the first area, which is the farthest from one end of the platform where the railcar enters, among the first area, second area, and third area defined on the platform, the determination unit determines that the safety of the person is in the first state. The first area, second area, and third area are, for example, the above-mentioned areas R1, R2, and R3, respectively. Furthermore, when the first detection unit detects a person in the third area, which is the closest to one end of the platform where the railcar enters, among the first area, second area, and third area, the determination unit determines that the safety of the person is in the second state. Then, when a person is detected by the first detection unit in a second area between the first area and the third area among the first area, the second area, and the third area, the judgment unit judges that the person's safety is in a first state when the person has moved from the first area and is in the second area, and judges that the person's safety is in a second state when the person has moved from the third area and is in the second area.

[0291] In the present embodiment, the monitoring device may include a first detection unit and a determination unit. The monitoring device is, for example, an information processing device that realizes the functions of both the tracking unit 20 and the determination unit 30 described above. Specifically, the first detection unit detects a person present on the platform based on an image acquired by an imaging unit that images the station platform. The determination unit determines whether the person's safety is in a first state in which the person's safety is relatively high or in a second state in which the person's safety is relatively low, depending on the position of the person detected by the first detection unit. More specifically, the determination unit determines whether the person's safety is in a first state or a second state, based on the position of the person detected by the first detection unit and the movement history on the platform. For example, when the first detection unit detects a person in the first area, which is the farthest from one end of the platform where the railroad car enters, among the first area, the second area, and the third area defined on the platform, the determination unit determines that the person's safety is in the first state. Furthermore, when the first detection unit detects a person in the third area, which is the closest to one end of the platform where the railcar enters, of the first area, the second area, and the third area, the determination unit determines that the safety of the person is in the second state. Then, when the first detection unit detects a person in the second area, which is between the first area and the third area, of the first area, the second area, and the third area, the determination unit determines that the safety of the person is in the first state when the person has moved from the first area and is in the second area, and determines that the safety of the person is in the second state when the person has moved from the third area and is in the second area.

[0292] In the present embodiment, the information processing device may execute the monitoring method. The information processing device is, for example, an information processing device that realizes the functions of both the tracking unit 20 and the determination unit 30 described above. The monitoring method includes a detection step and a determination step. Specifically, in the detection step, the information processing device detects a person present on the platform based on an image acquired by an imaging unit that images the station platform. In addition, in the determination step, the information processing device determines whether the person's safety is in a first state in which the person's safety is relatively high or in a second state in which the person's safety is relatively low, depending on the position of the person detected in the detection step. In addition, in the determination step, if the person is detected in the first area, which is the farthest from one end of the platform where the railroad vehicle enters, among the first area, the second area, and the third area defined on the platform, in the detection step, the safety of the person is determined to be in the first state. In addition, in the determination step, if the person is detected in the third area, which is the closest to one end of the platform where the railroad vehicle enters, among the first area, the second area, and the third area, in the detection step, the safety of the person is determined to be in the second state. Then, in the determination step, if a person is detected in the detection step in a second area between the first area and the third area among the first area, the second area, and the third area, when the person has moved from the first area and is in the second area, the safety of the person is determined to be in a first state, and when the person has moved from the third area and is in the second area, the safety of the person is determined to be in a second state.

[0293] In the present embodiment, the information processing device may execute a program. Specifically, the program executes a detection step and a determination step.

[0294] This allows the second area to be set as a buffer area between the first area corresponding to the first state in which human safety is relatively high and the third area corresponding to the second state in which human safety is relatively low. Therefore, the monitoring system or the like can suppress a situation in which the human safety frequently switches between the first state and the second state.

[0295] In the present embodiment, when a person is detected by the first detection unit in the third area, the determination unit may determine that the safety of the person is in the first state when the person moves from the second area and stays in the third area for a relatively long time compared to the predetermined time. The same may be true for the determination step. The predetermined time is, for example, the above-mentioned predetermined time T1.

