Train control system and obstacle detection device

Abstract

Train control system (1), with: an image information conversion device (22) mounted on a train (10) for outputting initial image information obtained by adding train information to image information acquired by an image acquisition device (21) capable of capturing an area in front of the train (10); and an obstacle detection device (100) which is installed on the ground and includes: an image information storage unit (103) for recording the first image information output from the image information conversion device (22); an image information comparison unit (101) for comparing the first image information with a second image information recorded in the image information storage unit (103); and an image information result output unit (104) for outputting a result from the image information comparison unit (101), wherein the image information comparison unit (101) determines a degree of similarity between the first image information and the second image information, and, if the degree of similarity is greater than or equal to a second threshold and less than a first threshold, performs obstacle detection by comparing the first image information with a third image information recorded in the image information storage unit (103).

Description

Area

[0001] The present invention relates to a train control system that performs obstacle detection. background

[0002] A technique for detecting an obstacle on a train track is known by capturing an image of an area in front of the train with a camera mounted on the train and comparing the captured image with an image recorded in the past, for example see Japanese patent application publication no. JP 2010 - 63260 A.

[0003] Publication JP 2016-88325A1 describes a monitoring unit that generates monitoring information about a position ahead of a vehicle's direction of travel based on wave motions incident from the front in the direction of travel. A distance determination unit determines the distance from the vehicle to a specific position based on position information of the vehicle and position information of the specific position. Using a monitoring control unit, the monitoring unit can extract information about the specific position based on the distance specified by the distance indication unit. Brief description of the technical problem

[0004] However, a conventional obstacle detection system stores the captured image in a storage device installed in the train, which has presented a problem in that the storage device's capacity for recording the images has increased.

[0005] The present invention was made to solve the aforementioned problem and it is an object of the present invention to reduce the capacity of the storage device mounted on a train to perform obstacle detection. Solution to the problem

[0006] A train control system according to a first aspect of the invention comprises: an image information conversion device mounted on a train to output first image information obtained by adding train information to image information acquired by an image acquisition device capable of detecting an area in front of the train; and an obstacle detection device installed on the ground, comprising: an image information storage unit for recording the first image information output from the image information conversion device; an image information comparison unit for comparing the first image information with second image information recorded in the image information storage unit; and an image information result output unit for outputting a result from the image information comparison unit.

[0007] An obstacle detection device according to a second aspect of the invention comprises: an image information storage unit installed on the ground which records first image information; an image information comparison unit which determines a degree of similarity of image information by comparing the first image information with second image information recorded in the image information storage unit, and performing a second obstacle detection pass if the degree of similarity is greater than or equal to a second threshold and less than a first threshold; and an image information result output unit which outputs a result from the image information comparison unit. Advantageous effects of the invention

[0008] The train control system according to the present invention comprises: the image information conversion device, which is mounted on the train and outputs the first image information obtained by adding the train information to the image information acquired by the image acquisition device, which is mounted on the train and is capable of detecting the area in front of the train; and the obstacle detection device, which is installed on the ground and comprises the image information storage unit, which records the first image information output from the image information conversion device, and the image information comparison unit, which compares the first image information with the second image information recorded in the image information storage unit, and the image information result output unit, which outputs the result from the image information comparison unit.Therefore, the train control system according to the present invention can reduce the capacity of the storage device mounted on the train for performing obstacle detection.

[0009] Furthermore, the obstacle detection device according to the present invention comprises: the image information storage unit, which is installed on the ground and records the first image information; the image information comparison unit, which determines the degree of similarity of the image information by comparing the first image information with the second image information recorded in the image information storage unit, and performs the second obstacle detection pass if the degree of similarity is greater than or equal to the second threshold and less than the first threshold; and the image information result output unit, which outputs the result from the image information comparison unit. Therefore, the obstacle detection device according to the present invention can reduce the capacity of the storage device mounted on the train for obstacle detection. Brief description of drawings Fig. Figure 1 shows a diagram representing a schematic configuration of a train control system according to a first embodiment. Fig. Figure 2 shows a diagram illustrating a schematic configuration of a carriage of the train control system according to the first embodiment. Fig. Figure 3 shows a diagram illustrating a configuration of an obstacle detection device of the train control system according to the first embodiment. Fig. Figure 4 shows a flowchart illustrating the operation of the obstacle detection device of the train control system according to the first embodiment. Fig. Figure 5 shows a diagram illustrating an example of on-board image information from the train control system according to the first embodiment. Fig. Figure 6 shows a diagram that provides another example of the on-board image information of the train control system according to the first embodiment. Fig. Figure 7 shows a diagram that provides yet another example of the on-board image information of the train control system according to the first embodiment. Fig. Figure 8 shows a flowchart illustrating the operation of an operational management unit of the train control system according to the first embodiment. Fig. Figure 9 shows a flowchart illustrating the operation of a ground control device of the train control system according to the first embodiment. Fig. Figure 10 shows a diagram illustrating a configuration of an obstacle detection device of the train control system according to a second embodiment. Fig. Figure 11 shows a flowchart illustrating the operation of the obstacle detection device of the train control system according to the second embodiment. Fig. Figure 12 shows a diagram illustrating a configuration of an obstacle detection device of the train control system according to a third embodiment. Fig. Figure 13 shows a flowchart illustrating the operation of the obstacle detection device of the train control system according to the third embodiment. Fig. Figure 14 shows a diagram illustrating an example of on-board image information from the train control system according to the third embodiment. Fig. Figure 15 shows a table that presents a history of the on-board image information of the train control system according to the third embodiment. Fig. Figure 16 shows a diagram illustrating a configuration of an obstacle detection device of the train control system according to a fourth embodiment. Fig. Figure 17 shows a flowchart illustrating the operation of the obstacle detection device of the train control system according to the fourth embodiment. Fig. Figure 18 shows a diagram illustrating an example of a general hardware configuration representing the train control system according to each of the first to fourth embodiments. Description of embodiments: First embodiment

[0010] Fig. Figure 1 shows a diagram illustrating a schematic configuration of a train control system 1 according to a first embodiment of the present invention. The train control system 1, which is described in Fig. Figure 1 shows a train 10, an operations management unit 11, ground control units 13 connected to the operations management unit 11 via a train control system network 12, wireless base stations 15 each connected to a corresponding ground control unit 13 via a corresponding base network 14, and an obstacle detection unit 100.

