A train control method, device, train, storage medium and electronic device

Abstract

An embodiment of the present application provides a train control method, device, train, storage medium, and electronic device. The train includes: a carriage and an image sensor; the image sensor is fixedly connected to the train and is used to obtain a first image including the carriage. The method includes: obtaining the first image from the image sensor in real time; based on the first image, locating and identifying the border of the carriage to obtain the current pose data of the carriage; comparing the current pose data with the initial pose data of the carriage to determine whether the current tilt angle of the carriage is greater than a preset angle threshold; and sending a warning signal to the human-machine interaction port of the train when the current tilt angle is greater than the preset angle threshold. By monitoring in real time whether the carriage of the train is in a rollover state, the embodiment of the present application improves the monitoring efficiency and immediacy of the carriage state, and efficiently guarantees the personal safety of the staff and the safe operation of the train.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of rail transit, and in particular, to a train control method, a train control device, a train, a computer-readable storage medium, and an electronic device. Background Art

[0002] Currently, freight ports and railway freight stations mainly control the personnel, vehicles, goods, operation processes, etc. within the site through human prevention means, with a low level of informatization and intelligence. There are deficiencies in both the supervision intensity and supervision efficiency. For this reason, safety accidents frequently occur at freight ports and railway freight stations. For example, after the train carriage in a freight port overturns and dumps out the goods, it needs to be reset in a timely manner, otherwise it will seriously affect the personal safety of the staff and the operation safety of the train. Currently, the method of confirming whether the train carriage is reset is to check manually. This detection method not only consumes a large amount of manpower, has low efficiency, but also cannot monitor in real time, has a lag, and it is difficult to nip the danger in the bud. Therefore, the current train control urgently needs to be improved, and the personal safety of the staff and the operation safety of the train cannot be fully guaranteed. Summary of the Invention

[0003] The embodiments of the present application provide a train control method, a train control device, a train, a computer-readable storage medium, and an electronic device, aiming to monitor in real time whether the carriage of the train is in a rollover state, improve the monitoring efficiency and immediacy of the carriage state, and efficiently guarantee the personal safety of the staff and the safe operation of the train.

[0004] In one aspect, the embodiments of the present application provide a train control method. The train includes: a carriage and an image sensor; the image sensor is fixedly connected to the train and is used to obtain a first image including the carriage. The method includes:

[0005] Obtain the first image from the image sensor in real time;

[0006] Based on the first image, perform positioning and recognition on the border of the carriage to obtain the current pose data of the carriage;

[0007] Compare the current pose data with the initial pose data of the carriage to determine whether the current tilt angle of the carriage is greater than a preset angle threshold;

[0008] In the case where the current tilt angle is greater than the preset angle threshold, send a warning signal to the human-machine interaction port of the train.

[0009] Optionally, based on the first image, performing positioning and recognition on the border of the carriage to obtain the current pose data of the carriage includes:

[0010] Rectify the distortion of the first image to obtain a second image;

[0011] Based on the second image, locate and identify the border of the carriage to obtain the current pose data of the carriage.

[0012] Optionally, rectifying the distortion of the first image to obtain a second image includes:

[0013] Detect the distortion type of the first image;

[0014] According to the distortion type, select a rectification algorithm corresponding to the first image;

[0015] Use the rectification algorithm to rectify the distortion of the first image to obtain the second image.

[0016] Optionally, based on the second image, locating and identifying the border of the carriage to obtain the current pose data of the carriage includes:

[0017] Based on the contrast enhancement algorithm of the local area, normalize the luminance channel value of the second image, and adaptively enhance the luminance of the second image to obtain a third image;

[0018] Use a preset target segmentation algorithm to locate and identify the border of the carriage in the third image to obtain the current pose data of the carriage.

[0019] Optionally, compare the current pose data with the initial pose data to determine whether the current tilt angle of the carriage is greater than a preset angle threshold, including:

[0020] Calculate the offset distance between the border of the carriage in the current pose data and the border of the carriage in the initial pose data;

[0021] Determine whether the current tilt angle of the carriage is greater than a preset angle threshold;

[0022] Among them, when the offset distance is greater than a preset distance threshold, it is determined that the current tilt angle of the carriage is greater than the preset angle threshold.