[0296] As a result, even if a person is detected in the third area, if that person continues to stay in the third area for a certain period of time, the monitoring system etc. can determine that the person is unlikely to move toward one end of the platform where railroad cars enter, and that the person's safety is relatively high. Therefore, the monitoring system etc. can more appropriately determine the safety of people in the third area.

[0297] In the present embodiment, the determination unit may determine that the safety of a person is in the second state when the first detection unit detects the person in the fourth region, which is closer to one end than the third region, among the first region, the second region, the third region, and the fourth region. The same may be true for the determination step. The fourth region is, for example, the above-mentioned region R4.

[0298] This allows the monitoring system etc. to evaluate the safety of people who stay for a long time in a place relatively close to the end where the railroad vehicle enters as relatively high, while evaluating the safety of people very close to the end where the railroad vehicle enters as relatively low. Therefore, the monitoring system etc. can more appropriately determine the safety of people in the areas close to the end of the platform where the railroad vehicle enters (the third area and the fourth area).

[0299] In the present embodiment, when a person is detected in the second area by the first detection unit, the determination unit may determine that the person's safety is in the first state when the person has moved from the third area to the second area and the safety of the person was determined to be in the first state in the third area. The same may be applied to the determination step.

[0300] This allows the monitoring system etc. to determine, for example, that a person who has been staying near the border between the second area and the third area for a certain period of time is unlikely to move toward the fourth area and that the person's safety is relatively high. Therefore, the monitoring system etc. can more appropriately determine the safety of people in the second area.

[0301] In the present embodiment, the determination unit may determine that the safety of a person is in the first state when the first detection unit detects a person in the third area, the person moves from the third area to the second area and then returns to the third area, and the safety of the person was determined to be in the first state in the second area. The same may be true for the determination step.

[0302] This allows the monitoring system etc. to determine, for example, that a person who has been staying near the border between the second area and the third area for a certain period of time is unlikely to move toward the fourth area and that the person's safety is relatively high. Therefore, the monitoring system etc. can more appropriately determine the safety of people in the third area.

[0303] In the present embodiment, when a person is detected by the first detection unit in the second area or the third area, the determination unit may determine that the safety of the person is in the first state when the person moves from the third area to the second area and stays in the combined area of ​​the second area and the third area for a relatively long time relative to the predetermined time. The same may be applied to the determination step. The predetermined time is, for example, the above-mentioned predetermined time T1.

[0304] This allows the monitoring system etc. to determine, for example, that a person who has been staying near the border between the second and third areas for a certain period of time is unlikely to move toward the fourth area and that the person's safety is relatively high. Therefore, the monitoring system etc. can more appropriately determine the safety of people in the second and third areas.

[0305] In the present embodiment, when a person is detected by the first detection unit in the third area, the determination unit may determine that the person's safety is in the first state if the person moves from the third area to the fourth area and then returns to the third area in a relatively short time compared to the predetermined time and the person's safety was previously determined to be in the first state in the third area. The same may be true for the determination step. The predetermined time is, for example, the above-mentioned predetermined time T2.

[0306] As a result, the monitoring system etc. can determine, for example, that a person who moves from the third area to the fourth area, but realizes that the train is not the intended one and returns to the third area is unlikely to return to the fourth area, and that the person's safety is relatively high. Therefore, the monitoring system etc. can more appropriately determine the safety of people in the third area.

[0307] In the present embodiment, the imaging unit may capture an image of the platform and the railcar. The railcar is the above-mentioned railcar TRN. The first detection unit may detect the platform and a person in the railcar entering the platform based on the image acquired by the imaging unit. When the first detection unit detects a person in the third area, the determination unit may determine that the person's safety is in the first state when the person moves from the third area to the fourth area, boards the railcar from the fourth area in a relatively short time compared to the predetermined time, gets off the railcar, and returns from the fourth area to the third area, and the person's safety was determined to be in the first state in the previous third area. The same may be applied to the determination step. The predetermined time is, for example, the above-mentioned predetermined time T3.

[0308] As a result, the monitoring system etc. can determine that, for example, a person who boarded a train from the third area, but realized that it was not the intended train and returned to the third area is unlikely to head to the fourth area again, and that the person's safety is relatively high. Therefore, the monitoring system etc. can more appropriately determine the safety of people in the third area.

