Fall detection device and fall detection system equipped therewith, fall detection method and fall detection program

Abstract

To provide a fall detection device enabling a station clerk or the like to immediately recognize which car of a train a fall event occurs when the occurrence of the fall event is detected while suppressing introduction cost and enabling the station clerk or the like to take appropriate action.SOLUTION: A fall detection device 4 includes a scanning data acquisition unit 14, a fall detection unit 15, a wheel detection area setting unit 16, a wheel detection unit 17, a wheel detection frequency detection unit 18, and a car number specification unit 19. The wheel detection area setting unit 16 sets such that a length L in a car traveling direction is made shorter than a length X between a front wheel position and a rear wheel position of the same car, and is made longer than a length Y between a center position of a rear wheel group of a preceding car of adjacent cars and a center position of a front wheel group of a succeeding car, in a wheel detection area set in a stop position of the rear wheel of the car of a train. The car number specification unit 19 determines that a car stopped in the wheel detection area is n-th of the train when the detection frequency of a wheel detected by the wheel detection frequency detection unit 18 is n.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a fall detection device for detecting a person who has fallen onto a track through which a train passes from a platform of a railway station, a fall detection system including the same, a fall detection method, and a fall detection program.

Background Art

[0002] At railway stations, railway users may accidentally fall from the platform onto the track. Also, when a train is stopped at a railway station, a railway user may fall into the gap between the platform and the train. Conventionally, systems for detecting these track falls and gap falls and reporting them to relevant personnel (such as train dispatchers, station staff, conductors, or drivers) have been proposed. For example, Patent Document 1 discloses a system including a laser scanner that emits a laser while deflecting it to form a two-dimensional scanning range. The laser scanner is installed at a height below the platform and above the track in a posture where the laser is emitted horizontally. An arbitrary area within the horizontal scanning range formed at that height is set as a monitoring area. The laser scanner detects whether an object exists within the monitoring area.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, the above conventional system has the following problems. In other words, on a platform, for example, the fifth car of train A might stop at the same position as the third car of another train B, which has a different number of cars. Therefore, when a fall detection sensor detects a fall, simply displaying the location of the sensor as an alarm is not enough for station staff to immediately recognize which car the fall occurred on. This could lead to delays in responding to the scene of the fall or making announcements, preventing appropriate action from being taken.

[0005] In order to accommodate trains with different numbers of carriages, it is conceivable that multiple fall detection sensors be installed to cover all areas where carriages may stop. In this case, a carriage detection area is set up to check for the presence of carriages on the track, and each carriage detection area detects stopped carriages. By checking the position and number of carriages, it is possible to determine which carriage each fall detection sensor is monitoring.

[0006] However, since a system that covers all vehicles would result in high implementation costs for the fall detection system, in order to reduce implementation costs, the installation locations of fall detection sensors are sometimes limited to places where falls are likely to occur, such as curved sections of the platform. In this case, if trains A and B with different numbers of vehicles stop at the location where the fall detection sensor is installed, it is difficult to immediately recognize which vehicle it is.

[0007] Alternatively, station staff could carry a correspondence table for each train schedule, linking the fall detection sensor to the train car number, and use it to verify the location of the fall and the train car number each time a fall occurs. However, even in this case, it would be extremely time-consuming for station staff, making immediate response difficult and potentially leading to delays in rushing to the scene of the fall or making announcements, thus preventing them from taking appropriate action.

[0008] The object of the present invention is to provide a fall detection device, a fall detection system equipped with the same, a fall detection method, and a fall detection program that can suppress introduction costs and immediately allow station staff to recognize which train car the fall occurred in when a fall incident occurs, enabling them to take appropriate action. [Means for solving the problem]

[0009] The fall detection device according to the first invention is a fall detection device that detects a person who has fallen from the platform of a target station onto the track through which a train passes, and comprises a light receiving data acquisition unit, an object detection unit, a wheel detection area setting unit, a wheel detection unit, a wheel detection count detection unit, and a vehicle number identification unit. The light receiving data acquisition unit acquires reflected light data at predetermined angles from an optical scanner that scans light within a predetermined angular range in a fall detection area set on the track and receives the reflected light. The object detection unit detects the intrusion of an object onto the track based on the light receiving data acquired by the light receiving data acquisition unit. The wheel detection area setting unit sets the length (L) in the direction of vehicle travel of the wheel detection area set at the stopping position of the rear wheels of the train vehicles to be shorter than the length (X) between the front wheel position and the rear wheel position in the same vehicle, and longer than the length (Y) between the center position of the rear wheel group of the front vehicle and the center position of the front wheel group of the rear vehicle in adjacent vehicles. The wheel detection unit detects the wheels of a train entering the platform within the wheel detection area set by the wheel detection area setting unit. The wheel detection count detection unit counts the number of times the wheels have been detected by the wheel detection unit when the train enters the platform. The vehicle number identification unit determines that the vehicle that has stopped in the wheel detection area is the nth vehicle of the train, assuming that the number of wheel detections by the wheel detection count detection unit is n.

[0010] Here, the wheel detection area is set to recognize the front and rear wheels of two adjacent vehicles as a single wheel group. The number of times (n times) that wheels pass through the wheel detection area is counted, and the number of vehicles (nth vehicle) that has stopped in the wheel detection area is determined. Here, the length (L) of the wheel detection area in the direction of vehicle travel, as set by the wheel detection area setting unit, is a distance set along the longitudinal direction of the platform according to the length of the train cars, for example, 5m.

[0011] As a result, the wheel detection area is set up to recognize the front and rear wheels of two adjacent vehicles as a single wheel group. Therefore, by counting the number of times wheels are detected within the wheel detection area, it is easy to determine which vehicle the stopped train is located in. Consequently, in the event of a fall, the vehicle in which the fall occurred can be reported to the station staff waiting room or displayed on a fall location indicator. As a result, while keeping implementation costs down, when a fall incident occurs, station staff can immediately recognize which train car the incident occurred in, allowing them to take appropriate action.

[0012] The fall detection device according to the second invention is the same as the fall detection device according to the first invention, wherein the wheel detection area setting unit sets a detection time (T1) for detecting wheels passing through the wheel detection area when a train enters the platform, and a detection time (T2) for detecting wheels that have stopped in the wheel detection area.

[0013] Here, the detection time in the wheel detection area is set separately for when the vehicle is passing and when it is stopped. This allows for setting appropriate detection times for wheels within the wheel detection area, such as setting a shorter time when a vehicle is passing by and a longer time when the vehicle is stopped.

[0014] The fall detection device according to the third invention is the fall detection device according to the second invention, wherein the detection time (T1) is set to be shorter than the detection time (T2). Here, the detection time for detecting passing wheels is set to be short (e.g., 100ms), while the detection time for detecting stopped wheels is set to be long (e.g., 5000ms). This allows the system to immediately detect when a train enters the wheel detection area within a pre-set short detection time, and to reliably detect when a wheel stops within the wheel detection area within a pre-set long detection time.

[0015] The fall detection device according to the fourth invention is a fall detection device according to any one of the first to third inventions, wherein the wheel detection unit detects wheels based on light-receiving data acquired by the light-receiving data acquisition unit. This allows for the detection of not only falling objects but also wheels, using the light data acquired from the photodetector.