[0023] Optionally, the image sensor is located above the connection between the carriages, so that the first image includes two upper side borders of the carriage along the train arrangement direction;

[0024] Alternatively, the image sensor is located above one side of the carriage along the display direction of the train, so that the first image includes at least one side upper border of the carriage along the display direction of the train, and in the case where the carriage is tilted, the image includes two side upper borders of the carriage along the display direction of the train.

[0025] In yet another aspect, an embodiment of the present application further provides a train control device, where the train includes: a carriage and an image sensor; the image sensor is fixedly connected to the train and is configured to obtain a first image including the carriage; the device includes:

[0026] An acquisition unit, configured to acquire the first image from the image sensor in real time;

[0027] A positioning unit, configured to perform positioning and recognition on the border of the carriage based on the first image to obtain current pose data of the carriage;

[0028] A comparison unit, configured to compare the current pose data with initial pose data of the carriage to determine whether a current tilt angle of the carriage is greater than a preset angle threshold;

[0029] An alarm unit, configured to send an alarm signal to a human-machine interaction port of the train when the current tilt angle is greater than the preset angle threshold.

[0030] In yet another aspect, an embodiment of the present application further provides a train, including:

[0031] A carriage;

[0032] An image sensor, fixedly connected to the train and configured to obtain a first image including the carriage;

[0033] A train control device, configured to implement the steps in the method of any one of the above embodiments of the present application.

[0034] In yet another aspect, another embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps in the method of any one of the above embodiments of the present application are implemented.

[0035] In yet another aspect, another embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes, the steps in the method of any one of the above embodiments of the present application are implemented.

[0036] Compared with the prior art, the advantages of the embodiments of the present application are as follows:

[0037] In the embodiment of the present application, an image sensor is arranged on the train car body to monitor the state of the train during operation at any time, and determine whether the tilt angle of the carriage is greater than a preset angle threshold through the acquired images in real time, so as to monitor in real time whether the train carriage is in a rollover state. On the one hand, it can avoid the train rollover caused by the failure of the freight train carriage to reset in time after dumping goods. On the other hand, it can also give a timely warning of the dangerous tilt of the vehicle body during the train running. By improving the monitoring efficiency and timeliness of the train state, it can effectively ensure the personal safety of the staff and the safe operation of the train, and reduce potential safety hazards. Description of the Drawings

[0038] The drawings are only for reference and illustration purposes and are not intended to limit the protection scope of the present application. The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0039] Figure 1 The flowchart of the steps of a train control method provided by an embodiment of the present application is shown;

[0040] Figure 2 The schematic diagram of an image distortion provided by an embodiment of the present application is shown;

[0041] Figure 3 The schematic diagram of the offset of the border of the carriage in a top view image provided by an embodiment of the present application is shown;

[0042] Figure 4 The schematic diagram of the change of the border of the carriage in a side view image provided by an embodiment of the present application is shown;

[0043] Figure 5 The comparison schematic diagram of an adaptive enhanced image provided by an embodiment of the present application is shown;

[0044] Figure 6 The block diagram of the structure of a train control device provided by an embodiment of the present application is shown. Detailed Embodiments

[0045] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0046] With the rapid development of information technology and Internet of Things technology, it is of great significance to build an intelligent railway port station. To strengthen the construction of port station intelligence, adhere to the principle of safety production management, and tightly grasp the investigation and treatment of potential safety hazards in equipment conditions, it has become particularly urgent to form a complete monitoring system through informatization construction and realize the intelligent supervision of shunting operations within the entire station. After a freight train dumps its cargo, there is a potential safety hazard that the carriage cannot be fully reset. It is necessary to improve it through train automation control to effectively ensure the personal safety of staff and the safe operation of the train. Taking the current port freight station as an example, it is difficult to safely and efficiently determine whether the carriage of a coal transport train has been successfully reset after tipping and cleaning the coal, which easily leads to the overall rollover of the train when it runs without being fully reset, thus causing personal safety accidents.

[0047] In view of this, the embodiments of the present application propose a train control method, a train control device, a train, a computer-readable storage medium, and an electronic device, aiming to monitor in real time whether the vehicle of the train is in a rollover state, improve the monitoring efficiency and immediacy of the train state, and effectively ensure the personal safety of staff and the safe operation of the train.

[0048] The embodiments of the present application will be described below with reference to the accompanying drawings.

[0049] Refer to Figure 1 , Figure 1 which shows the flowchart of the steps of a train control method in an embodiment provided by the present application. As Figure 1 shown, the embodiments of the present application provide a train control method. The train includes: a carriage and an image sensor; the image sensor is fixedly connected to the train and is used to obtain a first image including the carriage.