[0011] In the train control system 1, a rail 17 is divided into a plurality of control areas and the ground control devices 13 manage the train 10 in the control areas.

[0012] The operations management unit 11 monitors the operational status of the overall train control system 1, its settings, etc. For example, the operations management unit 11 sets a temporary speed limit.

[0013] The train control system network 12 is a network across the entire train control section and connects the operations management facility 11, the obstacle detection facility 100 and the ground control facilities 13.

[0014] A ground control unit 13 is provided for each area. The ground control units 13 each receive train location information from the train 10 via the corresponding wireless base stations 15 and the base networks 14, determine the location of each train 10 from the received train location information, and transmit the information to the operations management unit 11 via the train control system network 12. Furthermore, the ground control units 13 each receive operations management information from the train 10 from the operations management unit 11 and transmit the received operations management information to the train 10 via the corresponding base networks 14.

[0015] The ground control units 13 each calculate a movement authorization restriction (a position corresponding to a restriction at which the train 10 must stop) of the train 10 in the area based on the train location information and transmit opening information from a locking control unit (not shown). Information about the movement authorization restriction that has been calculated is transmitted to the train 10 via the corresponding wireless base stations 15.

[0016] Here, the information about the restriction of movement authorization is based on the location of the preceding train 10 and the position at which train 10 is to stop. This information is then passed to the following train 10, indicating the position at which train 10 can safely stop. For example, if there is an obstacle or similar obstruction on the track 17 that train 10 is traveling on, the position of the obstacle is set as the position at which train 10 must stop.

[0017] The base networks 14 are each a network provided for each area and connect corresponding ground control equipment 13 and wireless base stations 15.

[0018] Rail 17 is a structure on the track bed on which train 10 travels. The operation of train 10 is controlled by train control system 1, which is located in Fig. Figure 1 is shown. On rail 17, a plurality of ground units 16 or position correction ground units 16, to which IDs (ground unit information) are assigned, are installed at the intervals.

[0019] The wireless base stations 15 each communicate with an on-board wireless device 24, which is described below, of the train 10. Fig. Two of the wireless base stations 15 are installed for one of the ground control units 13, but the number is not limited to two. One, three, or more of the wireless base stations 15 can be installed for one of the ground control units 13.

[0020] The obstacle detection device 100 is connected to the operations management device 11 and the train control system network 12. The obstacle detection device 100 detects an obstacle on the track 17 on which the train 10 is traveling. Details are described below.

[0021] Fig. Figure 2 shows a diagram representing a configuration of a car of train 10 traveling in the train control system 1 according to the first embodiment. Note that although Fig. 2 only represents the equipment necessary for describing the train control system 1 according to the first embodiment; other equipment and functions may be installed.

[0022] The in Fig. The two depicted carriages demonstrate the on-board wireless device 24, an on-board antenna 25, an on-board unit 27, an on-board control unit 26, a speedometer generator 28, an image acquisition device 21, an image information conversion device 22, and an on-board information input / output device 23. The on-board control unit 26 comprises a train control unit 30, a ground unit information receiving unit 31, and a storage unit 32.

[0023] Storage unit 32 stores an on-board database (not shown). The on-board database stores information (track information) about the track 17 on which the train 10 travels. The track 17 information includes details about the kilometer spacing at which each of the ground units 16 is installed (ground unit location information).

[0024] The onboard unit 27 detects the ID of each of the ground units 16 when it passes over them (when the onboard unit 27 and the ground unit 16 overlap). The ID is received by the ground unit information receiver 31 as the ground unit information. The ground unit information is then output by the ground unit information receiver 31 to the train control unit 30.

[0025] The tachometer generator 28 generates a pulse corresponding to the rotational speed of a wheel, calculates the distance traveled and a speed of the train 10 and outputs the calculated distance traveled and speed to the train control unit 30.

[0026] The train control unit 30 calculates the location of train 10 based on the train's travel distance, output by the speedometer generator 28, and the ground unit location information stored in the memory unit 32. The calculated location of train 10 is treated as the train location information. This information includes kilometer markers, entry / exit line information, track number information in station facilities, and similar data. Furthermore, the speed of train 10, output by the speedometer generator 28, is treated as the train speed information. The train control unit 30 is connected to the on-board information input / output device 23 and outputs the train location information, train speed information, and similar data to the on-board information input / output device 23.

[0027] The image acquisition device 21 is mounted on the first car and can capture an image in an area in front of the train 10. Specifically, the image in the area in front of the train 10 is captured at regular intervals and output to the image information conversion device 22. A camera, a video camera, or similar device is suitable as the image acquisition device 21. The number of image acquisition devices 21 can be one, two, or more. The image acquisition device 21 captures the image at each predetermined kilometer or at each predetermined time, which are fixed in advance.Furthermore, the angle of image capture of the image acquisition device 21 is adjusted in relation to the head position of the train 10, so that the center of the image has a position at which the train 10 can be stopped, even if a certain extra distance is added when the train 10 is traveling at the maximum speed that is intended for the route, for example.

[0028] The image information conversion unit 22 is connected to the image acquisition unit 21 and can receive image information from the image acquisition unit 21, which is captured by the image acquisition unit 21. The image information conversion unit 22 is also connected to the on-board control unit 26 and can receive train information such as a train ID, train location information, train speed information, and train acceleration information from the on-board control unit 26. The image information conversion unit 22 generates new image information by adding the train information received from the on-board control unit 26 and the date and time of image capture to the image information received from the image acquisition unit 21. Here, the new image information generated by the image information conversion unit 22 is referred to as on-board image information.The image information conversion unit 22 is connected to the on-board information input / output unit 23 and outputs the generated on-board image information to the on-board information input / output unit 23.

[0029] The on-board information input / output unit 23 is connected to the on-board control unit 26 and the image information conversion unit 22. The on-board information input / output unit 23 can receive train information output from the on-board control unit 26 and similar devices. The on-board information input / output unit 23 can also receive on-board image information output from the image information conversion unit 22. The on-board information input / output unit 23 outputs the received train information and on-board image information to the on-board wireless device 24.