[0016] The fall detection device according to the fifth invention is a fall detection device according to any one of the first to fourth inventions, wherein the wheel detection area setting unit presets a first value as the length (L) of the wheel detection area in the direction of vehicle travel for detecting the wheels when the wheels of a train entering the platform pass through the wheel detection area, and presets a second value different from the first value for detecting the wheels when the wheels of the train stop in the wheel detection area. This allows for more accurate wheel detection by setting the length (L) of the wheel detection area in the direction of vehicle movement to different values ​​when the train is passing and when it is stopped.

[0017] The fall detection device according to the sixth invention is the fall detection device according to the fifth invention, wherein the second value is smaller than the first value. As a result, the second value set when detecting a train wheel while it is stopped in the wheel detection area is smaller than the first value set when detecting a wheel passing through the wheel detection area. Therefore, even when the second value is set to a smaller value than the first value, wheel detection can be performed effectively. Also, by setting the second value to be smaller than the first value, it is possible to avoid misjudging a non-stopped wheel as a stopped wheel because the train has stayed in the wheel detection area for a long time during slow operation.

[0018] The fall detection device according to the seventh invention is a fall detection device according to any one of the first to sixth inventions, and the wheel detection area setting unit sets the length (D) in the vehicle width direction of the vehicle of the train in the wheel detection area to a length slightly larger than the vehicle width of the vehicle. Thereby, when detecting the presence or absence of a wheel by detecting the reflected light of the light irradiated from the photodetector, it is possible to appropriately detect the front wheel and the rear wheel as viewed from the photodetector in the wheel detection area.

[0019] The fall detection device according to the eighth invention is a fall detection device according to any one of the first to seventh inventions, and further includes a storage unit that stores the determination result in the vehicle number specifying unit. Thereby, when a fall event occurs, it is possible to immediately recognize and notify which vehicle is being monitored as the determination result in the vehicle number specifying unit stored in the storage unit.

[0020] The fall detection system according to the ninth invention includes a fall detection device according to any one of the first to eighth inventions, and a light scanner that scans light within a predetermined angle range in a fall detection area set on a track and receives the reflected light of the light. Thereby, as described above, it is possible to configure a system that can immediately let station staff and the like recognize which vehicle of the train the fall event has occurred in when detecting the occurrence of a fall event while suppressing the introduction cost and enable appropriate countermeasures to be taken.

[0021] The fall detection system according to the tenth invention is the fall detection system according to the ninth invention, wherein the light scanner is installed near the position where the connection part of two adjacent vehicles stops. The object detection unit performs fall detection on two vehicles. As a result, for example, when the number of detections in the wheel detection number detection unit is n times, the object detection unit can monitor a person who has fallen or the like while recognizing that the fall detection is being performed at the n-th vehicle and the (n + 1)-th vehicle.

[0022] The fall detection system according to the eleventh invention is the fall detection system according to the ninth or tenth invention, and further includes a fall position indicator that indicates the position where the object is detected in the object detection unit or the position where the fall event has occurred. As a result, for example, in the fall position indicator installed on the platform of a station, by displaying the occurrence of the fall event detected by the fall detection device and the position (vehicle number) where the fall event has occurred, station staff and the like can quickly rush to the position where the fall event has occurred and take appropriate measures.

[0023] The fall detection method according to the 12th invention is a fall detection method for detecting a person who has fallen from the platform of a target station onto the track through which a train passes, and comprises a wheel detection area setting step, a light reception data acquisition step, a wheel detection step, a wheel detection count detection step, a vehicle number identification step, an object detection step, and a notification step. The wheel detection area setting step sets the length (L) in the direction of vehicle travel of the wheel detection area, which is set at the stopping position of the rear wheels of the train vehicles, so that it is shorter than the length (X) between the front wheel position and the rear wheel position on the same vehicle, and longer than the length (Y) between the center position of the rear wheel group of the front vehicle and the center position of the front wheel group of the rear vehicle between adjacent vehicles. The light reception data acquisition step acquires reflected light reception data at predetermined angles from an optical scanner that scans light within a predetermined angular range in the fall detection area set on the track and receives reflected light. The wheel detection step detects the wheels of a train entering the platform in the wheel detection area set in the wheel detection area setting step, based on the light data received in the light data acquisition step. The wheel detection count detection step counts the number of times the wheels detected by the wheel detection step have been detected when the train enters the platform. The vehicle number identification step determines that the vehicle stopped in the wheel detection area is the nth car of the train, assuming that the number of wheel detections in the wheel detection count detection step is n. The object detection step detects the intrusion of an object onto the track based on the light data received in the light data acquisition step. The notification step notifies the result of the object detection step, which detected the intrusion of an object, and that the location where the object was detected is the nth car of the train.

[0024] Here, the wheel detection area is set to recognize the front and rear wheels of two adjacent vehicles as a single wheel group. The number of times (n times) that wheels pass through the wheel detection area is counted, and the number of vehicles (nth vehicle) that has stopped in the wheel detection area is determined. Here, the length (L) of the wheel detection area in the direction of vehicle travel, as set by the wheel detection area setting unit, is a distance set along the longitudinal direction of the platform according to the length of the train cars, for example, 5m.

[0025] As a result, the wheel detection area is set up to recognize the front and rear wheels of two adjacent vehicles as a single wheel group. Therefore, by counting the number of times wheels are detected within the wheel detection area, it is easy to determine which vehicle the stopped train is located in. Consequently, in the event of a fall, the vehicle in which the fall occurred can be reported to the station staff waiting room or displayed on a fall location indicator. As a result, while keeping implementation costs down, when a fall incident occurs, station staff can immediately recognize which train car the incident occurred in, allowing them to take appropriate action.

[0026] The fall detection program according to the 13th invention is a fall detection program that detects a person who has fallen from the platform of a target station onto the track through which a train passes, and causes a computer to execute a fall detection method comprising a wheel detection area setting step, a light reception data acquisition step, a wheel detection step, a wheel detection count detection step, a vehicle number identification step, an object detection step, and a notification step. The wheel detection area setting step sets the length (L) in the direction of vehicle travel of a wheel detection area set at the stopping position of the rear wheels of a train vehicle so that it is shorter than the length (X) between the front wheel position and the rear wheel position of the same vehicle, and longer than the length (Y) between the center position of the rear wheel group of the front vehicle and the center position of the front wheel group of the rear vehicle in adjacent vehicles. The light reception data acquisition step acquires reflected light reception data at predetermined angles from an optical scanner that scans light within a predetermined angular range in the fall detection area set on the track and receives reflected light. The wheel detection step detects the wheels of a train entering the platform in the wheel detection area set in the wheel detection area setting step, based on the light data received in the light data acquisition step. The wheel detection count detection step counts the number of times the wheels detected by the wheel detection step have been detected when the train enters the platform. The vehicle number identification step determines that the vehicle stopped in the wheel detection area is the nth car of the train, assuming that the number of wheel detections in the wheel detection count detection step is n. The object detection step detects the intrusion of an object onto the track based on the light data received in the light data acquisition step. The notification step notifies the result of the object detection step, which detected the intrusion of an object, and that the location where the object was detected is the nth car of the train.

[0027] Here, the wheel detection area is set to recognize the front and rear wheels of two adjacent vehicles as a single wheel group. The number of times (n times) that wheels pass through the wheel detection area is counted, and the number of vehicles (nth vehicle) that has stopped in the wheel detection area is determined. Here, the length (L) of the wheel detection area in the direction of vehicle travel, as set by the wheel detection area setting unit, is a distance set along the longitudinal direction of the platform according to the length of the train cars, for example, 5m.

