Train type recognition system and method

Abstract

This disclosure relates to a train type recognition system and method, wherein the train type recognition system comprises a first imaging device, which is arranged at a first side of a traveling path of the train and is configured to capture an image toward the traveling path; a first detection device, which is in signal connection with the first imaging device and is configured to detect whether the image captured by the first imaging device contains at least part of an image of the train, and record arrival or departure time of the train when detecting that the image contains at least part of an image of the train; a train speed acquisition device, which acquires a train speed; an adjustment device, which is in signal connection with the train speed acquisition device and adjusts a line rate of the first imaging device according to the train speed; and a recognition device, which is in signal connection with the first imaging device and recognizes a train type according to the image captured by the first imaging device.

Description

Related Application

[0001] This application is based on the Chinese patent application with application number 202411833094.0, filed on December 12, 2024, and titled "Train Type Recognition System and Method", and from which priority is claimed. The disclosure of this Chinese patent application is hereby incorporated herein by reference in its entirety.Technical Field

[0002] This disclosure relates to the field of train type recognition technology, and in particular a train type recognition system and method.Background Art

[0003] At present, train type recognition mostly adopts an axle distance recognition method. The principle of this method is: wheel sensors mounted on the inner side of the track are first used to detect train wheel axles and calculate the axle distance, then the current axle distance information is used to match the train type in a database, and finally the carriage type of the current carriage is determined. In addition, according to the axle distance information, the coupler position can be judged, and the carriages can be divided.

[0004] However, in some countries and regions, the type of the train chassis does not match the type of the carriages. Using wheel sensors to recognize the train type can only recognize the type of the train chassis, but not the type of the carriages. Moreover, wheel sensors need to be installed on the inner side of the track, which has high installation requirements and requires on-track operations, resulting in greater construction difficulty.

[0005] It should be noted that the information disclosed in the Background Art of this disclosure is merely intended to increase the understanding of the overall background of this disclosure, and should not be regarded as admitting or implying in any form that the information constitutes existing technology known to those skilled in the art. The above statement is only used to provide background technical information related to the present application and does not necessarily constitute prior art.Content of Invention

[0006] The embodiments of this disclosure provide a train type recognition system and method, which can reduce the construction difficulty.

[0007] According to an aspect of this disclosure, a train type recognition system is provided, comprising: a first imaging device, which is arranged at a first side of a traveling path of the train and is configured to capture an image toward the traveling path, a first detection device, which is in signal connection with the first imaging device and is configured to detect whether the image captured by the first imaging device contains at least part of an image of the train, and record arrival or departure time of the train when detecting that the image contains at least part of an image of the train, a train speed acquisition device, which is configured to acquire a train speed, an adjustment device, which is in signal connection with the train speed acquisition device and is configured to adjust a line rate of the first imaging device according to the train speed, and a recognition device, which is in signal connection with the first imaging device and is configured to recognize a train type according to the image captured by the first imaging device.

[0008] In some embodiments, the train type recognition system further comprises an image processing device, which is in signal connection with the first imaging device, the train comprising a plurality of carriages, the image processing device being configured to segment the image captured by the first imaging device to form independent images of the plurality of carriages.

[0009] In some embodiments, the train speed acquisition device comprises: a second imaging device, which is arranged at the first side of the traveling path and has a preset distance from the first imaging device, the second imaging device being configured to capture an image toward the traveling path, a second detection device, which is in signal connection with the second imaging device and is configured to detect whether the image captured by the second imaging device contains at least part of an image of the train and record arrival or departure time of the train when detecting that the image contains at least part of an image of the train, and a calculation device, which is in signal connection with the first detection device and the second detection device and is configured to calculate the train speed according to the magnitude of the preset distance and the time difference between the arrival time recorded by the first detection device and the arrival time recorded by the second detection device.

[0010] In some embodiments, the train type recognition system further comprises a first reference object, which is arranged at a second side of the traveling path opposite to the first side and is configured to make the image captured by the first imaging device contain the first reference object when the train has not arrived; and / or a second reference object, which is arranged at a second side of the traveling path opposite to the first side and is configured to make the image captured by the second imaging device contain the second reference object when the train has not arrived.

[0011] In some embodiments, the distance between the first imaging device and the traveling path is equal to the distance between the second imaging device and the traveling path.

[0012] In some embodiments, the height of the first imaging device from the ground is equal to the height of the second imaging device from the ground.

[0013] In some embodiments, the train type recognition system further comprises a first support rod, on which the first imaging device is mounted; and / or a second support rod, on which the second imaging device is mounted.

[0014] In some embodiments, the train type recognition system further comprises two first lighting lamps, both of which are mounted on the first support rod and are respectively located above and below the first imaging device; and / or two second lighting lamps, both of which are mounted on the first support rod and are respectively located above and below the first imaging device.