[0030] The onboard wireless device 24 communicates with the wireless base station 15 according to the current location of the train 10. Specifically, information output from the onboard information input / output device 23 is transmitted to the corresponding wireless base station 15 via the onboard antenna 25. Furthermore, the onboard wireless device 24 receives information output from a corresponding ground control device 13 via the corresponding wireless base station 15 using the onboard antenna 25 and outputs the information to the onboard information input / output device 23.

[0031] Fig. Figure 3 shows a diagram illustrating a configuration of the obstacle detection device 100 of the train control system 1 according to the first embodiment. Note that although Fig. 3 only represents the devices necessary to describe the train control system 1 according to the first embodiment; other devices and functions may be installed.

[0032] The obstacle detection device 100, which is in Fig. Figure 3 shows an input unit 102, an image information storage unit 103, an image information comparison unit 101, and an image information result output unit 104.

[0033] The input unit 102 is connected to the train control system network 12 and can receive the on-board image information output from the train 10.

[0034] The image information storage unit 103 is connected to the input unit 102 and the image information comparison unit 101. The image information storage unit 103 stores and accumulates the onboard image information received by the input unit 102.

[0035] The image information comparison unit 101 is connected to the input unit 102 and the image information storage unit 103. The image information comparison unit 101 can receive the on-board image information from the input unit 102. The image information comparison unit 101 can also receive past on-board image information from the image information storage unit 103. The image information comparison unit 101 compares the on-board image information received from the input unit 102 with the past on-board image information received from the image information storage unit 103 and makes a determination regarding obstacle detection.

[0036] Here, the on-board image information received by input unit 102 is referred to as on-board image information A, and the past on-board image information received by image information storage unit 103 is referred to as on-board image information B. On-board image information A and on-board image information B share the same train location information. This means that on-board image information A and on-board image information B are image information captured at the same train location. Furthermore, on-board image information B can be image information captured by train 10, which generated on-board image information A, or it can be image information captured by a preceding train traveling ahead of train 10, which generated on-board image information A. In other words, the train ID information added to on-board image information A and B can be the same or different.

[0037] The image information comparison unit 101 compares the on-board image information A with the on-board image information B and makes a determination regarding obstacle detection. This determination ascertains whether or not there is an obstacle in front of the train that affects the train's journey. The image information comparison unit 101 is connected to the image information result output unit 104. If an obstacle has been detected, the image information comparison unit 101 outputs a result indicating the obstacle detection to the image information result output unit 104. Fig. 3. The image information comparison unit 101 is configured to receive the on-board image information A from the input unit 102, but can also receive the on-board image information A from the image information storage unit 103. If no obstacle has been detected, the image information comparison unit 101 can output a result to the image information result output unit 104 indicating that no obstacle has been detected.

[0038] The image information output unit 104 is connected to the image information comparison unit 101 and receives a result indicating an obstacle detection when the image information comparison unit 101 has detected an obstacle. The image information output unit 104 also outputs the result indicating the obstacle detection to the operations management unit 11. The image information output unit 104 and the operations management unit 11 can be connected directly or via the train control system network 12.

[0039] Next, an operation of an obstacle detection is described, which is carried out by the obstacle detection device 100, which is used in the train control system 1 according to the first embodiment.

[0040] Fig. Figure 4 shows a flowchart illustrating the operation of an obstacle detection system performed by the obstacle detection device 100, which is used in the train control system 1 according to the first embodiment. First, the input unit 102 receives the on-board image information A-output from the train 10 (S101). Next, the on-board image information A-output from the input unit 102 and the on-board image information B-output from the image information storage unit 103 are input to the image information comparison unit 101 (S102). The image information comparison unit 101 compares the on-board image information A with the on-board image information B and makes a determination regarding obstacle detection (S103). If the image information comparison unit 101 has not detected an obstacle (No in S104), the obstacle detection process ends.If the image information comparison unit 101 has detected an obstacle (Yes in S104), a result indicating the obstacle detection would be output to the image information result output unit 104 (S105).

[0041] Next, the determination of an obstacle detection is described, which is made by the obstacle detection device 100, which is used in the train control system 1 according to the first embodiment.

[0042] Fig. Figure 5 shows a diagram illustrating the determination via an obstacle detection made by the obstacle detection device 100, which is used in the train control system 1 according to the first embodiment. Fig. Figure 5 shows a diagram representing a case where the obstacle detection device 100 does not detect an obstacle. Fig. 5(a) corresponds to the on-board image information A input to the image information comparison unit 101. Fig. 5(b) corresponds to the on-board image information B input to the image information comparison unit 101.

[0043] The image information comparison unit 101 divides the on-board image information into a plurality of areas in order to compare the on-board image information. Fig. Figure 6 shows a diagram representing on-board image information divided into multiple areas. Fig. Figure 6(a) shows a diagram that provides an example in which the on-board image information A input to the image information comparison unit 101 is divided into the plurality of areas. Fig. Figure 6(b) shows a diagram illustrating an example where the on-board image information B input to the image information comparison unit 101 is divided into the plurality of areas. Fig. The onboard image information is divided into ten parts in each of the vertical and horizontal directions, resulting in a total of 100 sections. The number of divided sections is set as appropriate.

[0044] Among the shared areas in Fig. The area that is first from the left and first from the bottom is designated (1, 1). The area that is second from the left and first from the bottom is designated (2, 1). Similarly, each of the divided areas is designated differently.

[0045] The image information comparison unit 101 compares (1, 1) of onboard image information A with (1, 1) of onboard image information B. Similarly, the 100 shared areas are compared. If a predetermined percentage or more of the image information matches as a result of the shared area comparison, it is determined that no obstacle is present. In other words, onboard image information A is compared with onboard image information B, and if the degree of similarity between them is greater than or equal to a predetermined percentage, it is determined that no obstacle is present. If the degree of similarity is less than a predetermined percentage, it is determined that an obstacle has been detected.In other words, it is determined that no obstacle exists if the degree of agreement is greater than or equal to a predetermined first threshold, or that an obstacle has been detected if the degree of agreement is less than the first threshold.

[0046] Fig. Figure 7 shows a diagram illustrating a case where the obstacle detection device 100 detects an obstacle. Fig. 7(a) corresponds to the on-board image information A input to the image information comparison unit 101. Fig. 7(b) corresponds to the on-board image information B input to the image information comparison unit 101. Here, the degree of agreement to determine that an obstacle has been detected is set to 98% or higher.