[0028] As a result, the wheel detection area is set up to recognize the front and rear wheels of two adjacent vehicles as a single wheel group. Therefore, by counting the number of times wheels are detected within the wheel detection area, it is easy to determine which vehicle the stopped train is located in. Consequently, in the event of a fall, the vehicle in which the fall occurred can be reported to the station staff waiting room or displayed on a fall location indicator.

[0029] As a result, while keeping implementation costs down, when a fall incident occurs, station staff can immediately recognize which train car the incident occurred in, allowing them to take appropriate action. [Effects of the Invention]

[0030] The fall detection device according to the present invention allows for the detection of a fall incident while keeping implementation costs down. It also enables station staff to immediately recognize which train car the fall occurred in, allowing them to take appropriate action. [Brief explanation of the drawing]

[0031] [Figure 1] A diagram showing the configuration of a fall detection system equipped with a fall detection device according to one embodiment of the present invention. [Figure 2] A control block diagram showing the configuration of the fall detection device included in the fall detection system in Figure 1. [Figure 3] (a) is a plan view showing the scanning range of the optical scanner when there is no train on the station platform. (b) is a plan view showing the scanning range of the optical scanner when there is a train on the station platform. [Figure 4]A cross-sectional view, seen from the direction of the train's movement, showing the positional relationship between the optical scanner installed in the space beneath the station platform and the train stopped at the station. [Figure 5] (a) is a perspective view showing the fall detection area when the train is not on the station platform. (b) is a perspective view showing the gap fall detection area when the train is on the station platform. [Figure 6] This diagram shows the distance (X) between the front and rear wheel positions of a train entering a station platform, and the distance (Y) between the center of the rear wheel group of the front car and the center of the front wheel group of the rear car in adjacent cars. [Figure 7] A schematic diagram showing two trains, A and B, with different lengths (number of carriages), entering and stopping at a station platform. [Figure 8] (a) is a diagram showing the wheel detection area for detecting the wheels of a train entering the platform, and the output result from the lower scanner shown in Figure 4, etc. (b) is a diagram showing the output result from the lower scanner when the front wheels of the first car passing through the wheel detection area are detected. [Figure 9] (a) is a diagram showing the output from the lower scanner when a group of wheels, including the rear wheels of the first car and the front wheels of the second car, passing through the wheel detection area, is detected. (b) is a diagram showing the output from the lower scanner when a group of wheels, including the rear wheels of the second car and the front wheels of the third car, passing through the wheel detection area, is detected. [Figure 10] (a) is a diagram showing the output from the lower scanner when a group of wheels, including the rear wheels of the 6th car and the front wheels of the 7th car, passing through the wheel detection area, is detected. (b) is a diagram showing the output from the lower scanner when the rear wheels of the 7th car, which has stopped in the wheel detection area, are detected. [Figure 11] A flowchart showing the processing flow of the fall detection method performed by the fall detection device in Figure 2. [Modes for carrying out the invention]

[0032] A fall detection system 1 equipped with a fall detection device 4 according to one embodiment of the present invention will be described below with reference to Figures 1 to 11. In this embodiment, unnecessary details may be omitted. For example, detailed explanations of already well-known matters or redundant explanations of substantially identical configurations may be omitted. This is to avoid the following explanation becoming unnecessarily verbose and to facilitate understanding for those skilled in the art.

[0033] Furthermore, the applicant provides the accompanying drawings and the following description so that those skilled in the art may fully understand the present invention, and not intends to limit the subject matter described in the claims. In the following explanation, "longitudinal direction" refers to the longitudinal direction of the railway station platform P (left-right direction in Figure 3, perpendicular direction in Figure 4). The longitudinal direction is approximately parallel to the extension direction of the track R adjacent to platform P, and the length direction of the vehicle C on track R. "Width direction" refers to the width direction of platform P (up-down direction in Figure 3, left-right direction in Figure 4). The width direction is approximately parallel to the gauge direction of the track R adjacent to platform P, and the width direction of the vehicle C on track R. "Inside" or "inward" in the width direction refers to the side or direction approaching the center of width of platform P. "Outside" or "outward" in the width direction refers to the side or direction away from the center of width of platform P.

[0034] The track R includes a roadbed and a pair of rails placed on it. The train T travels along the rails on the track R. The train T consists of one vehicle C, or two or more vehicles C are connected sequentially via couplers. The platform P is set adjacent to the track R in the width direction, and its upper surface is located above the track R. The states of track R include the "absent state" shown in Figure 3(a) and the "occupied state" shown in Figure 3(b). In the absent state, there is no train T on track R, and the area above track R is wide open. In the absent state, railway users wait for the arrival of train T on the upper surface of platform P. In the occupied state, train T is stopped on track R. In the occupied state, railway users step over the gap D formed between platform P and train T to board train T from the upper surface of platform P, or to disembark from train T onto the upper surface of platform P.

[0035] (1) Configuration of fall detection system 1 The fall detection system 1 according to this embodiment is applied to a railway station equipped with a platform P and detects railway users who have accidentally fallen from the platform P. The detection targets include both "track fallers" who have fallen from the platform P onto the track R while the user is away, and "gap fallers" who have fallen into the gap D while the user is present on the track. Among gap fallers, some may fall all the way onto the track R, while others may become trapped in the gap D and be suspended in mid-air, but both are subject to detection.

[0036] The state of track R includes the "entering state," which is a transitional state from an unoccupied state to an occupied state, and the "exiting state," which is a transitional state from an occupied state to an unoccupied state. Furthermore, as shown in Figure 1, the fall detection system 1 according to this embodiment includes a scanning device 3 installed in the space below the platform P, a fall detection device 4 connected to the scanning device 3, a fall alarm 5 that issues an alarm according to the detection result of the fall detection device 4, and a fall location indicator 6 that displays the vehicle on which the fall event detected by the fall detection device 4 occurred.

[0037] As shown in Figures 2 to 4, the scanning device 3 is installed in the space below the platform P, along the track R adjacent to the platform P, and has a plurality of scanning units 3a, 3b, and 3c. Each scanning unit 3a, 3b, and 3c includes an upper scanner 31 and a lower scanner (optical scanner) 32 arranged in a pair. The plurality of scanning units 3a, 3b, and 3c are installed at multiple installation positions set at intervals in the longitudinal direction.

[0038] The upper and lower scanners 31 and 32 are, for example, 2D laser scanners that scan a laser beam within a predetermined angular range, and as shown in Figure 2, they include a light-emitting unit 33, a deflection unit 34, a light-receiving unit 35, a detection unit 36, and a distance calculation unit 37. The light-emitting unit 33 emits scanning light SL, such as laser light. The deflection unit 34 includes a deflector such as a galvanometer mirror and an actuator that rotates the deflector. The scanning light SL emitted from the light-emitting unit 33 is deflected and emitted by the deflection unit 34. This forms a two-dimensional scanning range SR. If an object is present within the scanning range SR, the scanning light SL is reflected by that object.

[0039] The light receiving unit 35 receives the reflected light of the scanning light SL. The detection unit 36 ​​detects whether or not an object exists within the scanning range SR (more precisely, the monitoring area set within the scanning range SR) based on the time from when the light is emitted from the light-emitting unit 33 until the reflected light is received by the light-receiving unit 35. The distance calculation unit 37 calculates the distance to the object that reflected the light, based on the time information from when the reflected light received by the light receiving unit 35 was received.