[0309] In the present embodiment, the determination unit may determine whether the safety of each person detected by the first detection unit is in the first state or the second state, and may perform a determination regarding the overall safety of the people in the image based on the individual determination results. The same may be true for the determination step.

[0310] This enables the monitoring system or the like to make a judgment regarding the overall safety of all people within the image range on the platform based on the results of judging the safety of each individual person.

[0311] In addition, in this embodiment, the monitoring system may include a display unit that displays the image captured by the imaging unit and information related to the determination result of the determination unit.

[0312] This allows, for example, station staff or train crew members to check the situation on the platform as well as the assessment result of the level of safety of people corresponding to that situation through the display unit.

[0313] In addition, in this embodiment, the judgment unit may use a different method for determining whether the safety of the person detected by the first detection unit is in a first state or a second state, depending on whether the degree of congestion on the platform is relatively high compared to a predetermined standard.

[0314] This enables the determination unit to appropriately determine the level of safety of occupants on the platform according to the level of congestion on the platform.

[0315] In addition, in this embodiment, when the congestion level of the platform is relatively high compared to a predetermined standard and a person is detected on the platform by the first detection unit, the judgment unit may determine whether the person's safety is in a first state or a second state based only on the person's location and locations in the person's past movement history.

[0316] For example, when the degree of congestion on a platform is relatively high, the space available for movement is limited, and it is possible that a judgment of the level of safety based on a person's past movement history does not function properly. In addition, when the degree of congestion on a platform is relatively high, it may become difficult to track a person's movement due to people overlapping each other on an image, etc. In contrast, when the degree of congestion on a platform is relatively high, a monitoring system or the like can judge the level of safety of a person based only on the person's position, regardless of the person's movement history. Therefore, the monitoring system or the like can more appropriately judge the level of safety of a person on the platform in accordance with the degree of congestion on the platform.

[0317] In this embodiment, when the first detection unit detects a person on the platform in a state where the degree of congestion on the platform is relatively high with respect to a predetermined standard, the determination unit may determine that the safety of the person is in a first state if the person is located outside a predetermined area including one end of the platform, and may determine that the safety of the person is in a second state if the person is located within the predetermined area. The predetermined area is, for example, the above-mentioned area R5.

[0318] This allows the monitoring system, etc. to determine the safety of a person based on whether or not the person is detected within a specified area near one end of the platform where railway cars enter when the platform is relatively crowded.

[0319] In this embodiment, the boundary line on the opposite side to one end of the predetermined region may be set closer to the one end than the boundary line between the second region and the third region.

[0320] This allows the monitoring system etc. to determine that the safety of a person is relatively low, limited to a person who is taking or is more likely to take the action of getting off or boarding a railway vehicle near one end of the platform where railway vehicles enter the station. Therefore, for example, when the platform is relatively crowded, a situation in which a person whose safety is determined to be relatively low continues to be present, and this can prevent a situation in which railway vehicle operation is disrupted.

[0321] In the present embodiment, the monitoring system or the like may also include an acquisition unit. The acquisition unit is, for example, the sensor 50 described above. Specifically, the acquisition unit is provided separately from the imaging unit, and acquires information regarding the presence of a person near one end of the platform where the railroad car enters. Then, when the first detection unit detects a person in a state where the degree of congestion on the platform is relatively high with respect to a predetermined standard, the determination unit may determine whether the safety of the person is in a first state or a second state based on the information acquired by the acquisition unit.

[0322] As a result, when the platform is relatively crowded, the monitoring system, etc. can determine the safety of a detected person by taking into consideration the information acquired by the acquisition unit, i.e., information regarding the presence of a person near one end of the platform, and information about the detected person in a comprehensive manner.

[0323] In addition, in this embodiment, the monitoring system etc. may include a second detection unit. The second detection unit is, for example, the detection unit 60 described above. Specifically, the second detection unit detects a person near one end of the platform based on information acquired by the acquisition unit. The platform door is, for example, the platform door PSD described above. Then, when the degree of congestion on the platform is relatively high with respect to a predetermined standard and a person is detected by the first detection unit, the determination unit may determine that the safety of the person is in a second state when the person is detected by the second detection unit.