[0015] According to another aspect of this disclosure, a method for recognizing a train type based on the above-mentioned train type recognition system is provided, comprising the following steps: capturing an image toward the traveling path by means of the first imaging device, detecting whether the image captured by the first imaging device contains at least part of an image of the train, if yes, recording arrival or departure time of the train, acquiring the train speed, adjusting the line rate of the first imaging device according to the train speed; and recognizing a train type according to the image captured by the first imaging device.

[0016] In some embodiments, the method for recognizing a train type further comprises: setting the initial line rate of the first imaging device as a first line rate F0, adjusting the line rate of the first imaging device according to the train speed after detecting the arrival of the train, and restoring the line rate of the first imaging device to the first line rate F0 after detecting the departure of the train.

[0017] In some embodiments, the train comprises a plurality of carriages, and the method for recognizing a train type further comprises: segmenting the image captured by the first imaging device to form independent images of the plurality of carriages, and then recognizing the type of each carriage based on the independent image.

[0018] In some embodiments, the method for recognizing a train type further comprises: providing a first reference object on the opposite side of the first imaging device, and making the image captured by the first imaging device contain the first reference object when the train has not arrived, and determining whether the proportion of the first reference object in the image captured by the first imaging device decreases, and if yes, it is determined that the train arrives.

[0019] Based on the above technical solution, in the embodiments of this disclosure, when recognizing the train type, the arrival or departure of the train is detected in such a way that images are captured by the first imaging device and there is no need to install sensors on the train running track or wheels, which can overcome problems such as high installation requirements, need for online operation, and large construction difficulty, etc. due to the installation of sensors, thereby effectively reducing the construction difficulty, the amount of construction, and the construction cost.Brief Description of the Drawings

[0020] The drawings described herein are used to provide a further understanding of this disclosure and constitute a part of the present application. The exemplar embodiments of this disclosure and their description are construed to explain this disclosure, and do not constitute improper limitations on this disclosure. In the drawings: FIG. 1 is a structural schematic view of some embodiments of the train type recognition system provided by this disclosure. FIG. 2 is a top view of some embodiments of the train type recognition system provided by this disclosure. FIG. 3 is a view of principle of some embodiments of the train type recognition system provided by this disclosure. FIG. 4 is a flowchart of some embodiments of the train type recognition method provided by this disclosure.

[0021] In the figures: 10, traveling path; 1, first imaging device; 2, first detection device; 3, train speed acquisition device; 31, second imaging device; 32, second detection device; 33, calculation device; 4, adjustment device; 5, image processing device; 6, recognition device; 71, first reference object; 72, second reference object; 81, first support rod; 82, second support rod; 91, first lighting lamp; 92, second lighting lamp.Specific Embodiments

[0022] The technical solutions in the embodiments of this disclosure will be described clearly and completely below with reference to the drawings in the embodiments of this disclosure. Obviously, the embodiments described are only some embodiments of this disclosure, rather than all the embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by ordinary technicians in the art without making creative effort fall within the scope of protection of this disclosure.

[0023] In the description of this disclosure, it is appreciated that terms such as "center", "transverse", "longitudinal", "front", "rear", "left", "right", "up", "down", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, which are merely for the purpose of facilitating the description of this disclosure and simplifying the description, but not for indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be construed as limiting the scope of protection of this disclosure.

[0024] As shown in FIGS. 1-3, in some embodiments of the train type recognition system provided by this disclosure, the train type recognition system comprises a first imaging device 1 which is arranged at a first side of a traveling path 10 of the train and is configured to capture an image toward the traveling path 10, and a first detection device 2, which is in signal connection with the first imaging device 1 and is configured to detect whether the image captured by the first imaging device 1 contains at least part of an image of the train, and record arrival or departure time of the train when detecting that the image contains at least part of an image of the train.

[0025] In the embodiment of this disclosure, the train type recognition system comprises a first imaging device 1 through which an image can be captured toward the traveling path 10 of the train, and a first detection device 2 which can detect whether the image captured by the first imaging device 1 contains at least part of an image of the train. When the captured image does not contain any part of the image of the train, it indicates that the train has not arrived or has left the position corresponding to the first imaging device 1. When the captured image contains at least part of the image of the train, it indicates that the train has arrived or has not yet left the position corresponding to the first imaging device 1. When the head of the train appears for the first time in the captured image, it indicates that the train has just arrived at the position corresponding to the first imaging device 1. When the tail of the train appears lastly in the captured image, it indicates that the train is about to leave the position corresponding to the first imaging device 1. In such a way, the arrival and departure of the train can be detected through the first imaging device 1 and the first detection device 2, and the first detection device 2 can also record the arrival time and departure time of the train to calculate the travel speed of the train using these times subsequently.