[0050] Since other objects in the viewfinder of the image sensor except the carriage need to be used as reference objects to judge the pose of the carriage, the image sensor can be stationary relative to the whole train and does not follow the carriage.

[0051] Among them, the first image can exist in the form of a video of a specified number of frames per second. Exemplarily, the first image can be an image in video data including 24 consecutive images per second.

[0052] Specifically, the train can be a freight train. Exemplarily, the train can be a coal transport train.

[0053] Correspondingly, the carriage can be a box body with an open top. Among them, the carriage can be flipped by more than 90° around the display direction of the train, pour out the cargo it carries by tipping the open top, and reset to the initial position after the cargo is emptied.

[0054] Among them, the image sensor can be a CCD image sensor. To facilitate monitoring during night operations of the train, the image sensor can be an infrared image sensor to achieve night-time photography.

[0055] The method includes:

[0056] Step S301, obtain the first image in real time from the image sensor.

[0057] Among them, to ensure the accuracy of the calculation, the first image can have a preset clarity. Exemplarily, the first image can be video data with a resolution of 1280×720.

[0058] Step S302, based on the first image, locate and identify the border of the carriage to obtain the current pose data of the carriage.

[0059] In the case where the first image is provided in the form of a video, in order to reduce the amount of calculation, the method of frame extraction can also be selected to locate and identify the border of the carriage.

[0060] In some alternative embodiments, key frame images can be processed to locate and identify the border of the carriage.

[0061] In some alternative embodiments, the first images obtained can also be selected at a preset time interval, and the border of the carriage can be located and identified based on the selected first images. Exemplarily, frame extraction can be performed every 0.5 seconds.

[0062] Specifically, an image neural network algorithm can be used to locate and identify the border of the carriage.

[0063] Among them, the current pose data can be data presenting the position and pose of the carriage in the viewfinder of the first image. Specifically, the current pose data can include the relative distance between the two upper side borders of the current carriage in the viewfinder of the first image. The current pose data can also include the position coordinates of the two upper side borders of the current carriage in the viewfinder of the first image.

[0064] Step S303, compare the current pose data with the initial pose data of the carriage to determine whether the current tilt angle of the carriage is greater than a preset angle threshold.

[0065] Among them, the initial pose data can be the pose data of the carriage when the carriage is fully reset to enable the train to enter the driving state. Specifically, the initial pose data can include the relative distance between the two upper side borders of the carriage in the viewfinder of the first image when the carriage is in the fully reset state. The current pose data can also include the position coordinates of the two upper side borders of the carriage in the viewfinder of the first image when the carriage is in the fully reset state.

[0066] Step S304, when the current tilt angle is greater than a preset angle threshold, send an alarm signal to the human - machine interaction port of the train.

[0067] Specifically, the preset angle threshold can be the maximum tilt angle of the carriage to ensure the safe operation of the train. Exemplarily, the preset angle threshold can be any value between 5° and 10°.

[0068] Among them, an alarm signal can also be sent to the monitoring computer and external systems subscribing to the alarm, that is, the alarm is remotely realized.

[0069] The embodiments of the present application can be executed by an algorithm server set remotely. Therefore, after the image sensor acquires the first image meeting the method requirements, it can be transmitted to the algorithm server through the network link, and the algorithm server executes the method described in the embodiments of the present application.

[0070] Refer to Figure 3 , Figure 3 shows a schematic diagram of the offset of the border of the carriage in a top - down image in an embodiment provided by the present application. As Figure 3 shown, when the image sensor is set above the carriage, by the change between the two borders of the carriage in the top - down image, it can be determined whether the carriage is tilted and the tilt angle. For this reason, the embodiments of the present application consider setting the image sensor above the carriage. In an alternative embodiment, the image sensor is located above the connection between the carriages, so that the image includes the two upper - side borders of the train along the track direction.

[0071] Refer to Figure 4 , Figure 4 shows a schematic diagram of the change of the border of the carriage in a side - view image in an embodiment provided by the present application. As Figure 4 shown, when the image sensor is set above the side of the carriage, the upper - side border of this side of the carriage is included in the side - view image. And when the carriage does not tilt, only the upper - side border of this side of the carriage can be recorded. When the carriage tilts, the first image includes the upper - side border of the other side of the carriage, and the tilt angle of the carriage can be determined by the distance between the two borders on both sides in the viewfinder of the first image. For this reason, in another alternative embodiment, the image sensor is located above one side of the carriage along the display direction of the train, so that the image includes at least one upper - side border of the train along the track direction, and when the train tilts, the image includes the two upper - side borders of the train along the track direction.