[0324] This allows the monitoring system, etc. to determine that when the platform is relatively crowded, the safety of people detected based on images of the platform near one end of the platform where railway cars enter is relatively low.

[0325] In addition, in this embodiment, the monitoring system or the like may include an acquisition unit and a second detection unit. The acquisition unit is, for example, the sensor 50 described above. The second detection unit is, for example, the detection unit 60 described above. Specifically, the acquisition unit acquires information regarding the presence of a person near the opening of the platform door provided along one end of the platform. Furthermore, the second detection unit detects a person near the opening of the platform door based on the information acquired by the acquisition unit. Then, the determination unit may determine whether the safety of the person detected by the first detection unit is in a first state or a second state, and when a person is detected by the second detection unit, determine that the safety of the person is in the second state.

[0326] This allows the monitoring system etc. to use the information from the acquisition unit to detect a child or a hairdresser in a wheelchair who is hidden by the platform door and cannot be detected in the image from the imaging unit, and to determine that the safety of that person is relatively low. Therefore, the monitoring system etc. can further improve the safety of the platform.

[0327] In addition, in this embodiment, the acquisition unit may be a sensor that acquires information about people passing through an opening in a platform door provided along one end of the platform, or a camera attached to a railway vehicle entering the platform.

[0328] This allows the surveillance system or the like to detect the presence of a person near one end of the platform where the railway car enters.

[0329] Although the embodiments have been described in detail above, the present disclosure is not limited to such specific embodiments, and various modifications and variations are possible within the scope of the gist described in the claims. [Explanation of symbols]

[0330] 1. Surveillance System 10. Imaging unit 20 Tracking Department 30 Judgment section 40 Display section 50 Sensors 60 Detection unit B1~B4 Boundary line PF Home R1~R5 area ST Station TRN rail vehicles

Claims

1. An imaging unit for imaging the station platform, A detection unit that detects people present on the platform based on the image acquired by the imaging unit, The system includes a determination unit that determines, based on the location of the person detected by the detection unit, whether the person is in a first state where their safety is relatively high or a second state where their safety is relatively low. The determination unit determines that if a person detected by the detection unit is present in a predetermined area including one end of the platform into which a railway vehicle enters, the person's safety is in the second state. The outer edge of the predetermined region opposite to the one end is closer to the one end when the congestion level of the platform is relatively high compared to the predetermined standard, compared to when the congestion level is relatively low compared to the predetermined standard. A monitoring system.

2. A detection unit that detects a person present on a station platform based on an image acquired by an imaging unit that images the station platform, The system includes a determination unit that determines, based on the location of the person detected by the detection unit, whether the person is in a first state where their safety is relatively high or a second state where their safety is relatively low. The determination unit determines that if a person detected by the detection unit is present in a predetermined area including one end of the platform into which a railway vehicle enters, the person's safety is in the second state. The outer edge of the predetermined region opposite to the one end is closer to the one end when the congestion level of the platform is relatively high compared to the predetermined standard, compared to when the congestion level is relatively low compared to the predetermined standard. Monitoring equipment.

3. An information processing device includes a detection step of detecting a person present on a station platform based on an image acquired by an imaging unit that images the station platform, The information processing device includes a determination step that determines, based on the location of the person detected in the detection step, whether the person is in a first state where their safety is relatively high or a second state where their safety is relatively low. In the determination step, if a person detected in the detection step is present in a predetermined area including one end of the platform into which the railway vehicle enters, it is determined that the person's safety is in the second state. The outer edge of the predetermined region opposite to the one end is closer to the one end when the congestion level of the platform is relatively high compared to the predetermined standard, compared to when the congestion level is relatively low compared to the predetermined standard. Monitoring method.