[0026] By detecting the arrival or departure of the train through the image captured by the first imaging device 1, there is no need to install sensors on the train running track or wheels. This can overcome problems such as high installation requirements, need for on-track operations, and large construction difficulty, etc. due to the installation of sensors, thereby effectively reducing the construction difficulty, the amount of construction, and the construction cost.

[0027] In the embodiments of this disclosure, the first imaging device 1 may comprise a camera, a video recorder, a video camera or the like.

[0028] In some embodiments, the first imaging device 1 includes a line scan camera.

[0029] The line scan camera captures rows of pixels of an image continuously, and then assembles these rows into a complete image. The line scan camera is very suitable for detecting continuously-moving objects and can handle high-speed movement well and capture high-resolution images.

[0030] By adjusting the line rate of the line scan camera, the number of lines scanned by the camera per second can be varied. The line rate determines the refresh rate of the camera and the ability to capture data. A higher line rate can improve the ability of the camera to capture moving objects and the clarity of the image, and thus is particularly suitable for application scenarios that require capturing details and tracking fast-moving objects.

[0031] In some embodiments, the train type recognition system further comprises a train speed acquisition device 3, which is configured to acquire a train speed.

[0032] By providing the train speed acquisition device 3, the train speed can be acquired.

[0033] There may be multiple options for specific structure of the train speed acquisition device 3, and correspondingly there may also be multiple options for the way of acquiring the train speed.

[0034] For example, in some embodiments, the train speed acquisition device 3 is in signal connection with an existing speed monitoring device on the train, thereby acquiring the train speed from the speed monitoring device.

[0035] In some other embodiments, the train speed acquisition device 3 is a dedicated device for measuring the train speed. After the train arrives, it measures the train speed in real time. Specific implementation solutions of this method of acquiring the train speed will be given later.

[0036] In some embodiments, the train type recognition system further comprises an adjustment device 4, which is in signal connection with the train speed acquisition device 3 and is configured to adjust a line rate of the first imaging device 1 according to the train speed.

[0037] By providing the adjustment device 4, the line rate of the first imaging device 1 can be adjusted according to the train speed, so that the line rate of the first imaging device 1 matches the train speed, thereby avoiding deformation of the train in the images captured by the first imaging device 1 due to train speed being too fast or too slow, reducing or avoiding subsequent image processing procedures, saving image processing time, and improving the efficiency of type recognition.

[0038] In some embodiments, the train type recognition system further comprises an image processing device 5, which is in signal connection with the first imaging device 1. The train comprises a plurality of carriages. The image processing device 5 is configured to segment the image captured by the first imaging device 1 to form independent images of the plurality of carriages.

[0039] By providing the image processing device 5, images captured by the first imaging device 1 can be segmented to form independent images of the plurality of carriages, so as to facilitate subsequent type recognition of each carriage.

[0040] By providing the image processing device 5, the purpose of dividing carriages can be achieved through image processing, and it is not necessary to divide carriages by using sensors through coupler information. Therefore, some embodiments of this disclosure can completely omit sensors arranged on the inner side of the track or on wheels, solve the problems such as high installation requirements of sensors, need for online operation, and large construction difficulty, and effectively reduce construction difficulty, construction volume and construction cost.

[0041] By dividing carriages through the image processing device 5, dividing errors caused by inconsistency between the type of chassis and the type of a carriage can also be avoided, thereby effectively improving the accuracy of carriage division.

[0042] In some embodiments, the train type recognition system further comprises a recognition device 6, which is in signal connection with the first imaging device 1 and is configured to recognize a train type according to the image captured by the first imaging device 1.

[0043] By providing the recognition device 6, a train type can be recognized according to the image captured by the first imaging device 1. For example, some type pictures may be pre-stored in the recognition device 6, and by comparing features of images captured by the first imaging device 1 with those of the pre-stored pictures, it can be determined which type the carriage in the images captured by the first imaging device 1 belongs to.

[0044] In embodiments with the image processing device 5, the recognition device 6 may be in signal connection with the image processing device 5, i.e., the recognition device 6 is indirectly in signal connection with the first imaging device 1, so that the recognition device 6 can recognize the train type according to images processed by the image processing device 5, thereby improving the accuracy of type recognition.

[0045] A type of a carriage may be a locomotive for drivers, a passenger carriage for passengers, or a freight carriage for transporting goods, etc.

[0046] By recognizing the type through image recognition, recognition errors caused by inconsistency between the type of chassis and the type of carriage can also be avoided, thereby effectively improving the accuracy of type recognition.