[0047] As described above, the image information comparison unit 101 compares (1, 1) of onboard image information A with (1, 1) of onboard image information B. Similarly, the 100 shared areas are compared. As a result of comparing the shared areas, the onboard image information does not match in seven areas, namely (3, 4), (4, 4), ..., and (9, 4). In this case, the degree of agreement is 93%, which is less than the 98% degree of agreement required to determine that an obstacle has been detected. Therefore, the image information comparison unit 101 determines that "an obstacle has been detected" and outputs the result to the image information result output unit 104. The image information result output unit 104 outputs the result to the operations management unit 11.

[0048] As described above, the determination of obstacle detection by the obstacle detection device 100, which is used in the train control system 1 according to the first embodiment, is carried out by the image information comparison unit 101. This unit divides each of the on-board image information A and the on-board image information B into a plurality of areas and compares the areas between the on-board image information A and B. If the degree of agreement is greater than or equal to the predetermined percentage as a result of the comparison between on-board image information A and on-board image information B, it is determined that no obstacle is present. If the degree of agreement is less than the predetermined percentage, it is determined that an obstacle has been detected, and this is output to the image information result output unit 104.The image information output unit 104 outputs the result that "an obstacle has been detected" to the operations management unit 11.

[0049] As described above, in the train control system 1 according to the first embodiment, the obstacle detection device 100, which is installed on the ground, makes the determination via obstacle detection using the on-board image information acquired by the train 10 and the past on-board image information. Therefore, the capacity of the storage device mounted on the train 10 can be reduced.

[0050] Next, the operation of the operations management device 11 is described when the obstacle detection device 100 has detected an obstacle. Fig. Figure 8 shows a flowchart illustrating the operation of the operations management device 11, which is used in the train control system 1 according to the first embodiment.

[0051] First, it is detected whether or not a result indicating obstacle detection has been received from obstacle detection device 100 (S201). Processing is terminated if the result indicating obstacle detection was not received from obstacle detection device 100 (No in S201). If the result indicating obstacle detection was received from obstacle detection device 100 (Yes in S201), a protected area is defined for train 10 for which the obstacle has been detected (S202). Here, the protected area refers to an area in which train 10 can safely stop or an area in which train 10 can safely proceed. For example, the protected area for train 10 is defined directly in front of the detected obstacle in the direction of travel.Next, the protected area information regarding the protected area that has been established is transmitted to a corresponding ground control device 13 via the train control system network 12 (S203).

[0052] The protected area can be set for train 10 for which the obstacle has been detected and for a subsequent train traveling behind train 10 for which the obstacle has been detected.

[0053] The safety of train 10, which is traveling on track 17, can be ensured by the operational management device 11, which detects that the obstacle has been detected.

[0054] Next, the operation of the ground control device 13 is described when the obstacle detection device 100 has detected an obstacle. Fig. Figure 9 shows a flowchart illustrating the operation of the ground control device 13, which is used in the train control system 1 according to the first embodiment.

[0055] First, it is detected whether or not the protected area information has been received from the operations management unit 11 (S301). Processing is terminated if the protected area information has not been received from the operations management unit 11 (No in S301). If the protected area information has been received from the operations management unit 11 (Yes in S301), a protection command information is generated (S302). Here, the protection command information refers to information that is transmitted to train 10 and enables train 10 to stop or proceed safely. The protection command information includes, for example, protected position information that specifies a position enabling train 10 to stop safely. Next, the generated protection command information is transmitted to train 10 via a corresponding one of the base networks 14 (S303).Here, although the protection command information is transmitted wirelessly to train 10, as in . Fig. As shown in Figure 1, the protection command information that has been generated can be transmitted from the ground control unit 13 to the train 10 via the ground unit(s) 16.

[0056] The protection order information can be transmitted to train 10 for which the obstacle has been detected and to a following train that is behind train 10 for which the obstacle has been detected.

[0057] The on-board control unit 26 of train 10, which has received the protection command information, generates a speed check pattern for train 10 to brake and stop at the position where the train can stop safely, based on the protected position information contained in the protection command information. Train 10 brakes according to the speed check pattern and stops before the obstacle in the direction of travel.

[0058] The train control system 1 according to the first embodiment comprises: the image information conversion device 22, which is mounted on the train 10 and provides first image information obtained by adding the train information to the image information acquired by the image acquisition device 21, which is mounted on the train 10 and is capable of detecting the area in front of the train 10; and the obstacle detection device 100, which is installed on the ground and comprises the image information storage unit 103, which records the first image information output from the image information conversion device 22, the image information comparison unit 101, which compares the first image information with a second image information recorded in the image information storage unit 203, and the image information result output unit 104, which outputs a result from the image information comparison unit 101.Therefore, according to the first embodiment, the train control system 1 can reduce the capacity of the storage device mounted on the train 10 to perform obstacle detection.

[0059] Furthermore, in the train control system 1 according to the first embodiment, the image information result output unit 104 outputs a result indicating an obstacle detection when an obstacle is in front of the train 10, thereby reducing the capacity of the storage device mounted on the train 10 for performing an obstacle detection.

[0060] Furthermore, in the train control system 1 according to the first embodiment, the image information comparison unit 101 determines the degree of agreement of the image information, thereby reducing the capacity of the storage device mounted on the train 10 for carrying out obstacle detection.

[0061] Furthermore, in the train control system 1 according to the first embodiment, the image information comparison unit 101 determines that an obstacle has been detected if the degree of agreement is less than the first threshold, thereby reducing the capacity of the storage device mounted on the train 10 for carrying out obstacle detection.

[0062] Furthermore, according to the first embodiment, the train control system 1 includes the operations management unit 11, which manages the operation of the train 10 and sets the protected area for the train 10 when the result has been received from the image information output unit 104 that an obstacle has been detected, thereby reducing the capacity of the storage device mounted on the train for carrying out the obstacle detection. Second embodiment

[0063] A second embodiment is characterized by an obstacle detection device that compares the on-board image information multiple times. Fig. Figure 10 shows a diagram illustrating a configuration of an obstacle detection device 200 of the train control system according to the second embodiment. Note that, although Fig. 10 only represents the equipment necessary to describe the train control system according to the second embodiment; other equipment and functions may be installed. The basic configuration of the Fig. 10 is similar to the Fig. 3. The image information comparison unit 101 in Fig. 3 compares the on-board image information once, using an image information comparison unit 201 in Fig. 10. The on-board image information compares multiple times or several times.