[0040] As shown in Figure 4, the upper and lower scanners 31 and 32 are installed below the platform P and above the track R. The upper scanner 31 is installed above the lower scanner 32. The lower scanner 32 primarily detects people who have fallen off the track and, when the vehicle is on the track, detects the wheels W of the vehicle C supported on the rails. On the other hand, the lower scanner 32 is required not to detect components of the track R. The vertical position of the lower scanner 32 is adjusted to an appropriate value for performing this role (for example, approximately 250 mm above the track).

[0041] The upper scanner 31 primarily detects people who have fallen through the gap D and are suspended in mid-air. The vertical position of the upper scanner 31 is adjusted to an appropriate value for this role (for example, approximately 700 mm above the track, or approximately 600 mm below the platform P). As will be described later, the upper scanner 31 according to this embodiment is used not only for its primary role of detecting people who have fallen through the gap, but also to accurately detect people who have fallen through the track when it is not present.

[0042] The upper and lower scanners 31 and 32 are installed further inward in the width direction than the track R, and even further inward in the width direction than the edge of the platform P. Below the platform P, a space may be formed that is set inward in the width direction when viewed from the edge of the platform P. The installation positions of the scanner units 3a, 3b, and 3c are set, for example, within such a space, and the vertical and widthwise positions of the upper and lower scanners 31 and 32 are adjusted within this space.

[0043] The upper and lower scanners 31 and 32 are installed in a position where the scanning light SL is emitted horizontally outward in the width direction and deflected in the horizontal plane. As a result, the upper and lower scanners 31 and 32 form two horizontal scanning ranges SR below the platform P and outward in the width direction as viewed from the upper and lower scanners 31 and 32. The deflection range of the scanning light SL is set to an angular range of approximately -5° to 185°, as shown in Figure 3. The scanning range SR is formed in a semicircular shape that is symmetrical with respect to the reference line RL extending in the width direction from the corresponding upper and lower scanners 31 and 32 in a plan view. Multiple scanning ranges SR are arranged in the longitudinal direction and partially overlap each other, and are formed by multiple scanner units 3a, 3b, and 3c.

[0044] A monitoring area is set within each scanning range SR. The upper and lower scanners 31 and 32 can each detect the presence or absence of an object (fallen person or wheel) within the scanning range SR. When the upper and lower scanners 31 and 32 detect the presence of an object within their respective set monitoring areas, they output a detection signal to that effect. The size and location of the monitoring area are changed according to the trajectory conditions. In the upper and lower scanners 31 and 32 according to this embodiment, it is possible to simultaneously set multiple monitoring areas in different areas within the same scanning range SR. It is also possible to partially overlap two or more simultaneously set monitoring areas within the same scanning range SR.

[0045] In this embodiment, the installation positions of the scanner units 3a, 3b, and 3c are set at intervals corresponding to the length of two vehicles in the longitudinal direction. Each installation position is located opposite the vehicle coupling section of the stationary train T in the width direction. The scanner unit 3a, located at the end of one longitudinal side (the left side of the page in Figure 3), faces the coupling section between the first and second cars. The scanner unit 3b, located next to it, is separated from the end scanner unit 3a by the length of two cars and faces the coupling section between the third and fourth cars.

[0046] The monitoring areas set for the upper and lower scanners 31 and 32, respectively, have a length equivalent to two vehicles in the longitudinal direction. The monitoring area corresponds to the area where two vehicles, one vehicle C on one side in the longitudinal direction and the other vehicle C on the other side (right side of the page in Figure 3), are stopped, as viewed from the installation position. The multiple monitoring areas set in the multiple scanner units 3a, 3b, and 3c are aligned longitudinally. Therefore, the entire area where train T stops is subject to monitoring.

[0047] However, the number of scanner units 3a, 3b, and 3c may be singular or plural, and is not particularly limited. The number of scanner units 3a, 3b, and 3c can be appropriately changed depending on the longitudinal dimensions of the monitoring area, the effective length of the platform P, and the maximum length of the train T that is assumed to stop on the track R. For example, scanner units 3a, 3b, and 3c may be installed only in curved sections where the gap between platform P and train T tends to widen, and fall detection may be performed in a portion of platform P.

[0048] As shown in Figure 2, the fall detection device 4 is connected to the scanning device 3 and detects a person who has fallen from platform P and the wheels W of a train T entering platform P based on the detection results output from the scanning device 3. When the fall detection device 4 detects a person who has fallen, it causes the fall alarm device 5 to activate in order to notify the relevant parties. Furthermore, the fall detection device 4 detects the wheels W of the train T entering platform P within a predetermined wheel detection area and determines the position (which car number) of the vehicle on which the scanner units 3a, 3b, and 3c are installed. The process from wheel detection by the fall detection device 4 to the identification of the vehicle number will be described in detail later.

[0049] The fall warning device 5 includes a warning device installed in a control center that manages train operations and emits a warning to the control center operator, or installed in a railway station and emits a warning to station staff, or installed in the control car of train T and emits a warning to the driver or conductor. The warning device may be a speaker or buzzer that outputs sound information such as a warning sound, a display that shows a warning message, or a lamp that emits a warning light.

[0050] Based on the warning, those involved can take measures to deal with the fall and prevent secondary accidents that may result from the fall. The fall location indicator 6 is installed, for example, in a station staff waiting area, on a platform P, in the driver's cab or conductor's room of a train T, and displays the location where the fall incident detected by the fall detection device 4 occurred as the vehicle number.

[0051] (2) Fall detection device 4 The fall detection device 4 is, for example, composed of a computer equipped with a CPU, memory such as ROM, RAM, and EEPROM, and an input / output interface. The computer may be a single unit or a combination of multiple physically distributed computers. The memory stores a fall detection program that causes such a computer to execute the fall detection method. The CPU reads the fall detection program stored in the memory and performs information processing related to the fall detection method according to the fall detection program. In addition to the fall detection program, the memory can also temporarily store information or data necessary for executing the fall detection method.

[0052] The fall detection device 4, as shown in Figure 2, has a control unit 10, a signal receiving unit 11, a track information acquisition unit 12, a monitoring area setting unit 13, a scanning data acquisition unit 14, a fall detection unit 15, a wheel detection area setting unit 16, a wheel detection unit 17, a wheel detection count detection unit 18, a vehicle number identification unit 19, a storage unit 20, and an output unit 21, which are configured by the CPU reading and executing a fall detection program.

[0053] The control unit 10 is connected to the signal receiving unit 11, track information acquisition unit 12, monitoring area setting unit 13, scanning data acquisition unit 14, fall detection unit 15, wheel detection area setting unit 16, wheel detection unit 17, wheel detection count detection unit 18, vehicle number identification unit 19, storage unit 20, and output unit 21, and controls each of these units. The signal receiving unit 11 receives data from the upper and lower scanners 31 and 32, such as the time until the light receiving unit 35 receives reflected light, the detection result detected by the detection unit 36, and the distance data to the object located within the fall detection area calculated by the distance calculation unit 37.

[0054] In particular, the signal receiving unit 11 detects the wheels W of train T in the wheel detection area described later and identifies the vehicle number (which car it is). To do this, it receives distance data to the object that reflected the light at predetermined angles, calculated based on information about the time it takes to receive the reflected light from the lower scanner 32 of the optical scanning units 3a, 3b, and 3c. The track information acquisition unit 12 acquires information indicating the state of the track. In this embodiment, the track information acquisition unit 12 acquires information indicating the state of the track, such as whether it is in a occupying state, an absent state, or neither of these two states (a transient state).