[0072] Among them, the greater the distance between the two upper - side borders in the viewfinder, the greater the tilt angle of the carriage.

[0073] In an alternative embodiment, a third-party image sensor may also be provided on the ground of the freight station, with the third-party image sensor located on the side of the train, thereby obtaining a fourth image. Its application is similar to that of the side view image obtained when the image sensor is provided above the side of the carriage, and it can also be used to detect the inclination angle of the carriage.

[0074] Through the above embodiments, an image sensor is provided on the train body of the present application to be able to monitor the state of the train during operation at any time, and to determine whether the inclination angle of the carriage is greater than a preset angle threshold through the images obtained in real time, so as to monitor in real time whether the train is in a rollover state, thereby improving the monitoring efficiency and immediacy of the carriage state, efficiently ensuring the personal safety of the staff and the safe operation of the train, and reducing potential safety hazards. The method provided by the embodiments of the present application can be used for the automatic identification of the safety of railway freight carriages, can perform real-time inclination detection of the carriage in real time and effectively, reduce the consumption of manpower and material resources, and ensure the operation safety during the turnover and reset process of the coal-carrying train carriage.

[0075] In the embodiments of the present application, since the image sensor can monitor the state of the train during operation at any time, the method provided by the embodiments of the present application can also be used to detect the lateral inclination angle of the train during driving, give an alarm in time in the case of a risk of the whole train rolling over, ensure the safe operation of the train, and can also be used to improve the speed of obtaining rescue after the train rolls over.

[0076] The lens used by the image sensor will introduce distortion due to the deviation of manufacturing precision and assembly process, resulting in the distortion of the original image. Moreover, the image sensor is fixed on the vehicle body. Although it can monitor the state of the train during operation at any time through the integrated setting with the train, thereby ensuring the safe operation of the train. However, the inventor found that due to the too-close setting of the image sensor and the vehicle body, restricted by the characteristics of the lens of the image sensor itself, it is more likely to cause image distortion, and thus subsequent comparison calculations cannot be accurately performed.

[0077] In view of this, in an alternative embodiment, the present application also provides a method for obtaining the current pose data of the carriage, including:

[0078] Step S401, correct the distortion of the image to obtain a second image.

[0079] Among them, the distortion can include two categories: radial distortion and tangential distortion.

[0080] Refer to Figure 2 , Figure 2 shows a schematic diagram of an image distortion in an embodiment provided by the present application. As Figure 2As shown, the distortion at the center of the optical axis of the radial distortion imager is 0. Moving along the radius direction of the lens towards the edge, the distortion becomes more and more serious. The mathematical model of distortion can be described by the first few terms of the Taylor series expansion around the principle point. Usually, the first two terms, namely k1 and k2, are used. For lenses with large distortion, such as fish-eye lenses, barrel distortion is mainly generated.

[0081] As Figure 2 shown, tangential distortion is caused by the non - parallelism between the lens itself and the imaging plane of the camera sensor. This situation is mostly due to the mounting deviation when the lens is attached to the lens module, mainly causing pincushion distortion.

[0082] Step S402: Locate and identify the border of the carriage in the second image to obtain the current pose data of the carriage.

[0083] Through the above - mentioned embodiments, by locating and identifying the border of the carriage in the second image with corrected distortion, the accuracy of the current pose data of the carriage can be improved, thereby enhancing the accuracy of sending warning signals and avoiding false triggering of warnings or untimely warnings.

[0084] Considering that there are different situations of image distortion, different algorithms can be selected to correct different types of distortion. For this purpose, in an alternative embodiment, the present application also provides a method for obtaining a second image, including:

[0085] Step S501: Detect the distortion type of the first image.

[0086] In an alternative example, for radial distortion / barrel distortion, the third term k3 can be additionally used for description. For a certain point on the imager, according to its distribution position in the radial direction, formula (1) is adjusted as:

[0087] x0 = x(1 + k1r 2 + k2r 4 + k3r 6 )

[0088] y0 = y(1 + k1r 2 + k2r 4 + k3r 6 ) (1)

[0089] Among them, (x0, y0) is the original position of the distorted point on the imager, (x, y) is the new position after correction, k1, k2, and k3 are distortion parameters of the Taylor series expansion around the distorted point, and r is the radius with the center of the imager optical axis as the center of the circle.