[0064] The operation of an obstacle detection system, which is carried out by the obstacle detection device 200 used in the train control system according to the second embodiment, is described. Fig. Figure 11 shows a flowchart illustrating the operation of the obstacle detection performed by the obstacle detection device 200, which is used in the train control system according to the second embodiment. First, the input unit 102 receives the on-board image information A-output from the train 10 (S401). Next, the image information comparison unit 201 compares the on-board image information A with on-board image information C and makes a determination regarding obstacle detection (S402).

[0065] The on-board image information C is past on-board image information received by the image information storage unit 103. On-board image information A and on-board image information C share the same train location information. This means that on-board image information A and on-board image information C are image information captured at the same train location. Furthermore, on-board image information C can be image information captured by train 10, which generated on-board image information A, or it can be image information captured by a preceding train traveling ahead of train 10, which generated on-board image information A. In other words, the train ID information added to on-board image information A and C can be the same or different.In the following description, the past image information captured by the preceding train, which is ahead of train 10 that generated the on-board image information A, is referred to as the on-board image information C.

[0066] The image information comparison unit 201 compares the on-board image information A with the on-board image information C and makes a determination about obstacle detection based on the degree of agreement. Specifically, on-board image information A is compared with on-board image information C, and it is determined whether the degree of agreement between them is greater than or equal to a predetermined first threshold. If the degree of agreement is greater than or equal to the first threshold (Yes in S403), obstacle detection is terminated with the determination that no obstacle is present. If the degree of agreement is less than the first threshold (No in S403), it is determined whether the degree of agreement is greater than or equal to a predetermined second threshold (S404). The second threshold is a value lower than the first threshold.If the degree of agreement is less than the second threshold (No in S404), it is determined that an obstacle has been detected and a result indicating obstacle detection is output to the image information result output unit 104 (S407). The processing after the result indicating obstacle detection has been output to the image information result output unit 104 is similar to that of the first embodiment. If the degree of agreement is greater than or equal to the second threshold (Yes in S404), the on-board image information A is compared with on-board image information D and a determination of obstacle detection is made (S405).

[0067] The on-board image information D is past on-board image information received by the image information storage unit 103 and is different from on-board image information C. On-board image information A and on-board image information D share the same train location information. This means that on-board image information A and on-board image information D are image information captured at the same train location. Furthermore, on-board image information D can be image information captured by train 10, which generated on-board image information A, or it can be image information captured by a preceding train traveling ahead of train 10, which generated on-board image information A. In the following description, the image information captured by train 10, which generated on-board image information A, is referred to as on-board image information D.

[0068] The image information comparison unit 201 compares the on-board image information A with the on-board image information D and determines whether the degree of agreement between them is greater than or equal to a predetermined third threshold. If the degree of agreement is greater than or equal to the third threshold (Yes in S406), obstacle detection is terminated by determining that no obstacle is present. The third threshold is a value greater than or equal to the second threshold. If the degree of agreement is less than the third threshold (No in S406), it is determined that an obstacle has been detected, and a result indicating obstacle detection is output to the image information result output unit 104 (S407).

[0069] The train control system according to the second embodiment performs obstacle detection at multiple points in time, utilizing different image data that contain the same train position information. Although the preceding description presented an example where obstacle detection was performed twice, it can be performed three times or more. Performing the determination multiple times reduces false detections and thus enables more accurate obstacle detection.

[0070] The train control system according to the second embodiment uses the on-board image information with the different train ID information in the first determination during obstacle detection and in the second and subsequent determinations of an obstacle detection, thereby making it possible to eliminate false detections due to the installation angle, since the installation angle of the image acquisition device varies depending on the train. False detections due to a difference in the installation angle of the image acquisition device depending on the train can be eliminated.

[0071] In the train control system according to the second embodiment, the image information comparison unit 201 performs obstacle detection by comparing the first image information with a third image information recorded in the image information storage unit 103. The detection occurs when the degree of similarity is greater than or equal to the second threshold and less than the first threshold. This reduces the capacity of the storage device mounted on the train 10 for performing obstacle detection. Furthermore, performing the determination multiple times reduces false detections and thus enables more accurate obstacle detection.

[0072] Furthermore, in the train control system according to the second embodiment, the image information comparison unit 201 determines that an obstacle has been detected if the degree of agreement is less than the third threshold, thereby reducing the capacity of the storage device mounted on the train 10 for performing obstacle detection. The multiple times the determination is performed also reduce false detections and thus enable a more accurate determination of obstacle detection.

[0073] Furthermore, the obstacle detection device 200 according to the second embodiment comprises: the image information storage unit 103, which is installed on the ground and records the first image information; the image information comparison unit 201, which determines the degree of similarity of the image information by comparing the first image information with a second image information recorded in the image information storage unit 103 and performs a second obstacle detection pass if the degree of similarity is greater than or equal to the second threshold and less than the first threshold; and the image information result output unit 104, which outputs the result from the image information comparison unit 201. Therefore, the obstacle detection device 200 according to the second embodiment can reduce the capacity of the storage device mounted on the train 10 for performing obstacle detection.Furthermore, performing the determination multiple times reduces the risk of false detection, thereby enabling a more accurate determination of obstacle detection.

[0074] Furthermore, in the obstacle detection device 200 according to the second embodiment, the image information comparison unit 201 performs obstacle detection by comparing the first image information with the third image information recorded in the image information storage unit 103. This reduces the capacity of the storage device mounted on the train 10 for performing obstacle detection. It also reduces the number of false detections performed multiple times, thus enabling more accurate obstacle detection. Third embodiment

[0075] A third embodiment is characterized by an obstacle detection device 300 which uses on-board image information which has various train location information, wherein the on-board image information is captured by the train 10 for which a determination via an obstacle detection is made. Fig. Figure 12 shows a diagram illustrating a configuration of the obstacle detection device 300 of the train control system according to the third embodiment. Note that although Fig. 12 only represents the equipment necessary to describe the train control system according to the third embodiment; other equipment and functions may be fitted.