[0055] The monitoring area setting unit 13 selects one of a predetermined pattern of monitoring areas to be set on the upper and lower scanners 31 and 32 respectively, according to the acquired trajectory status, and sets the selected pattern of monitoring area on the upper and lower scanners 31 and 32 respectively. The monitoring area setting unit 13 switches the pattern of the monitoring area, the target of detection by the fall detection system 1, or the operating mode of the fall detection system 1 according to the trajectory status.

[0056] The scanning data acquisition unit 14 acquires the detection results regarding the presence or absence of objects within the monitoring area output by each scanning unit 3a to 3c (upper and lower scanning units 31 and 32). The fall detection unit 15 determines whether or not a person has fallen from the platform P, based on the detection results acquired by the scanning data acquisition unit 14, according to a predetermined determination logic. The wheel detection area setting unit 16 sets the length (L) in the direction of vehicle travel of the wheel detection area set at the stopping position of the rear wheel W of vehicle C to be sufficiently shorter than the length (X) between the position of the front wheel W and the position of the rear wheel W on the same vehicle C, and longer than the length (Y) between the center position of the rear wheel group of the front vehicle C and the center position of the front wheel group of the rear vehicle C between adjacent vehicle Cs (see X,Y in Figure 6 and L in Figure 8(a)).

[0057] This allows the wheel detection area to be set so that the front and rear wheels W of two adjacent vehicles C are recognized as a single group of wheels. By counting the number of times (n times) that wheels W pass through the wheel detection area, it is possible to determine which vehicle (nth vehicle) the vehicle C that has stopped in the wheel detection area belongs to. Furthermore, the wheel detection area setting unit 16 sets a detection time (T1) for detecting wheels W passing through the wheel detection area when train T enters platform P, and a detection time (T2) for detecting wheels that have stopped in the wheel detection area.

[0058] Here, the detection time (T1) (e.g., 100ms) is set to be shorter than the detection time (T2) (e.g., 5000ms). This allows for immediate detection of wheel W entering the wheel detection area with a short detection time when train T is passing through the wheel detection area, and for reliable detection of the stopped wheel W with a longer detection time when wheel W has stopped in the wheel detection area.

[0059] Furthermore, the wheel detection area setting unit 16 pre-sets a first value as the length L of the wheel detection area in the direction of vehicle travel for detecting the wheels W of a train T that has entered the platform P when the wheels W pass through the wheel detection area, and sets a second value different from the first value for detecting the wheels W when the wheels W of the train T have stopped in the wheel detection area. Here, the second value of the wheel detection area length L when detecting wheels W while train T is stopped (e.g., 2-3m) is smaller than the first value of the wheel detection area length L when train T is passing by (e.g., 5m).

[0060] As a result, the second value, which is pre-set to detect when the wheels W of train T are stopped in the wheel detection area, can detect wheels W even if it is within a smaller range than the first value, which is pre-set to detect wheels W passing through the wheel detection area. Therefore, even when the second value is set to a smaller value than the first value, the wheels W can be reliably detected.

[0061] Furthermore, the wheel detection area setting unit 16 sets the length D of the wheel detection area in the vehicle width direction of the train T vehicle C to a length slightly larger than the track width (spacing) (for example, 1.5 to 2.0 m). This allows for the detection of the presence or absence of wheels W by detecting the reflected light emitted from the lower scanner 32, and enables the appropriate detection of the wheels W on the near side and the wheels W on the far side from the perspective of the lower scanner 32 within the wheel detection area.

[0062] Furthermore, the size of the object detected within the wheel detection area is set to, for example, 100 mm. The wheel detection unit 17 detects the wheels W of the train T entering platform P within the wheel detection area set by the wheel detection area setting unit 16. Specifically, the wheel detection unit 17 detects the wheels W based on the data acquired by the scanning data acquisition unit 14.

[0063] The wheel detection count detection unit 18 counts the number of times the wheels W detected by the wheel detection unit 17 are detected when the train T enters the platform P. The vehicle number identification unit 19 determines that, given that the number of times the wheel W is detected by the wheel detection count detection unit 18 is n, the vehicle C that has stopped in the wheel detection area is the nth car of train T. The memory unit 20 stores various signals received by the signal receiving unit 11, track information acquired by the track information acquisition unit 12, monitoring areas set by the monitoring area setting unit 13, scanning data acquired by the scanning data acquisition unit 14, and the results of the fall detection by the fall detection unit 15. In particular, the memory unit 20 stores information related to the wheel detection area, such as information on the length and width dimensions of the wheel detection area in the direction of travel of the train T, information on the detection time of wheels in the wheel detection area, and information on the vehicle number identified according to the number of times wheels W detected in the wheel detection area are detected.

[0064] The output unit 21 outputs a command to the fall alarm 5 to activate an alarm when the fall detection unit 15 detects a person who has fallen. The output unit 21 also outputs and displays the vehicle number, which is determined according to the number of times the wheels W of the train T stopped in the wheel detection area (described later) have been detected, to the fall position indicator 6. In this embodiment, the fall detection device 4, with the configuration described above, can recognize which vehicle C the wheel W that has stopped in the wheel detection area belongs to. The fall detection device 4 then starts monitoring for people who have fallen in the fall detection area of ​​each lower scanner 32, and when a fall occurs, it can quickly notify the fall location indicator 6 or the like of the occurrence of the fall and which vehicle C the fall occurred in.

[0065] (monitoring area) In the fall detection device 4 of this embodiment, as shown in Figure 3, the following patterns of monitoring areas are set to be selectively set in the scanning device 3 according to the track conditions: a track monitoring area AR, which is set when the vehicle is absent, and a gap monitoring area AD, which is set when the vehicle is present.

[0066] As shown in Figure 3(a), the orbital monitoring area AR is set to cover the space above the orbit R. A person who falls off the orbit enters the orbital monitoring area AR during the process of falling. Therefore, based on the detection result that an object is present within the orbital monitoring area AR, it is possible to detect the person who fell off the orbit across the entire width of the orbit R. As shown in Figure 3(b), the gap monitoring area AD is narrower in width than the trajectory monitoring area AR and is set within the gap D. Those who fall through the gap and reach trajectory R enter the gap monitoring area AD during their fall. Those who become suspended in mid-air also remain within the gap monitoring area AD (particularly the upper gap monitoring area ADU). Therefore, based on the detection result that an object is present within the gap monitoring area AD, those who have fallen through the gap can be detected.

[0067] (Trajectory monitoring area) The scanning device 3 is composed of multiple scanning units 3a, 3b, and 3c arranged in the longitudinal direction. As shown in Figure 3(a), the orbital monitoring area AR is constructed by arranging orbital monitoring areas ARa, ARb, and ARc, which are set for each of the multiple scanner units 3a, 3b, and 3c, in a longitudinal direction.

[0068] Each scanning unit 3a, 3b, and 3c consists of an upper and lower scanning unit 31 and 32, which are provided as a pair, one above the other, as described above. The orbital monitoring area ARa of the scanner unit 3a consists of the upper orbital monitoring area ARaU, which is set on the upper scanner 31, and the lower orbital monitoring area ARaL, which is set on the lower scanner 32. The same applies to the orbital monitoring areas ARb and ARc of the other scanner units 3b and 3c.