[0047] In some embodiments, the train speed acquisition device 3 comprises a second imaging device 31, which is arranged at the first side of the traveling path 10 and has a preset distance from the first imaging device 1, the second imaging device 31 being configured to capture an image toward the traveling path 10; a second detection device 32, which is in signal connection with the second imaging device 31 and is configured to detect whether the image captured by the second imaging device 31 contains at least part of an image of the train and record arrival or departure time of the train when detecting that the image contains at least part of an image of the train; and a calculation device 33, which is in signal connection with the first detection device 2 and the second detection device 32 and is configured to calculate the train speed according to the magnitude of the preset distance and the time difference between the arrival (departure) time recorded by the first detection device 2 and the arrival (departure) time recorded by the second detection device 32. The second imaging device 31 may adopt a camera, a video recorder or a video camera.

[0048] By providing the second imaging device 31 and the second detection device 32, it is possible to detect the arrival and departure of the train at a position having a preset distance from the first imaging device 1. According to the time difference between the arrival (departure) time recorded by the first detection device 2 and the arrival (departure) time recorded by the second detection device 32, as well as the magnitude of the preset distance between the first imaging device 1 and the second imaging device 31, the travel speed of the train can be calculated.

[0049] By providing the second imaging device 31, the second detection device 32 and the calculation device 33, it is possible that the detection of the travel speed of the train can also be performed without using sensors, so as to overcome problems such as high installation requirements, need for online operation, and large construction difficulty due to the installation of sensors, thereby effectively reducing construction difficulty, construction volume and construction cost.

[0050] In an embodiment, the first imaging device 1 is disposed upstream of the second imaging device 31, so that the train arrives at the first imaging device 1 first, and then arrives at the second imaging device 31. An arrival time recorded by the first detection device 2 is t1, an arrival time recorded by the second detection device 32 is t2, and the distance between the first imaging device 1 and the second imaging device 31 is L, then a travel speed V of the train is V=L / (t2-t1).

[0051] In some embodiments, the adjustment device 4 is further configured to adjust a line rate of the second imaging device 31 according to the travel speed of the train. In this embodiment, the adjustment device 4 may synchronously adjust the line frequencies of the first imaging device 1 and the second imaging device 31 to keep them consistent, thereby simplifying the subsequent processing difficulty of the images captured by the first imaging device 1 and the second imaging device 31 and improving recognition efficiency.

[0052] In some embodiments, the train type recognition system further comprises a first reference object 71, which is disposed on a second side opposite to the first side of the traveling path 10 and is configured to make the image captured by the first imaging device 1 contain the first reference object 71 when the train does not arrive.

[0053] When the train does not arrive, the image captured by the first imaging device 1 contains the first reference object 71. After the train arrives, since the first imaging device 1 and the first reference object 71 are respectively located on two sides of the traveling path 10 of the train, the train will block the first reference object 71 after arriving, thereby reducing the proportion of an area occupied by the first reference object 71 in the total area of the entire image. After the train leaves, the blocking of the first reference object 71 is removed, and the proportion of the first reference object 71 in the entire image will recover.

[0054] Therefore, by providing the first reference object 71 on the opposite side of the first imaging device 1, the first reference object 71 can be used as a reference object in an image captured by the first imaging device 1, so that whether the train arrives or leaves can be determined by determining a proportion of the first reference object 71 contained in the image captured by the first imaging device 1 in the entire image.

[0055] By providing the first reference object 71, a clear basis can be provided for determining whether the train arrives or leaves, which is conducive to improving the efficiency and accuracy of the determination.

[0056] The structure, shape and color of the first reference object 71 can be flexibly selected.

[0057] As shown in FIG. 1, in some embodiments, the first reference object 71 may be an elongated plate-shaped structure, and black and white stripes are provided on the first reference object 71 at intervals. Such a configuration can enhance visual impact and facilitate recognition and differentiation.

[0058] In some embodiments, the train type recognition system further comprises a second reference object 72, which is arranged at a second side of the traveling path 10 opposite to the first side and is configured to make the image captured by the second imaging device 31 contain the second reference object 72 when the train has not arrived.

[0059] When the train does not arrive, the image captured by the second imaging device 31 contains the second reference object 72. After the train arrives, since the second imaging device 31 and the second reference object 72 are located on two sides of the traveling path 10 of the train respectively, the train will block the second reference object 72 after arriving, thereby reducing a proportion of an area occupied by the second reference object 72 in the image captured by the second imaging device 31 in a total area of the entire image. After the train leaves, blocking of the second reference object 72 is removed, and a proportion of the second reference object 72 in the entire image will recover.