[0076] The obstacle detection device 300, which is in Fig. As shown in Figure 12, the input unit 102, the image information storage unit 103, a comparison result storage unit 305, an image information comparison unit 301, and the image information result output unit 104 are all present. The comparison result storage unit 305 is connected to the image information comparison unit 301 and stores the result of a comparison previously performed by the image information comparison unit 301. The result of the past comparison includes a record of the degree of agreement of the past comparison and similar information. Here, the comparison result storage unit 305 is configured separately, but the image information storage unit 103 can also contain the comparison result storage unit 305.

[0077] The operation of an obstacle detection system, which is carried out by the obstacle detection device 300 used in the train control system according to the third embodiment, is described. Fig. Figure 13 shows a flowchart illustrating the operation of the obstacle detection performed by the obstacle detection device 300, which is used in the train control system according to the third embodiment. First, the input unit 102 receives the on-board image information A-output from the train 10 (S501). Next, the on-board image information A is compared with on-board image information E in the image information comparison unit 301, which then makes a determination regarding obstacle detection (S502).

[0078] The on-board image information E is past on-board image information received by the image information storage unit 103. On-board image information A and on-board image information E share the same train location information. This means that on-board image information A and on-board image information E are image information captured at the same train location. Furthermore, on-board image information E can be image information captured by train 10, which generated on-board image information A, or it can be image information captured by a preceding train traveling ahead of train 10, which generated on-board image information A. In other words, the train ID information associated with on-board image information A and E can be the same or different. In the following description, the past image information captured by train 10, which generated on-board image information A, is referred to as on-board image information E.

[0079] The image information comparison unit 301 compares the on-board image information A with the on-board image information E and makes a determination about obstacle detection based on the degree of agreement. Specifically, on-board image information A is compared with on-board image information E, and it is determined whether the degree of agreement between them is greater than or equal to a predetermined first threshold. If the degree of agreement is greater than or equal to the first threshold (Yes in S503), obstacle detection is terminated by determining that no obstacle is present. If the degree of agreement is less than the first threshold (No in S503), it is determined whether the degree of agreement is greater than or equal to a predetermined second threshold (S504). The second threshold is a value lower than the first threshold.If the degree of agreement is less than the second threshold (No in S504), it is determined that an obstacle has been detected and a result indicating obstacle detection is output to the image information result output unit 104 (S507). The processing after the result indicating obstacle detection has been output to the image information result output unit 104 is similar to that of the first embodiment. If the degree of agreement is less than the first threshold but greater than or equal to the second threshold (Yes in S504), the image information comparison unit 301 makes a determination of obstacle detection according to a transition of the degree of agreement (S505).

[0080] Image information comparison unit 301 refers to compared comparison results stored in comparison result storage unit 305. Specifically, image information comparison unit 301 refers to the results of an image information comparison for on-board image information A1, which was generated immediately before on-board image information A, and for on-board image information A2, which was generated before on-board image information A and A1. For example, the results of the degree of agreement are referred to as the results of the image information comparison. The train ID information added to on-board image information A, A1, and A2 is the same.

[0081] For example, it is assumed that the results of the image information comparison (degree of agreement) for the onboard image information A, A1, and A2 are 94%, 95%, and 96%, respectively. In other words, it is assumed that the degree of agreement for onboard image information A2, which was compared before onboard image information A and A1, is 96%, where the degree of agreement for onboard image information A1, which was compared immediately before onboard image information A, is 95%, and the degree of agreement for onboard image information A, which is now being compared, is 94%. In this case, the degree of agreement as a value decreases over time.

[0082] The image information comparison unit 301 refers to past comparison results stored in the comparison result storage unit 305 and determines that an obstacle has been detected if the degree of agreement decreases in value over time (Yes in S506), thereby sending a result indicating obstacle detection to the image information result output unit 104 (S507). The image information comparison unit 301 refers to past comparison results stored in the comparison result storage unit 305 and determines that no obstacle is present if the degree of agreement does not decrease in value over time (No in S506), thereby ending obstacle detection.

[0083] Fig. Figure 14 shows a diagram to illustrate obstacle detection by the train control system according to the third embodiment. Fig. 14(a) shows a diagram representing the on-board image information A. Fig. 14(b) shows a diagram representing the on-board image information A1. Fig. 14(c) shows a diagram representing the on-board image information A2.

[0084] The image information comparison unit 301 performs an image information comparison on the on-board image information A and determines, as a result, that the on-board image information does not match (3, 4), (4, 4), ..., and (8, 4) in six areas, and that the degree of match is 94%. Similarly, the image information comparison unit 301 performs an image information comparison on the on-board image information A1 and determines, as a result, that the on-board image information does not match (3, 6), (4, 6), ..., and (7, 6) in five areas, and that the degree of match is 95%. Similarly, the image information comparison unit 301 performs an image information comparison on the on-board image information A2 and determines, as a result, that the on-board image information does not match (4, 8), (5, 8), (6, 8), and (7, 8) in four areas, and that the degree of match is 96%.

[0085] Fig. Figure 15 shows a table that presents a history of the on-board image information of the train control system according to the third embodiment. Fig. 15 presents an example of the time at which the on-board image information A, A1 and A2 is captured, and an example of the train location at which each of the on-board image information A, A1 and A2 is captured, and an example of the degree of match for each of the on-board image information A, A1 and A2. As from the Fig. As can be seen in section 15, the on-board information components A, A1, and A2 are recorded at different times and at different train locations. The on-board information components are also located in... Fig. 15 in the order of the on-board image information A2, A1 and A.

[0086] As from Fig. Since train 10 can be seen as 14, the distance between train 10 and an obstacle decreases over time. Because the distance between train 10 and the obstacle decreases, the number of areas of the obstacle's image information increases relative to all the image information areas. Comparing the on-board image information A2 and A1 in Fig. 14. The number of areas containing the obstacle's image information, relative to all image information areas, increases in onboard image information A1. Furthermore, in a comparison between onboard image information A1 and A, the number of areas containing the obstacle's image information, relative to all image information areas, is greater in onboard image information A. Over time, the number of areas containing the obstacle's image information, relative to all image information areas, increases. This means that the degree of similarity, compared to the case where no obstacle is present, decreases over time.