[0090] In an optional example, for tangential distortion / pincushion distortion, it is necessary to adjust the mounting deviation of the lens pasted onto the lens module. The adjustment formula (2) is as follows:

[0091] x0 = x + [2p1y + p2(r 2 + 2x 2 )]

[0092] y0 = y + [2p2x + p1(r 2 + 2y 2 )] (1)

[0093] where (x0, y0) is the original position of the distorted point on the imager, (x, y) is the new position after correction, p1 and p2 are distortion parameters, and r is the radius with the center of the imager optical axis as the center of the circle.

[0094] Step S502: Select a correction algorithm corresponding to the first image according to the distortion type.

[0095] Step S503: Use the correction algorithm to correct the distortion of the first image to obtain a second image.

[0096] In an optional example, radial and tangential distortions can be eliminated simultaneously. By combining formulas (1) and (2), a total of 5 distortion parameters can be obtained and adjusted simultaneously using formula (3):

[0097]

[0098] where (x0, y0) is the original position of the distorted point on the imager, (x, y) is the new position after correction, k1, k2, k3, p1, and p2 are distortion parameters, and r is the radius with the center of the imager optical axis as the center of the circle.

[0099] Among them, the distortion parameters k1, k2, k3, p1, and p2 can be calculated using a calibration picture.

[0100] If the calculated point (x0, y0) of the corresponding original image is not an integer, this point can be calculated using bilinear interpolation and then assigned to (x, y). In an optional example, assuming that the RGB values of point x1 are R1, G1, B1 respectively, and the RGB values of point x2 are R2, G2, B2 respectively, then the RGB value calculation formula (4) for point x3 between x1 and x2 is:

[0101] R3 = R1 * (1 - X3) + R2 * X3

[0102] G3 = G1 * (1 - X3) + G2 * X3

[0103] B3 = B1 * (1 - X3) + B2 * X3 (4)

[0104] Refer to Figure 5 , Figure 5 shows a comparison diagram of an adaptive enhanced image in an embodiment provided by the present application. As Figure 5 shown, image segmentation has requirements for the contrast of the target in the image. If the contrast is relatively low, it is difficult to accurately segment the target in the image. And the brightness of the image affects the contrast clarity of the picture, which in turn affects the accuracy of the positioning and recognition of the carriage border. In order to improve the accuracy of warning, the local brightness of the image can also be enhanced to improve the accuracy of the positioning and recognition of the carriage border. For this reason, in an optional implementation manner, the present application also provides a method for obtaining the current pose data of the carriage, including:

[0105] Step S601, based on the contrast enhancement algorithm of the local area, normalize the brightness channel value of the second image, and adaptively enhance the brightness of the second image to obtain a third image.

[0106] In an optional example, the following formula (5) can be used for brightness adaptive enhancement:

[0107]

[0108] where, I in (x, y) ∈ [0, 1] represents the brightness value at the (x, y) position after normalizing the brightness channel value of the second image, where the second image can be obtained by using the top-down shooting angle; q is an adaptive adjustment parameter; I E (x, y) is the brightness value after adaptive enhancement at the (x, y) position.

[0109] where, q can be adaptively adjusted according to the local value of the brightness image pixels according to the following formula (6):

[0110]

[0111] where, I ave (x, y) is; c1 and c2 are two constants that can be preset according to experience; ε = 0.01; I ave (x, y) ∈ [0, 1] is the normalized pixel mean.

[0112] where, according to the multi-scale idea, the normalized pixel mean can be obtained by the following formula (7):

[0113]

[0114] where, G iare Gaussian functions of different scales, where m and n are different scales in the X and Y coordinates respectively.

[0115] Among them, the Gaussian function has the relationships in the following formulas (8) and (9):

[0116] ∫∫G(x, y)dxdy = 1, (8)

[0117]

[0118] Among them, c is the standard deviation, and K is the Boltzmann constant.

[0119] After the original image is adaptively enhanced in brightness in the above example, the center-surround local contrast enhancement technology can also be used for adaptive local contrast enhancement, as shown in the following formulas (10) and (11):

[0120]

[0121]

[0122] Among them, p is the global brightness standard deviation in the first image, and σ g can be adaptively valued.