[0060] Therefore, by providing the second reference object 72 on the opposite side of the second imaging device 31, the second reference object 72 can be used as a reference object in an image captured by the second imaging device 31, so that whether the train arrives or leaves can be determined by determining a proportion of the second reference object 72 contained in the image captured by the second imaging device 31 in the entire image.

[0061] By providing the second reference object 72, a clear basis can be provided for determining whether the train arrives or leaves, which is conducive to improving the efficiency and accuracy of the determination.

[0062] The structure, shape and color of the second reference object 72 can be flexibly selected.

[0063] As shown in FIG. 1, in some embodiments, the second reference object 72 may be an elongated plate-shaped structure, and black and white stripes are provided on the second reference object 72 at intervals. Such a configuration can enhance visual impact and facilitate recognition and differentiation.

[0064] As shown in FIG. 2, in some embodiments, a distance s1 between the first imaging device 1 and the traveling path 10 is equals to a distance s2 between the second imaging device 31 and the traveling path 10.

[0065] By configuring the distance s1 between the first imaging device 1 and the traveling path 10 to be equal to the distance s2 between the second imaging device 31 and the traveling path 10, a proportion of the train in the image captured by the first imaging device 1 can be kept equal to a proportion of the train in the image captured by the second imaging device 31.

[0066] If the proportion of the train in the image captured by the first imaging device 1 is not equal to the proportion of the train in the image captured by the second imaging device 31, then when determining whether the train arrives or leaves, proportions of the determination basis or feature on the train based on which the determination is made in the images will also be different, thereby possibly resulting in determination faults or determination errors due to different determination standards, affecting the accuracy of the determination.

[0067] Therefore, by configuring the distance s1 between the first imaging device 1 and the traveling path 10 to be equal to the distance s2 between the second imaging device 31 and the traveling path 10, it is conducive to improving the accuracy of determining the arrival or departure of the train.

[0068] Of course, when it is necessary to simultaneously use the image captured by the first imaging device 1 and the image captured by the second imaging device 31 to recognize the train type in subsequent processes, if the proportion of the train in the image captured by the first imaging device 1 is not equal to the proportion of the train in the image captured by the second imaging device 31, then a step of adjusting the proportion of the images captured by the first imaging device 1 and the proportion of the images captured by the second imaging device 31 to be the same will be added, thereby reducing the recognition efficiency. Moreover, the failure rate and error rate of recognition may also increase.

[0069] In some embodiments, the height H1 of the first imaging device 1 from the ground is equal to the height H2 of the second imaging device 31 from the ground.

[0070] The height of the imaging device from the ground affects the relative position between the imaging device and the train, thereby affecting the shooting angle of the imaging device and the perspective and proportion of the train in the image captured by the imaging device.

[0071] If the height H1 of the first imaging device 1 from the ground is not equal to the height H2 of the second imaging device 31 from the ground, then the perspective and proportion of the train in the image captured by the first imaging device 1 and the perspective and proportion of the train in the image captured by the second imaging device 31 will be different, which not only affects the determination of the arrival or departure time of the train, but also affects the recognition of the train type.

[0072] Therefore, by configuring the height H1 of the first imaging device 1 from the ground to be equal to the height H2 of the second imaging device 31 from the ground, it is conducive to improving the accuracy of determining the arrival or departure of the train and is also conducive to improving the efficiency and accuracy of the train type recognition.

[0073] In some embodiments, the train type recognition system further comprises a first support rod 81, on which the first imaging device 1 is mounted.

[0074] By providing the first support rod 81, stable support can be provided for the first imaging device 1. The stability and reliability of the first imaging device 1 during the shooting process can be maintained, and the quality of the captured image can be improved.

[0075] There can be multiple options for the shape of the first support rod 81, for example, it can be cylindrical, quadrangular prism or triangular prism, etc.

[0076] The first support rod 81 can extend in the vertical direction, which facilitates adjustment of the height of the first imaging device 1 from the ground and meet the height requirements of different trains.

[0077] The first imaging device 1 can be mounted in flexible manner. For example, the first imaging device 1 can be sleeved onto the first support rod 81 through an annular clamp or can be engaged in a groove by providing a groove on the first support rod 81, etc.

[0078] In some embodiments, the first imaging device 1 is removably mounted on the first support rod 81 to facilitate adjustment of the height of the first imaging device 1 from the ground.

[0079] In some embodiments, the train type recognition system further comprises a second support rod 82, on which the second imaging device 31 is mounted.

[0080] By providing the second support rod 82, stable support can be provided for the second imaging device 31. The stability and reliability of the second imaging device 31 during the shooting process can be maintained, and the quality of the captured image can be improved.