[0087] The third embodiment performs obstacle detection using onboard image information that includes the same train ID information but different train location information. Specifically, the obstacle is detected by performing obstacle detection by comparing the results of the image information generated by train 10, which has the same train ID information but different train locations. Using image information that includes the same train ID information allows for a comparison based on image information obtained from the same image acquisition device, thus enabling more accurate obstacle detection.

[0088] In the preceding description, the determination is made using the transition of three degrees of agreement, representing the degree of agreement before the current and previous degrees of agreement up to the current degree of agreement. However, transitions of two, four, or more degrees of agreement can also be used. To reduce misdeterminations, it is preferable to use transitions of three or more degrees of agreement. Regarding the decrease in the value of the degree of agreement over time, it is not necessary for the value to decrease continuously, but only for it to decrease overall.

[0089] In the train control system according to the third embodiment, the image information comparison unit 301 performs obstacle detection based on the transition of the degree of agreement when the degree of agreement is greater than or equal to the second threshold and less than the first threshold, thereby reducing the capacity of the storage device mounted on the train 10 for performing obstacle detection. The comparison can also be made based on image information from the same image acquisition device, thus enabling more accurate obstacle detection.

[0090] Furthermore, in the train control system according to the third embodiment, the image information comparison unit 301 determines that an obstacle has been detected when the degree of agreement decreases over time, thereby reducing the capacity of the storage device on which the train 10 is mounted for performing obstacle detection. The comparison can also be made based on image information from the same image acquisition device, thus enabling more accurate obstacle detection. Fourth embodiment

[0091] A fourth embodiment is characterized in that, when comparing the on-board image information, an obstacle detection device compares train speed information and train acceleration information that are added to the on-board image information.

[0092] Fig. Figure 16 shows a diagram illustrating a configuration of the obstacle detection device 400 of the train control system according to the fourth embodiment. Note that although Fig. 16 only represents the equipment necessary to describe the train control system according to the fourth embodiment; other equipment and functions may be fitted.

[0093] The obstacle detection device 400, which is in Fig. As shown in Figure 16, the input unit 102, the image information storage unit 103, an image information comparison unit 401, and the image information result output unit 104 are included. The basic configuration of Fig. 16 is similar to that of the Fig. 3.

[0094] The operation of an obstacle detection system, which is carried out by the obstacle detection device 400 used in the train control system according to the fourth embodiment, is described. Fig. Figure 17 shows a flowchart illustrating the operation of the obstacle detection performed by the obstacle detection device 400, which is used in the train control system according to the fourth embodiment. First, the input unit 102 receives the on-board image information A-output from the train 10 (S601). Next, the on-board image information A is compared with on-board image information F in the image information comparison unit 401, which then makes a determination regarding obstacle detection (S602).

[0095] The on-board image information F is past on-board image information received by the image information storage unit 103. On-board image information A and on-board image information F contain the same train ID and train location information.

[0096] The image information comparison unit 401 compares the on-board image information A with the on-board image information F and determines whether obstacle detection is possible based on the degree of similarity. Specifically, it compares on-board image information A with on-board image information F and determines whether the degree of similarity between them is greater than or equal to a predetermined first threshold. If the degree of similarity is greater than or equal to the first threshold (Yes in S603), obstacle detection is terminated by determining that no obstacle is present. If the degree of similarity is less than the first threshold (No in S603), it determines whether the degree of similarity is greater than or equal to a predetermined second threshold (S604). The second threshold is a value lower than the first threshold.If the degree of agreement is less than the second threshold (No in S604), it is determined that an obstacle has been detected, and a result indicating obstacle detection is output to the image information result output unit 104 (S607). The processing after the result indicating obstacle detection has been output to the image information result output unit 104 is similar to that of the first embodiment. If the degree of agreement is less than the first threshold but greater than or equal to the second threshold (Yes in S604), a determination of obstacle detection is made using the train speed information and the train acceleration information added to the on-board image information A and the on-board image information F (S605).

[0097] The image information comparison unit 401 compares the train speed information between on-board image information A and on-board image information F and calculates a difference. Similarly, the image information comparison unit 401 compares the train acceleration information between on-board image information A and on-board image information F and calculates a difference. Here, the difference value calculated for the train speed information and the difference value calculated for the train acceleration information are designated D1 and D2, respectively.

[0098] The image information comparison unit 401 determines whether the difference value D1 in the train speed information is less than or equal to a predetermined fourth threshold. Similarly, the image information comparison unit 401 determines whether the difference value D2 in the train acceleration information is less than or equal to a predetermined fifth threshold. If the difference value D1 in the train speed information is less than or equal to the fourth threshold and the difference value D2 in the train acceleration information is less than or equal to the fifth threshold (Yes in S606), it is determined that an obstacle has been detected, and a result indicating the obstacle detection is output to the image information result output unit 104 (S607).If the difference value D1 in the train speed information is greater than the fourth threshold (No in S606), or if the difference value D2 in the train acceleration information is greater than the fifth threshold (No in S606), it is determined that there is no obstacle and the obstacle detection is terminated.

[0099] If the difference value D1 in the train speed information is less than or equal to the fourth threshold and the difference value D2 in the train acceleration information is less than or equal to the fifth threshold, then there is not much difference between the train speed and train acceleration values ​​at the time the on-board image information A and the on-board image information F are generated. This means that there is not much difference in the image acquisition conditions with respect to train speed and train acceleration. This implies that the parts of the image information with the relatively small differences in image acquisition conditions are being compared.Therefore, although the degree of agreement is greater than or equal to the second threshold, the image information comparison unit 401 determines that an obstacle has been detected because the degree of agreement is less than the first threshold, which is the threshold for determining an obstacle detection.

[0100] The train control system according to the fourth embodiment captures the train speed information and the train acceleration information that is added to the on-board image information and can therefore take into account the difference in the image capture conditions for the on-board image information and improve the reliability of the result of the comparison of the on-board image information.

[0101] In the train control system according to the fourth embodiment, when the degree of agreement is greater than or equal to the second threshold and less than the first threshold, the image information comparison unit 401 performs obstacle detection by comparing the train speed information between the first and second image information, as well as the train acceleration information between the first and second image information. This reduces the capacity of the storage device mounted on the train 10 for performing obstacle detection. Furthermore, the difference in the image acquisition conditions for the on-board image information can be taken into account, thus improving the reliability of the on-board image information comparison result.