[0069] In other words, multiple upper scanners 31 are arranged along the longitudinal direction, and multiple lower scanners 32 are arranged along the longitudinal direction. The orbital monitoring area AR of the scanning device 3 consists of an upper orbital monitoring area ARU and a lower orbital monitoring area ARL. The upper orbit monitoring area ARU is formed by arranging multiple upper orbit monitoring areas ARaU, ARbU, and ARcU, which are set on multiple upper scanners 31, in a longitudinal direction. The lower orbit monitoring area ARL is set at a lower position than the upper orbit monitoring area ARU and is formed by arranging multiple lower orbit monitoring areas ARaL, ARbL, and ARcL, which are set on multiple lower scanners 32, in a longitudinal direction.

[0070] Multiple upper orbit monitoring regions ARaU, ARbU, and ARcU are set up to be aligned along their longitudinal direction. Similarly, multiple lower orbit monitoring regions ARaL, ARbL, and ARcL are set up corresponding to each of the multiple upper orbit monitoring regions ARaU, ARbU, and ARcU. Each lower orbit monitoring region ARaL, ARbL, and ARcL is set up at approximately the same longitudinal position as its corresponding upper orbit monitoring region ARaU, ARbU, and ARcU.

[0071] Figure 5(a) is a perspective view showing the track monitoring area ARa set on the scanner unit 3a when the train T is absent. In this embodiment, the upper track monitoring area ARaU is formed by arranging multiple upper segments ARaU1, ARaU2, ARaU3, ARaU4 in the longitudinal direction within a single scanning range SR formed by the upper scanner 31 of the scanner unit 3a. The lower track monitoring area ARaL is also formed by arranging multiple lower segments ARaL1, ARaL2, ARaL3, ARaL4 in the longitudinal direction within a single scanning range SR formed by the lower scanner 32 of the scanner unit 3a.

[0072] In this way, multiple upper segments ARaU1, ARaU2, ARaU3, ARaU4 (multiple first monitoring areas) are configured to be aligned along the longitudinal direction. Furthermore, multiple lower segments ARaL1, ARaL2, ARaL3, ARaL4 (multiple second monitoring areas) are configured to correspond to each of the multiple upper segments ARaU1, ARaU2, ARaU3, ARaU4. Each lower segment ARaL1, ARaL2, ARaL3, ARaL4 is configured to be approximately in the same position along the longitudinal direction as its corresponding upper segment ARaU1, ARaU2, ARaU3, ARaU4.

[0073] The same applies to the other scanning units 3b and 3c. The number of segments can be set in any way, as long as the top and bottom segments are the same in a single scanning unit. In this embodiment, as just one example, the number of segments is 4 in each of the scanning units 3a, 3b, and 3c. As an example, the upper segments ARaU1, ARaU2, ARaU3, and ARaU4 are formed by dividing the upper track monitoring area ARaU into equal parts along its longitudinal direction. In this example, the upper track monitoring area ARaU is rectangular in shape with its longer side running along its longitudinal direction, and the length of its longer side is approximately the length of two vehicles (about 40m). The upper segments ARaU1, ARaU2, ARaU3, and ARaU4 are also roughly rectangular in shape when viewed from above, and the length of their longer sides is roughly equal to half the length of a vehicle. The same applies to the lower segments ARaL1, ARaL2, ARaL3, and ARaL4.

[0074] (Gap monitoring area) The gap monitoring area AD is similar to the orbit monitoring area AR. The gap monitoring area AD is constructed by arranging gap monitoring areas ADa, ADb, and ADc, which are set for each of the multiple scanner units 3a, 3b, and 3c, in a longitudinal direction. The gap monitoring area ADa of the scanner unit 3a consists of an upper gap monitoring area ADaU set on the upper scanner 31 and a lower gap monitoring area ADaL set on the lower scanner 32. The same applies to the gap monitoring areas ADb and ADc of the other scanner units 3b and 3c. The gap monitoring area AD of the scanning device 3 consists of an upper gap monitoring area ADU and a lower gap monitoring area ADL. The upper gap monitoring area ADU is formed by arranging multiple upper gap monitoring areas ADaU, ADbU, and ADcU set on multiple upper scanners 31 in the longitudinal direction. The lower gap monitoring area ADL is set at a lower position than the upper gap monitoring area ADU and is formed by arranging multiple lower gap monitoring areas ADaL, ADbL, and ADcL set on multiple lower scanners 32 in the longitudinal direction.

[0075] Multiple upper gap monitoring regions ADaU, ADbU, and ADcU are configured to be aligned along the longitudinal direction. Similarly, multiple lower gap monitoring regions ADaL, ADbL, and ADcL are configured to correspond to each of the multiple upper gap monitoring regions ADaU, ADbU, and ADcU. Each lower gap monitoring region ADaL, ADbL, and ADcL is configured to be approximately at the same longitudinal position as its corresponding upper gap monitoring region ADaU, ADbU, and ADcU.

[0076] Figure 5(b) is a perspective view showing the gap monitoring area ADa set on the scanner unit 3a when the train is on track. In this embodiment, as merely an example, the gap monitoring area ADa of the scanner unit 3a is not subdivided into multiple segments, unlike the track monitoring area ARa (see Figure 5(a)). However, multiple upper gap monitoring areas ADaU, ADbU, ADcU may partially overlap each other. Accordingly, multiple lower gap monitoring areas ADaL, ADbL, ADcL may also partially overlap each other.

[0077] (Wheel detection area) As described above, in this embodiment, the fall detection device 4 has a wheel detection area setting unit 16 that sets a wheel detection area as an area for detecting the front and rear wheels W of the vehicle C of the train T that has entered the platform P. Figure 6 shows the length X between the positions of the front wheels W and the rear wheels W on the same vehicle C of a train T entering a station platform P, and the length Y between the center position of the rear wheel group of the front vehicle C and the center position of the front wheel group of the rear vehicle C between two adjacent vehicles C.

[0078] Here, using Figure 7, we will explain the case where a 10-car train A and an 8-car train B are entering the station platform P, and the optical scanner (lower scanner 32) that detects the wheels W is installed in a position corresponding to the four rear cars C. As shown in Figure 7, the lower scanner 32 is installed in a position that covers the 7th and 8th cars and the 9th and 10th cars for train A, and in a position that covers the 5th and 6th cars and the 7th and 8th cars for train B.

[0079] In other words, although the lower scanner 32 is installed in a fixed position on the platform P, if the number of cars of the incoming train T changes, the target car C for fall detection changes depending on the detection area. Therefore, when the lower scanner 32 detects that a fall has occurred, it is not possible to immediately determine which car of the incoming train T the fall occurred in, which could lead to delays in the response of station staff and others.

[0080] Therefore, in the fall detection device 4 of this embodiment, the wheel detection area setting unit 16 sets the length L of the wheel detection area shown in Figure 8(a) in the direction of travel of the train T to be shorter than the length X shown in Figure 6 and longer than the length Y. In Figure 8(a), the origin of the vertical axis showing distance (m) is the position where the scanner units 3a, 3b, and 3c are installed; 0 to 1 on the vertical axis represents the fall detection area in the gap between platform P and train T; 1 to 4 on the vertical axis represents the area where train T may be present when train T is on the track (the area on the track); and 4 to 5 on the vertical axis represents the detection area where an object on the opposite platform P side was detected.

[0081] Similarly, the origin of the horizontal axis, which represents distance (m), is the position where the scanning units 3a, 3b, and 3c are installed, which is approximately directly in front of the connection point between the two vehicles C of train T when it is on the track. The negative side of the horizontal axis (left side in the figure) represents the left side of the scanning units 3a, 3b, and 3c, and the positive side (right side in the figure) represents the right side of the scanning units 3a, 3b, and 3c. Additionally, the range of 0.8 to 2.7 on the vertical axis and -5 to 0 on the horizontal axis indicates the wheel detection area.