4. Information processing device, A detection step that detects people present on the station platform based on images acquired by an imaging unit that images the station platform, A determination step is performed to determine whether the person's safety is in a first state (relatively high) or a second state (relatively low) according to the location of the person detected in the above detection step. In the determination step, if a person detected in the detection step is present in a predetermined area including one end of the platform into which the railway vehicle enters, it is determined that the person's safety is in the second state. The outer edge of the predetermined region opposite to the one end is closer to the one end when the congestion level of the platform is relatively high compared to the predetermined standard, compared to when the congestion level is relatively low compared to the predetermined standard. program.

5. An imaging unit that captures images of the station platform, A first detection unit detects people present on the platform based on the image acquired by the imaging unit, The system includes a determination unit that determines, based on the location of the person detected by the first detection unit, whether the person's safety is in a first state (relatively high) or a second state (relatively low), The determination unit determines that, among the first, second, and third regions defined on the platform, if a person is detected by the first detection unit in the first region, which is furthest from one end of the platform where a railway vehicle enters, the person's safety is in the first state; if a person is detected by the first detection unit in the third region, which is closest to the one end, the person's safety is in the second state; if a person is detected by the first detection unit in the second region, which is located between the first and third regions, the person's safety is in the first state when the person has moved from the first region to the second region; and if the person has moved from the third region to the second region, the person's safety is in the second state. A monitoring system.

6. The determination unit determines that when a person is detected in the third area by the first detection unit, the person has moved from the second area and has stayed in the third area for a relatively long period of time, and that the person's safety is in the first state. The monitoring system according to claim 5.

7. The determination unit determines that a person's safety is in the second state if the first detection unit detects a person in the fourth region, which is closer to one end than the third region, among the first, second, third, and fourth regions. The monitoring system according to claim 6.

8. The determination unit determines that a person is in the first state of safety when the first detection unit detects a person in the second region, and the determination unit determines that the person has moved from the third region to the second region and that the safety of the person was determined to be in the first state in the third region. The monitoring system according to claim 6 or 7.

9. The determination unit determines that a person is in the first state of safety when the first detection unit detects a person in the third region, the person has moved from the third region to the second region and then returned to the third region, and the determination unit has determined that the person's safety is in the first state in the second region. The monitoring system according to claim 8.

10. The determination unit determines that a person is in the first state of safety when the first detection unit detects a person in the second or third region, and the person moves from the third region to the second region and then stays in the combined region of the second and third regions for a relatively long period of time relative to a predetermined time. The monitoring system according to claim 6 or 7.

11. The determination unit determines that a person is in the first state of safety if, when a person is detected in the third area by the first detection unit, the person has moved from the third area to the fourth area and then returned to the third area in a relatively short time relative to a predetermined time, and the determination unit determined that the person's safety was in the first state in the previous instance in the third area. The monitoring system according to claim 7.

12. The imaging unit captures images of the platform and the railway vehicle. The first detection unit detects the platform and people inside the railway vehicle entering the platform based on the image acquired by the imaging unit. The determination unit determines that the person's safety is in the first state if, when the first detection unit detects a person in the third area, the person moves from the third area to the fourth area, boards the railway vehicle from the fourth area in a relatively short time relative to a predetermined time, disembarks from the railway vehicle, returns from the fourth area to the third area, and the determination unit determined that the person's safety was in the first state in the previous instance in the third area. The monitoring system according to claim 7.

13. The determination unit determines, for each person detected by the first detection unit, whether that person's safety is in the first state or the second state, and based on the individual determination results, makes a determination regarding the overall safety of the person in the image. The monitoring system according to any one of claims 5 to 7.

14. The system includes a display unit that displays an image acquired by the imaging unit and information related to the determination result of the determination unit. The monitoring system according to any one of claims 5 to 7.

15. The determination unit modifies the method for determining whether the safety of a person detected by the first detection unit is in the first state or the second state, depending on whether the level of congestion on the platform is relatively high compared to a predetermined standard. The monitoring system according to any one of claims 5 to 7.

16. The determination unit, when the congestion level of the platform is relatively high compared to the predetermined standard and the first detection unit detects a person on the platform, determines whether the person's safety is in the first state or the second state based only on the person's location and the location in their past movement history. The monitoring system according to claim 15.