[0102] Note that in the first to fourth embodiments described above, the image information conversion device 22 and the obstacle detection device 100 (200, 300, 400) each have at least one processor 1003, a memory circuit, a receiver 1004 and a transmitter 1001, and the operation of each device can be implemented by software. Fig. Figure 18 shows a diagram illustrating an example of a general hardware configuration representing each of the image information conversion devices 22 and the obstacle detection device 100 (200, 300, 400) of the train control system according to each of the first to fourth embodiments. A Fig.The device shown in Figure 18 comprises a processor 1003, a memory 1002, a receiver 1004, and a transmitter 1001. The processor 1003 performs calculations and control using software and received data. The memory 1002 stores received data or data required when the processor 1003 performs calculations and control, and also stores the software. The receiver 1004 is an interface that receives a signal or information to be sent to the image information conversion device 22 or the obstacle detection device 100 (200, 300, 400). The transmitter 1001 is an interface that transmits a signal or information output from the image information conversion device 22 or to the obstacle detection device 100 (200, 300, 400). Note that a plurality of each of the processors 1003, the memory 1002, the receiver 1004 and the transmitter 1001 can be provided.

[0103] Furthermore, the embodiments can be suitably combined. For example, a combination of the second and third embodiments can perform the determination twice using different parts of image information with the same train location information, and the determination can be performed based on the transition of the degree of agreement for the image information generated by train 10, which has the same train ID information but different train locations.

[0104] Note that the present invention is not limited to the foregoing embodiments and the embodiments may be modified or omitted as appropriate without departing from the scope of the concept of the present invention. Reference symbol list

[0105] 1 Train control system; 10 Train; 11 Operations management unit; 12 Train control system network; 13 Ground control unit; 14 Base network; 15 Wireless base station; 16 Ground unit; 17 Rail; 21 Image acquisition unit; 22 Image information conversion unit; 23 On-board information input / output unit; 24 On-board wireless unit; 25 On-board antenna; 26 On-board control unit; 27 On-board unit; 28 Speedometer generator; 30 Train control unit; 31 Ground unit information receiving unit; 32 Storage unit; 100, 200, 300, 400 Obstacle detection unit; 101, 201, 301, 401 Image information comparison unit; 102 Input unit; 103 Image information storage unit; 104 Image information result output unit; 305 Comparison result storage unit; 1001 Transducer; 1002 Memory; 1003 Processor; 1004 Receiver.

Claims

[1] Train control system (1), comprising: an image information conversion device (22) mounted on a train (10) for outputting initial image information obtained by adding train information to image information acquired by an image acquisition device (21) capable of capturing an area in front of the train (10); and an obstacle detection device (100) which is installed on the ground and includes: an image information storage unit (103) for recording the first image information output from the image information conversion device (22); an image information comparison unit (101) for comparing the first image information with a second image information recorded in the image information storage unit (103); and an image information result output unit (104) for outputting a result from the image information comparison unit (101), wherein the image information comparison unit (101) determines a degree of similarity between the first image information and the second image information, and, if the degree of similarity is greater than or equal to a second threshold and less than a first threshold, performs obstacle detection by comparing the first image information with a third image information recorded in the image information storage unit (103). [2] Train control system (1) according to claim 1, wherein the image information comparison unit (22) determines that an obstacle is detected when the degree of agreement is less than a third threshold. [3] Train control system (1), with: an image information conversion device (22) mounted on a train (10) for outputting initial image information obtained by adding train information to image information acquired by an image acquisition device (21) capable of capturing an area in front of the train (10); and an obstacle detection device (200) which is installed on the ground and includes: an image information storage unit (103) for recording the first image information output from the image information conversion device (22); an image information comparison unit (201) for comparing the first image information with a second image information recorded in the image information storage unit (103); and an image information result output unit (104) for outputting a result from the image information comparison unit (201), wherein the image information comparison unit (201) determines a degree of similarity between the first image information and the second image information, and, if the degree of similarity is greater than or equal to a second threshold and less than a first threshold, performs obstacle detection based on a transition of the degree of similarity. [4] Train control system (1) according to claim 3, wherein the image information comparison unit (201) determines that an obstacle is detected when the degree of agreement decreases over time. [5] Train control system (1), with: an image information conversion device (22) mounted on a train (10) for outputting initial image information obtained by adding train information to image information acquired by an image acquisition device (21) capable of capturing an area in front of the train (10); and an obstacle detection device (300) which is installed on the ground and includes: an image information storage unit (103) for recording the first image information output from the image information conversion device (22); an image information comparison unit (301) for comparing the first image information with a second image information recorded in the image information storage unit (103); and an image information result output unit (104) for outputting a result from the image information comparison unit (301), wherein the image information comparison unit (301) determines a degree of similarity between the first image information and the second image information, and, if the degree of similarity is greater than or equal to a second threshold and less than a first threshold, performs obstacle detection by comparing train velocity information between the first image information and the second image information, and by comparing train acceleration information between the first image information and the second image information. [6] Train control system (1) according to any one of claims 1 to 5, further comprising an operations management facility (11) for carrying out operations management on the train (10), wherein The operations management unit (11) establishes a protected area for the train when a result is received indicating an obstacle detection output by the image information result output unit (104). [7] Obstacle detection device (100), with: an image information storage unit (103) installed on the ground to record initial image information; an image information comparison unit (101) for determining a degree of agreement of image information by comparing the first image information with a second image information recorded in the image information storage unit (103), and performing a second pass of obstacle detection if the degree of agreement is greater than or equal to a second threshold and less than a first threshold; and an image information result output unit (104) for outputting a result from the image information comparison unit (101). [8] Obstacle detection device (100) according to claim 7, wherein the image information comparison unit (101) performs obstacle detection by comparing the first image information with a third image information recorded in the image information storage unit (103). [9] Obstacle detection device (100) according to claim 7 or 8, wherein the image information comparison unit (101) performs obstacle detection based on a transition of the degree of agreement. [10] Obstacle detection device (100) according to one of claims 7 to 9, wherein the image information comparison unit (101) performs obstacle detection by comparing train speed information between the first image information and the second image information, and by comparing train acceleration information between the first image information and the second image information.

Images