[0081] There may be multiple options for the shape of the second support rod 82, for example, it can be cylindrical, quadrangular prism or triangular prism, etc.

[0082] The second support rod 82 can extend in the vertical direction, so as to facilitate adjustment of the height of the second imaging device 31 from the ground and meet the height requirements of different trains.

[0083] The second imaging device 31 can be mounted in flexible manner. For example, the second imaging device 31 can be sleeved onto the second support rod 82 through an annular clamp or can be engaged in a groove by providing a groove on the second support rod 82, etc.

[0084] In some embodiments, the second imaging device 31 is removably mounted on the second support rod 82 to facilitate adjustment of the height of the second imaging device 31 from the ground.

[0085] In some embodiments, the train type recognition system further comprises two first lighting lamps 91, both of which are mounted on the first support rod 81 and are respectively located above and below the first imaging device 1.

[0086] By providing the first lighting lamp 91, the train can be provided with light source at night, and meanwhile the first imaging device 1 can be supplemented with light, thereby avoiding that the first imaging device 1 cannot capture a clear image of the train due to too dark light.

[0087] By respectively providing a first lighting lamp 91 above and below the first imaging device 1, it is possible to provide stronger light source for the train, to make light of the train and around the train uniform, and uniformly supplement the light for the first imaging device 1 from top to bottom, thereby avoiding the reduction of image quality caused by uneven light and further the reduction of recognition accuracy.

[0088] In some embodiments, the train type recognition system further comprises two second lighting lamps 92, both of which are mounted on the second support rod 82 and are respectively located above and below the second imaging device 31.

[0089] By providing the second lighting lamp 92, it is possible to provide stronger light source for the train and meanwhile supplement the light for the second imaging device 31, thereby avoiding that the second imaging device 31 cannot capture a clear image of the train due to too dark light.

[0090] By respectively providing a second lighting lamp 92 above and below the second imaging device 31, it is possible to provide stronger light source for the train, to make light of the train and around the train uniform, and uniformly supplement the light for the second imaging device 1 from top to bottom, thereby avoiding the reduction of image quality caused by uneven light and further the reduction of recognition accuracy.

[0091] As shown in FIG. 4, based on the train type recognition device 6 in the above embodiments, this disclosure further provides a method for recognizing a train type, comprising the following steps: capturing an image toward the traveling path 10 by means of the first imaging device 1, detecting whether the image captured by the first imaging device 1 contains at least part of an image of the train, if yes, recording arrival or departure time of the train.

[0092] By detecting arrival or departure of the train through capturing images using the first imaging device 1, there is no need to install sensors on the travelling running track of the train, thereby overcoming problems such as high installation requirements, need for online operation, and great construction difficulty faced when installing sensors, thereby effectively reducing construction difficulty, construction amount and construction cost.

[0093] In some embodiments, the method for recognizing a train type further comprises the following steps: acquiring the train speed, and adjusting the line rate of the first imaging device 1 according to the train speed.

[0094] By adjusting the line rate of the first imaging device 1 according to the travel speed of the train, the line rate of the first imaging device 1 can match the train speed, thereby avoiding deformation of the train in the images captured by the first imaging device 1 due to train speed being too fast or too slow, and thus reducing or avoiding subsequent image processing procedures, saving image processing time, and improving type recognition efficiency.

[0095] In some embodiments, the method for recognizing a train type further comprises: setting the initial line rate of the first imaging device 1 as a first line rate F0, adjusting the line rate of the first imaging device 1 according to the train speed after detecting the arrival of the train, and restoring the line rate of the first imaging device 1 to the first line rate F0 after detecting the departure of the train.

[0096] In the embodiments of this disclosure, before the train arrives and after the train leaves, the line rate of the first imaging device 1 is the first line rate F0, which can be lower than the line rate after the train arrives, which is conducive to reducing the scanning frequency of the first imaging device 1, reducing energy consumption, and reducing the amount of subsequent image processing work.

[0097] In some embodiments, the method for recognizing a train type further comprises: recognizing a train type according to an image captured by the first imaging device 1.

[0098] Specifically, some images of train type may be pre-stored. By comparing features of the image captured by the first imaging device 1 with the features of the pre-stored images, the type of the carriage in the image captured by the first imaging device 1 can be determined.

[0099] In some embodiments, the train comprises a plurality of carriages. The method for recognizing a train type further comprises: segmenting the image captured by the first imaging device 1 to form independent images of the plurality of carriages, and then recognizing the type of each carriage based on the independent image.

[0100] The purpose of dividing carriages can be achieved through image processing, and it is not necessary to divide carriages by using sensors through coupler information. Therefore, some embodiments of this disclosure can completely omit sensors arranged on the track, solve the problems such as high installation requirements of sensors, need for online operation, and large construction difficulty, and effectively reduce construction difficulty, construction volume and construction cost.