17. The determination unit, when the degree of congestion on the platform is relatively high compared to the predetermined standard, and when the first detection unit detects a person on the platform, determines that the person's safety is in the first state if the person's position is outside the predetermined area including one end of the platform, and determines that the person's safety is in the second state if the person's position is within the predetermined area. The monitoring system according to claim 16.

18. The boundary line of the predetermined region on the opposite side of the one end is set closer to the one end than the boundary line between the second region and the third region. The monitoring system according to claim 17.

19. A separate unit is provided in addition to the imaging unit, which acquires information regarding the presence of a person near one end of the platform. The determination unit, when the congestion level of the platform is relatively high compared to the predetermined standard and a person is detected by the first detection unit, determines, based on the information acquired by the acquisition unit, whether the person's safety is in the first state or the second state. The monitoring system according to claim 15.

20. The system includes a second detection unit that detects a person near one end of the platform based on the information acquired by the acquisition unit, The determination unit determines that the safety of a person is in the second state when the congestion level of the platform is relatively high compared to the predetermined standard, and when a person is detected by the first detection unit, and when that person is detected by the second detection unit. The monitoring system according to claim 19.

21. The acquisition unit is a sensor that acquires information about a person passing through the opening of a platform door provided along one end of the platform, or a camera attached to a railway vehicle entering the platform. The monitoring system according to claim 19.

22. An acquisition unit that acquires information regarding the presence of a person near the opening of a platform door provided along one end of the platform, The system includes a second detection unit that detects a person near the opening of the platform door based on the information acquired by the acquisition unit, The determination unit determines whether the safety of the person detected by the first detection unit is in the first state or the second state, and if a person is detected by the second detection unit, it determines that the person's safety is in the second state. The monitoring system according to any one of claims 5 to 7.

23. A detection unit detects people present on the station platform based on images acquired by an imaging unit that images the station platform, The system includes a determination unit that determines, based on the location of the person detected by the detection unit, whether the person is in a first state where their safety is relatively high or a second state where their safety is relatively low. The determination unit determines that, among the first, second, and third regions defined on the platform, if the detection unit detects a person in the first region, which is furthest from one end of the platform where a railway vehicle enters, the person's safety is in the first state; if the detection unit detects a person in the third region, which is closest to the one end, the person's safety is in the second state; if the detection unit detects a person in the second region, which is located between the first and third regions, the person's safety is in the first state if they have moved from the first region to the second region; and if the person has moved from the third region to the second region, the person's safety is in the second state. Monitoring equipment.

24. The information processing device performs a detection step of detecting a person present on the station platform based on an image acquired by an imaging unit that images the station platform, The information processing device includes a determination step that determines, based on the location of the person detected in the detection step, whether the person is in a first state where their safety is relatively high or a second state where their safety is relatively low. In the determination step, if a person is detected by the detection step in the first area, which is the furthest from one end of the platform where the railway vehicle enters, among the first, second, and third areas defined on the platform, it is determined that the person's safety is in the first state; if a person is detected by the detection step in the third area, which is closest to the one end, it is determined that the person's safety is in the second state; if a person is detected by the detection step in the second area, which is located between the first and third areas, it is determined that the person's safety is in the first state when the person has moved from the first area to the second area; and if the person has moved from the third area to the second area, it is determined that the person's safety is in the second state. Monitoring method.

25. In an information processing device, A detection step that detects people present on the station platform based on images acquired by an imaging unit that images the station platform, A determination step is performed to determine whether the person's safety is in a first state (relatively high) or a second state (relatively low) according to the location of the person detected in the above detection step. In the determination step, if a person is detected by the detection step in the first area, which is the furthest from one end of the platform where the railway vehicle enters, among the first, second, and third areas defined on the platform, it is determined that the person's safety is in the first state; if a person is detected by the detection step in the third area, which is closest to the one end, it is determined that the person's safety is in the second state; if a person is detected by the detection step in the second area, which is located between the first and third areas, it is determined that the person's safety is in the first state when the person has moved from the first area to the second area; and if the person has moved from the third area to the second area, it is determined that the person's safety is in the second state. program.

Images