[0082] As a result, the length L of the wheel detection area in the direction of travel (longitudinal direction) for detecting the wheels W is sufficiently shorter than the length X in Figure 6. Therefore, for example, it is possible to reliably detect when the front wheels W of the first vehicle have passed and count the passage of the first vehicle. Furthermore, since the length L of the wheel detection area is longer than the length Y in Figure 6, it is possible to avoid the detection of the rear wheel W of the first car and the front wheel W of the second car separately due to the gap between them, for example, and count them separately.

[0083] Therefore, as shown in Figure 8(a), in a wheel detection area with length L in the direction of travel of train T, as shown in Figure 8(b), the front wheel W of the first car C is detected and counted as the first car (N=1). Next, as shown in Figure 9(a), the rear wheels W of the first vehicle C and the front wheels W of the second vehicle C are detected together as a single wheel group and counted as the second vehicle (N=2). This allows for immediate recognition of the number of vehicles C that have passed through the wheel detection area and which vehicle C has stopped within the wheel detection area.

[0084] Next, as shown in Figure 9(b), the rear wheels W of the second vehicle C and the front wheels W of the third vehicle C are detected together as a single wheel group and counted as the third vehicle (N=3). Similarly, as shown in Figure 10(a), the rear wheels W of the 6th vehicle C and the front wheels W of the 7th vehicle C are detected together as a single wheel group and counted as the 7th vehicle (N=7).

[0085] Finally, as shown in Figure 10(b), if the train stops without passing the wheel W at the rear of the 7th car C in a wheel detection area with length L in the direction of travel of the train T, the detection count N is not incremented and remains at N=7. In the fall detection device 4 of this embodiment, as described above, a wheel detection area is set to detect the wheels W of a vehicle C so that it is possible to immediately recognize which vehicle C has stopped within the scanning range (fall detection area) of the lower scanner 32. The length L of the wheel detection area in the direction of travel of the train T is set to be sufficiently shorter than the length X between the position of the front wheel W and the position of the rear wheel W on the same vehicle C of the train T, and longer than the length Y between the center position of the rear wheel group of the front vehicle C and the center position of the front wheel group of the rear vehicle C between adjacent vehicles C, C.

[0086] As a result, the wheel detection area is set up to recognize the front and rear wheels W of two adjacent vehicles C, C as a single wheel group. Therefore, by counting the number of times wheels W are detected in the wheel detection area, it is easy to determine which vehicle C is located in. Thus, in the event of a fall, the vehicle number in which the fall occurred can be reported to the station staff waiting room or displayed on the fall position indicator 6, etc. As a result, while keeping implementation costs down, when a fall incident occurs, station staff can immediately recognize which car of train T the fall occurred in, allowing them to take appropriate action.

[0087] <Fall detection method> The fall detection device 4 of this embodiment, with the configuration described above, executes the fall detection method based on the flowchart shown in Figure 11.

[0088] In other words, the fall detection method of this embodiment is a fall detection method for detecting a person who has fallen from the platform of a target station onto the track through which a train passes. First, in step S11, the wheel detection area set at the stopping position of the rear wheel W of vehicle C of train T is set such that the length L in the direction of vehicle travel is shorter than the length X between the position of the front wheel W and the position of the rear wheel W on the same vehicle C, and longer than the length Y between the center position of the rear wheel group of the front vehicle C and the center position of the front wheel group of the rear vehicle C between adjacent vehicles C, C (wheel detection area setting step).

[0089] Next, in step S12, light is scanned within a predetermined angular range in the fall detection area set on the track, and reflected light data is acquired at predetermined angular intervals from an optical scanner that receives the reflected light (light data acquisition step). Next, in step S13, the wheels W of the train T entering platform P are detected in the wheel detection area set in the wheel detection area setting step, based on the light received data acquired in the light received data acquisition step (wheel detection step).

[0090] Next, in step S14, when train T enters platform P, the number of times the wheels W detected by the wheel detection step are counted is performed (wheel detection count detection step). Next, in step S15, assuming that the number of times the wheel W was detected in the wheel detection count detection step is n, it is determined that the vehicle C that stopped in the wheel detection area is the nth vehicle of train T (vehicle number identification step).

[0091] Next, in step S16, based on the light reception data acquired in the light reception data acquisition step, it is determined whether or not an object (person or object) has entered the orbit. If an object has been detected, the process proceeds to step S17 (object detection step). Next, in step S17, the result of detecting the intrusion of an object in the object detection step and the location where the object was detected, which is the nth car of train T, are reported (notification step).

[0092] As a result, by executing the fall detection method described above, the wheel detection area is set to recognize the front wheels W and rear wheels W of two adjacent vehicles C, C as a single wheel group. Therefore, by counting the number of times wheels W are detected in the wheel detection area, it is easy to determine which vehicle C is stopped in. Thus, in the event of a fall, the vehicle number in which the fall occurred can be reported to the station staff waiting room or displayed on the fall position indicator 6, etc. As a result, while keeping implementation costs down, when a fall incident occurs, station staff can immediately recognize which car of train T the fall occurred in, allowing them to take appropriate action.

[0093] [Other embodiments] Although one embodiment of the present invention has been described above, the present invention is not limited to the above embodiment, and various modifications are possible without departing from the spirit of the invention.

[0094] (A) In the above embodiments, examples of the present invention were described as a fall detection device and a fall detection method. However, the present invention is not limited thereto. For example, the present invention may be implemented as a fall detection program that causes a computer to execute the fall detection method described above.

[0095] This fall detection program is stored in the memory (storage unit) installed in the fall detection device, and the CPU reads the fall detection program stored in memory and causes the hardware to execute each step. More specifically, the CPU reads the fall detection program and executes the wheel detection area setting step, the light reception data acquisition step, the wheel detection step, the wheel detection count detection step, the vehicle number identification step, the object detection step, and the notification step described above, thereby achieving the same effect as above. Furthermore, the present invention may be implemented as a recording medium that stores a fall detection program.

[0096] (B) In the above embodiment, an example was given in which the detection time T1 for detecting wheels W passing through the wheel detection area when train T enters platform P is set to a different value from the detection time T2 for detecting wheels W stopped in the wheel detection area. However, the present invention is not limited thereto.

[0097] For example, you could set the detection times T1 and T2 to be the same. However, by setting the detection time T1 to a shorter time than T2, it is possible to detect the wheels as soon as they enter the wheel detection area with a short detection time when the train is passing through the wheel detection area, and to reliably detect the stopped wheels with a longer detection time when the wheels have stopped in the wheel detection area. Considering these effects, it is more preferable to use the settings as in the above embodiment.

[0098] (C) In the above embodiment, an example was given in which the length L of the wheel detection area in the direction of vehicle travel is set to a different value when the wheels W of the train T entering the platform P pass through the wheel detection area and when the wheels W of the train T come to a stop in the wheel detection area. However, the present invention is not limited to this. For example, the length L may be set to a constant value.

[0099] (D) In the above embodiment, an example was given in which the wheels W of the train T's vehicle C in the wheel detection area are detected by receiving reflected light from the scanning light of the lower scanner 32. However, the present invention is not limited thereto.