[0101] In some embodiments, the method for recognizing a train type further comprises: providing a first reference object 71 on the opposite side of the first imaging device 1, and making the image captured by the first imaging device 1 contain the first reference object 71 when the train has not arrived, determining that the train arrives when the proportion of the first reference object 71 in the image captured by the first imaging device 1 decreases.

[0102] In some embodiments, the method for recognizing a train type further comprises: providing a second reference object 72 on the opposite side of the second imaging device 31, and making the image captured by the second imaging device 31 contain the second reference object 71 when the train has not arrived, determining that the train arrives when the proportion of the second reference object 72 in the image captured by the second imaging device 31 decreases.

[0103] By providing the first reference object 71 and the second reference object 72, a clear basis can be provided for judging whether the train arrives or leaves, which is conducive to improving the efficiency and accuracy of the determination

[0104] The structure and working process of an embodiment of the train type recognition system and method provided by this disclosure will be described below with reference to FIGS. 1-4: As shown in FIGS. 1 and 2, the illustrated train is a train, and the traveling path 10 of the train is a railway track.

[0105] A first imaging devices 1 and second imaging devices 31 (both using line scan cameras) spaced apart by L are mounted on one side of the railway track. The mounting heights of the first imaging device 1 and the second imaging device 31 are on the same horizontal line, i.e., H1=H2. The mounting positions of the first imaging device 1 and the second imaging device 31 are at the same distance from the track, i.e., s1=s2.

[0106] The first imaging device 1 is disposed in front of the second imaging device 31. When the train enters, it passes through the first imaging device 1 first and then passes through the second imaging device 31.

[0107] A first reference object 71 and a second reference object 72 are respectively mounted opposite to the first imaging device 1 and the second imaging device 31. The first imaging device 1 captures the first reference object 71, and the second imaging device 31 captures the second reference object 72.

[0108] The recognition process of the train type recognition system is as follows: As shown in FIGS. 3 and 4, the first imaging device 1 and the second imaging device 31 work all day long, initialize line rate F0=1000, continuously collect images and save one image for 20 seconds, wherein each image has 20 thousand lines; The images captured by the first imaging device 1 and the second imaging device 31 are respectively transmitted to the first detection device 2 and the second detection device 32, which recognize the arrival of the train according to the situation of the first reference object 71 and the second reference object 72 captured by the first imaging device 1 and the second imaging device 31 respectively. When the arrival of the train is confirmed, the arrival time of the first imaging device 1 and the second imaging device 31 are recorded as t1 and t2 respectively, and the calculation device 33 in the train speed acquisition device 3 can calculate the train speed V=L / (t2-t1) accordingly; According to the train speed, a new line rate F=V*r / H can be calculated, wherein r is the resolution of the first imaging device 1 or the second imaging device 31, and H is the height of the train; The adjustment device 4 adjusts the line rate of the first imaging device 1 and the second imaging device 31 according to the train speed, and the first imaging device 1 and the second imaging device 31 continue to capture images at the new line rate; Then, the captured images are transmitted to the image processing device 5, which processes the images to achieve the purpose of dividing the carriages, and independent images of each carriage are saved; Next, the recognition device 6 identifies the type of each carriage by the comparison of the features between the independent images of each carriage and the pre-stored images, and the images and type information are stored; Finally, according to the imaging situation of the first reference object 71 and the second reference object 72, it is identified whether the train leaves. After the departure of the train has been confirmed, the line rate of the first imaging device 1 and the second imaging device 31 returns to 1000, and the above process is repeated.

[0109] In the aforesaid process, the photos processed and recognized by the image processing device 5 and the recognition device 6 may only come from the first imaging device 1, or only from the second imaging device 31, or simultaneously from the first imaging device 1 and the second imaging device 31.

[0110] The embodiments of the recognition system and method provided by this disclosure can save the process of installing sensors on the track or wheels, reducing construction difficulty, construction volume, and construction cost. By dividing the carriages and recognizing the types through images, the recognition error caused by the inconsistency between the chassis type and the carriage type can be avoided, and the accuracy of recognition can be effectively improved.

[0111] In the embodiments of this disclosure, the first detection device 2, the second detection device 32, the calculation device 33, the adjustment device 4, the image processing device 5, and the recognition device 6 may be a general-purpose processor, a programmable logic controller (PLC), a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or any appropriate combination thereof for performing the functions described herein.

[0112] The descriptions of the above embodiments tend to emphasize the differences between the embodiments. For the same or similar parts, reference can be made to each other. For the sake of brevity, they are not repeated here.