[0123] After the local area contrast of the original image is enhanced, the boundary contrast of the target area can become more prominent, which is beneficial to image segmentation.

[0124] Step S602, using a preset target segmentation algorithm, locate and identify the frame of the carriage in the third image to obtain the current pose data of the carriage.

[0125] Among them, in some optional embodiments, the watershed segmentation algorithm can be used to separate the frame of the carriage from other areas; then the canny edge detection algorithm is used to detect the boundaries of different segmented areas; finally, the hough line segment detection algorithm is used among all the boundaries to obtain line segments of different angles and lengths.

[0126] In an optional example, a length threshold and an angle threshold can be set. If the length of the line segment is less than the length threshold, or the absolute value of the angle of the line segment is less than the angle threshold, then the line segment can be filtered out. Finally, two adjacent side lines on the frame of each carriage are used as a group, and two sets of body frame lines of the carriage when it is tilted can be obtained.

[0127] By the same method provided in the above example, two sets of body frame lines of the carriage in the initial state when it is not tilted can also be obtained.

[0128] For this reason, in an optional implementation manner, the present application also provides a method for judging the current tilt angle of the carriage, including:

[0129] Step S701: Calculate the offset distance between the border lines of the carriage in the current pose data and the border lines of the carriage in the initial pose data.

[0130] Step S702: Determine whether the current tilt angle of the carriage is greater than a preset angle threshold.

[0131] Step S703: Among them, when the offset distance is greater than a preset distance threshold, it is determined that the current tilt angle of the carriage is greater than the preset angle threshold.

[0132] Through the above embodiments, the image sensor can take a top-down photo from the roof of the car. Then, by comparing and calculating the offset distance between the border of the currently tilted carriage and the border of the initial car body, it can be determined whether the flip angle of the carriage reaches the alarm threshold, and an alarm is realized when the offset distance is greater than the preset distance threshold.

[0133] Refer to Figure 6 , Figure 6 shows a structural block diagram of a train control device in an embodiment provided by the present application. As Figure 6 shown, based on the same inventive concept, an embodiment of the present application further provides a train control device, and the train includes: a carriage and an image sensor; the image sensor is fixedly connected to the train and is used to obtain a first image including the carriage.

[0134] The device includes:

[0135] An acquisition unit 901, configured to acquire the first image from the image sensor in real time;

[0136] A positioning unit 902, configured to perform positioning and recognition on the border of the carriage based on the first image to obtain the current pose data of the carriage;

[0137] A comparison unit 903, configured to compare the current pose data with the initial pose data of the carriage to determine whether the current tilt angle of the carriage is greater than a preset angle threshold;

[0138] An alarm unit 904, configured to send an alarm signal to the human-machine interaction port of the train when the current tilt angle is greater than the preset angle threshold.

[0139] Based on the same inventive concept, another embodiment of the present application provides a train, including:

[0140] A carriage;

[0141] An image sensor, fixedly connected to the train and used to obtain a first image including the carriage;

[0142] A train control device for implementing the steps in the method according to any one of the foregoing embodiments of the present application.

[0143] Based on the same inventive concept, another embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps in the method according to any one of the foregoing embodiments of the present application are implemented.

[0144] Based on the same inventive concept, another embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes, the steps in the method according to any one of the foregoing embodiments of the present application are implemented.

[0145] For the apparatus embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and for the relevant parts, reference may be made to the partial description of the method embodiment.

[0146] Each embodiment in this specification is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. For the same or similar parts among the embodiments, reference may be made to each other.

[0147] Those skilled in the art should understand that the embodiments of the present application may be provided as a method, an apparatus, or a computer program product. Therefore, the embodiments of the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0148] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the method, terminal device (system), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal devices generate a device for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0149] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured article including an instruction device, and the instruction device implements the operations in the process Figure 1 one process or multiple processes and / or blocks Figure 1 the functions specified in one block or multiple blocks.

[0150] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device, so that a series of operation steps are executed on the computer or other programmable terminal device to generate a computer-implemented process. Therefore, the instructions executed on the computer or other programmable terminal device provide steps for implementing the functions specified in one process or multiple processes and / or blocks Figure 1 one process or multiple processes and / or blocks Figure 1 the functions specified in one block or multiple blocks.