[0100] For example, the detection of train car wheels may be performed using a dedicated device for wheel detection other than an optical scanner. However, the configuration described in the above embodiment is preferable because it simplifies the system configuration by using the optical scanner that detects falls to also detect wheels.

[0101] (E) In the above embodiment, an example was described in which the storage unit 20, which stores the vehicle number and the like identified by the vehicle number identification unit 19, is provided inside the fall detection device 4. However, the present invention is not limited thereto. For example, a storage device that stores information such as the vehicle number may be located outside the fall detection device. [Industrial applicability]

[0102] The fall detection device of the present invention has the effect of suppressing installation costs while immediately allowing station staff to recognize which train car the fall occurred in, enabling them to take appropriate action, and is therefore widely applicable to fall detection systems installed at railway stations. [Explanation of symbols]

[0103] 1. Fall detection system 3 Scanning device 3a, 3b, 3c Scanner Unit 4. Fall detection device 5. Fall alarm 6. Fall position indicator 10 Control Unit 11 Signal receiving section 12 Orbit information acquisition section 13 Monitoring area setting section 14. Scanning data acquisition unit (light reception data acquisition unit) 15. Fall detection unit (object detection unit) 16 Wheel detection area setting unit 17 Wheel detection unit 18 Wheel detection count detection unit 19. Vehicle Number Identification Section 20 Memory section 21 Output section 31 Upper scanner 32. Lower Scanner (Optical Scanner) 33 Light-emitting part 34 Polarizing part 35 Light receiving part 36 Detection unit 37 Distance Calculation Unit AD gap monitoring area AR trajectory monitoring area C Vehicle D Gap L Length P Platform R orbit SR scanning range T train T1, T2 detection time W wheels X length Y length

Claims

1. A fall detection device that detects a person who has fallen from the platform of a target station onto the tracks through which a train passes, A light receiving data acquisition unit acquires light receiving data of reflected light from an optical scanner that scans light within a predetermined angular range in a fall detection area set on the aforementioned track and receives the reflected light, at predetermined angular intervals. Based on the light-receiving data acquired by the light-receiving data acquisition unit, an object detection unit detects the intrusion of an object into the trajectory, A wheel detection area setting unit sets the length (L) in the direction of vehicle travel of the wheel detection area set at the stopping position of the rear wheels of the aforementioned train vehicles so that it is shorter than the length (X) between the front wheel position and the rear wheel position on the same vehicle, and longer than the length (Y) between the center position of the rear wheel group of the front vehicle and the center position of the front wheel group of the rear vehicle in adjacent vehicles. A wheel detection unit detects the wheels of the train that has entered the platform within the wheel detection area set by the wheel detection area setting unit, When the aforementioned train enters the platform, a wheel detection count detection unit counts the number of times the wheels detected by the wheel detection unit are detected, If the number of times the wheel is detected by the wheel detection count detection unit is n, the vehicle number identification unit determines that the vehicle stopped in the wheel detection area is the nth car of the train. A fall detection device equipped with this feature.

2. The wheel detection area setting unit sets a detection time (T1) for detecting the wheels as they pass through the wheel detection area when the train enters the platform, and a detection time (T2) for detecting the wheels when they are stopped in the wheel detection area. The fall detection device according to claim 1.

3. The detection time (T1) is set to be shorter than the detection time (T2). The fall detection device according to claim 2.

4. The wheel detection unit detects the wheel based on the light-receiving data acquired by the light-receiving data acquisition unit. A fall detection device according to any one of claims 1 to 3.

5. The wheel detection area setting unit pre-sets a first value as the length (L) of the wheel detection area in the direction of vehicle travel for detecting the wheels when the wheels of the train entering the platform pass through the wheel detection area, and pre-sets a second value different from the first value for detecting the wheels when the wheels of the train stop in the wheel detection area. The fall detection device according to claim 1 or 2.

6. The second value is smaller than the first value. The fall detection device according to claim 5.

7. The wheel detection area setting unit sets the length (D) of the wheel detection area in the vehicle width direction of the train to a length slightly greater than the vehicle width. The fall detection device according to claim 1 or 2.

8. The system further includes a storage unit for storing the determination result in the vehicle number identification unit. The fall detection device according to claim 1 or 2.

9. A fall detection device according to claim 1 or 2, An optical scanner that scans light within a predetermined angular range in a fall detection area set on the aforementioned track and receives reflected light, A fall detection system equipped with this feature.

10. The aforementioned optical scanner is installed near the position where the connecting section of two adjacent vehicles stops. The object detection unit performs fall detection on two vehicles. The fall detection system according to claim 9.

11. The object detection unit further includes a fall location indicator that shows the location where an object was detected or the location where a fall occurred. The fall detection system according to claim 9.

12. A fall detection method for detecting a person who has fallen from the platform of a target station onto the tracks through which a train passes, A wheel detection area setting step involves setting the length (L) in the direction of vehicle travel of the wheel detection area set at the stopping position of the rear wheels of the train's vehicles to be shorter than the length (X) between the front wheel position and the rear wheel position on the same vehicle, and longer than the length (Y) between the center position of the rear wheel group of the front vehicle and the center position of the front wheel group of the rear vehicle in adjacent vehicles. A light receiving data acquisition step involves acquiring light receiving data of the reflected light from an optical scanner that scans light within a predetermined angular range in a fall detection area set on the aforementioned track and receives the reflected light, at predetermined angular intervals. A wheel detection step in which, in the wheel detection area set by the wheel detection area setting step, the wheels of the train that has entered the platform are detected based on the light received data acquired in the light received data acquisition step, When the aforementioned train enters the platform, a wheel detection count detection step counts the number of times the wheels detected by the wheel detection step are counted, If the number of times the wheels are detected in the wheel detection count detection step is n, then the vehicle number identification step determines that the vehicle stopped in the wheel detection area is the nth car of the train. An object detection step which detects the intrusion of an object into the trajectory based on the light reception data acquired in the light reception data acquisition step, The object detection step includes a notification step that notifies the result of detecting the intrusion of the object and the position where the object was detected, which is the nth car of the train. A fall detection method equipped with the following features.

13. A fall detection program that detects a person who has fallen from the platform of a target station onto the tracks through which a train passes, A wheel detection area setting step involves setting the length (L) in the direction of vehicle travel of the wheel detection area set at the stopping position of the rear wheels of the train's vehicles to be shorter than the length (X) between the front wheel position and the rear wheel position on the same vehicle, and longer than the length (Y) between the center position of the rear wheel group of the front vehicle and the center position of the front wheel group of the rear vehicle in adjacent vehicles. A light receiving data acquisition step involves acquiring light receiving data of the reflected light from an optical scanner that scans light within a predetermined angular range in a fall detection area set on the aforementioned track and receives the reflected light, at predetermined angular intervals. A wheel detection step in which, in the wheel detection area set by the wheel detection area setting step, the wheels of the train that has entered the platform are detected based on the light received data acquired in the light received data acquisition step, When the aforementioned train enters the platform, a wheel detection count detection step counts the number of times the wheels detected by the wheel detection step are counted, If the number of times the wheels are detected in the wheel detection count detection step is n, then the vehicle number identification step determines that the vehicle stopped in the wheel detection area is the nth car of the train. An object detection step which detects the intrusion of an object into the trajectory based on the light reception data acquired in the light reception data acquisition step, The object detection step includes a notification step that notifies the result of detecting the intrusion of the object and the position where the object was detected, which is the nth car of the train. A fall detection program that causes a computer to execute a fall detection method that includes the following features.

Images