[0113] Those skilled in the art can understand that in the above methods of specific embodiments, the written order of each step does not imply a strict execution order that constitutes any limitation on the implementation process. The specific execution order of each step should be determined by its function and possible internal logic.

[0114] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure but not for limiting them. Although the present disclosure has been described in detail with reference to the preferred embodiments, those ordinary skilled in the art should understand that: without departing from the principle of the present disclosure, modifications to the specific embodiments or equivalent replacements for part of the technical features can still be made. These modifications and equivalent replacements should all be covered within the scope of the technical solutions claimed in the present disclosure.

Claims

1. A train type recognition system, comprising: a first imaging device (1), arranged at a first side of a traveling path (10) of the train to capture an image toward the traveling path (10); a first detection device (2), in signal connection with the first imaging device (1) to detect whether the image captured by the first imaging device (1) contains at least part of an image of the train, and record arrival or departure time of the train when detecting that the image contains at least part of an image of the train; a train speed acquisition device (3), configured to acquire a train speed; an adjustment device (4), in signal connection with the train speed acquisition device (3) to adjust a line rate of the first imaging device (1) according to the train speed; and a recognition device (6), in signal connection with the first imaging device (1) to recognize a train type according to the image captured by the first imaging device (1).

2. The train type recognition system according to claim 1, comprising an image processing device (5), which is in signal connection with the first imaging device (1), the train comprising a plurality of carriages, the image processing device (5) being configured to segment the image captured by the first imaging device (1) to form independent images of the plurality of carriages.

3. The train type recognition system according to claim 1, wherein the train speed acquisition device (3) comprises: a second imaging device (31), arranged at the first side of the traveling path (10) to capture an image toward the traveling path (10); and there is a preset distance between the first imaging device (1) and the second imaging device (31); a second detection device (32), in signal connection with the second imaging device (31) to detect whether the image captured by the second imaging device (31) contains at least part of an image of the train and record arrival or departure time of the train when detecting that the image contains at least part of an image of the train, and a calculation device (33), in signal connection with the first detection device (2) and the second detection device (32) and configured to calculate the train speed according to the preset distance and the time difference between the arrival time recorded by the first detection device (2) and the arrival time recorded by the second detection device (32).

4. The train type recognition system according to claim 3, comprising: a first reference object (71), arranged at a second side of the traveling path (10) opposite to the first side to make the image captured by the first imaging device (1) contain the first reference object (71) when the train has not arrived; and / or a second reference object (72), arranged at a second side of the traveling path (10) opposite to the first side to make the image captured by the second imaging device (31) contain the second reference object (72) when the train has not arrived.

5. The train type recognition system according to claim 3, wherein a distance between the first imaging device (1) and the traveling path (10) is equal to a distance between the second imaging device (31) and the traveling path (10).

6. The train type recognition system according to claim 3, wherein the height of the first imaging device (1) from the ground is equal to the height of the second imaging device (31) from the ground.

7. The train type recognition system according to claim 3, comprising: a first support rod (81), on which the first imaging device (1) is mounted; and / or a second support rod (82), on which the second imaging device (31) is mounted.

8. The train type recognition system according to claim 7, comprising: two first lighting lamps (91), both of which are mounted on the first support rod (81), and are respectively located above and below the first imaging device (1); and / or two second lighting lamps (92), both of which are mounted on the second support rod (82) and are respectively located above and below the second imaging device (31).

9. A method for recognizing a train type based on the train type recognition system according to any one of claims 1 to 8, comprising: capturing an image toward the traveling path (10) by means of the first imaging device (1); detecting whether the image captured by the first imaging device (1) contains at least part of an image of the train; if yes, recording arrival or departure time of the train; acquiring the train speed; adjusting the line rate of the first imaging device (1) according to the train speed; and recognizing a train type according to the image captured by the first imaging device (1).

10. The method for recognizing a train type according to claim 9, comprising: setting the initial line rate of the first imaging device (1) as a first line rate F0, adjusting the line rate of the first imaging device (1) according to the train speed after detecting the arrival of the train, and restoring the line rate of the first imaging device (1) to the first line rate F0 after detecting the departure of the train.

11. The method for recognizing a train type according to claim 9, wherein the train comprises a plurality of carriages, and the method further comprises: segmenting the image captured by the first imaging device (1) to form independent images of the plurality of carriages; and recognizing the type of each carriage based on the independent image.

12. The method for recognizing a train type according to claim 9, comprising: providing a first reference object (71) on the opposite side of the first imaging device (1), and making the image captured by the first imaging device (1) contain the first reference object (71) when the train has not arrived; determining whether a proportion of the first reference object (71) in the image captured by the first imaging device (1) decreases: if yes, it is determined that the train arrives.

Images