[0151] Although the preferred embodiments of the embodiments of the present application have been described, those skilled in the art can make additional changes and modifications once they learn the basic creative concepts. Therefore, the appended claims are intended to be construed as including the preferred embodiments and all changes and modifications falling within the scope of the embodiments of the present application.

[0152] Finally, it should also be noted that in this text, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or terminal device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or terminal device including the element.

Claims

1. A train control method, characterized in that, The train includes: carriages and an image sensor; the image sensor is fixedly connected to the train and is configured to obtain a first image including the carriages; the method includes: Obtain the first image from the image sensor in real time; Based on the first image, locate and identify the border of the carriage to obtain the current pose data of the carriage; Compare the current pose data with the initial pose data of the carriage to determine whether the current tilt angle of the carriage is greater than a preset angle threshold; When the current tilt angle is greater than the preset angle threshold, send an alarm signal to the human-machine interaction port of the train; Wherein, the current pose data includes the relative distance between the two upper side borders of the current carriage in the viewfinder of the first image and the position coordinates of the two upper side borders of the current carriage in the viewfinder of the first image; The comparing the current pose data with the initial pose data to determine whether the current tilt angle of the carriage is greater than a preset angle threshold includes: Calculate the offset distance between the border of the carriage in the current pose data and the border of the carriage in the initial pose data; Determine whether the current tilt angle of the carriage is greater than a preset angle threshold; Wherein, when the offset distance is greater than a preset distance threshold, it is determined that the current tilt angle of the carriage is greater than the preset angle threshold.

2. The train control method according to claim 1, wherein Based on the first image, locating and identifying the border of the carriage to obtain the current pose data of the carriage includes: Correct the distortion of the first image to obtain a second image; Based on the second image, locate and identify the border of the carriage to obtain the current pose data of the carriage.

3. The train control method according to claim 2, wherein Correcting the distortion of the first image to obtain a second image includes: Detect the distortion type of the first image; According to the distortion type, select a correction algorithm corresponding to the first image; Use the correction algorithm to correct the distortion of the first image to obtain the second image.

4. The train control method according to claim 2, wherein Based on the second image, locating and identifying the border of the carriage to obtain the current pose data of the carriage includes: Based on a local region contrast enhancement algorithm, normalize the luminance channel value of the second image and adaptively enhance the luminance of the second image to obtain a third image; Use a preset target segmentation algorithm to locate and identify the border of the carriage in the third image to obtain the current pose data of the carriage.

5. The train control method according to claim 1, wherein The image sensor is located above the connection between the carriages, so that the first image includes the two upper side borders of the carriage along the train arrangement direction; Alternatively, the image sensor is located above one side of the carriage along the train arrangement direction, so that the first image includes at least one upper side border of the carriage along the train arrangement direction, and when the carriage is tilted, the image includes the two upper side borders of the carriage along the train arrangement direction.

6. A train control device, characterized in that, The train includes: a carriage and an image sensor; the image sensor is fixedly connected to the train and is configured to obtain a first image including the carriage; the device includes: An acquisition unit, configured to acquire the first image from the image sensor in real time; A positioning unit, configured to perform positioning and recognition on the frame of the carriage based on the first image to obtain the current pose data of the carriage; A comparison unit, configured to compare the current pose data with the initial pose data of the carriage to determine whether the current tilt angle of the carriage is greater than a preset angle threshold; An alarming unit, configured to send an alarm signal to the human-machine interaction port of the train when the current tilt angle is greater than the preset angle threshold; Wherein, the current pose data includes the relative distance between the two upper side frames of the current carriage in the viewfinder of the first image and the position coordinates of the two upper side frames of the current carriage in the viewfinder of the first image; The comparing the current pose data with the initial pose data to determine whether the current tilt angle of the carriage is greater than a preset angle threshold includes: Calculating the offset distance between the frame of the carriage in the current pose data and the frame of the carriage in the initial pose data; Determining whether the current tilt angle of the carriage is greater than a preset angle threshold; Wherein, when the offset distance is greater than a preset distance threshold, it is determined that the current tilt angle of the carriage is greater than the preset angle threshold.

7. A train, characterized in that, Includes: A carriage; An image sensor, fixedly connected to the train and configured to obtain a first image including the carriage; A train control device, configured to implement the steps in the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, A computer program is stored thereon, and when the program is executed by a processor, it implements the steps in the method according to any one of claims 1 to 5.

9. An electronic device, characterized in that, Includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes, it implements the steps in the method according to any one of claims 1 to 5.

Images