Train wheel position monitoring method, device and system

Abstract

According to the train wheel position monitoring method, device and system provided by the invention, the position of the train wheel is accurately measured by periodically utilizing the stereo camera and the thermal imager, and the measured image is automatically displayed, so that the accuracy and robustness of measurement are remarkably improved. When an image selection instruction is received, the system automatically marks a to-be-measured image and a to-be-measured wheel in the to-be-measured image, then the image is processed through an edge detector, the vertical edge of the wheel is extracted, and the horizontal coordinates of edge white points are further accurately extracted through an iterative algorithm. The wheel width is determined by calculating the radian angle between the horizontal center of the image and the edge white point, and the wheel width is compared with the preset template to obtain the offset so as to judge whether the wheel shifts or not. According to the method, the dependence on camera setting is reduced, the operation efficiency is improved, the implementation and operation cost is reduced, meanwhile, the strict requirements of a railway transportation system for high safety and high efficiency are met, and the huge popularization value is shown.

Description

Technical Field

[0001] The present invention relates to the technical field of position monitoring and positioning, and in particular to a method, device and system for monitoring the position of train wheels. Background Art

[0002] In the railway transportation system, accurate monitoring of train wheel position is crucial to ensure driving safety, improve operational efficiency, and prevent major accidents. First, abnormal wheel position may lead to serious consequences such as derailment or collision, so real-time monitoring becomes an effective preventive measure. Secondly, the railway operating environment is complex, including curves, ramps, and bridges. Accurate monitoring of wheel position helps trains travel safely under different conditions. With the continuous advancement of railway technology and the increasing safety requirements, wheel position monitoring is particularly important.

[0003] At present, accelerometers are mainly used to monitor the position of train wheels. Their principle is based on oscillation and can monitor the acceleration of an object in a specific direction, thereby measuring the state of motion. In this process, the accelerometer works by capturing the vibration signal between the wheel and the rail. However, factors such as rail quality, rail gaps, curves and track geometry will affect the characteristics of the vibration signal, which may lead to a decrease in measurement accuracy. In addition, the accelerometer needs to be calibrated regularly, which not only increases the cost but also increases the complexity of operation. Although it can monitor vibration signals, the measurement range is limited by factors such as wheel size and running speed.

[0004] Therefore, how to improve the accuracy and reliability of wheel position monitoring is an urgent problem that needs to be solved. Summary of the invention

[0005] In view of this, the embodiments of the present invention provide a train wheel position monitoring method, device and system to solve the problem of inaccurate monitoring of train wheel position at present.

[0006] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0007] A first aspect of the present invention discloses a method for monitoring the position of a train wheel, the method comprising:

[0008] Periodically measuring the position of the train wheels using a stereo camera and a thermal imager, and displaying the measured images; wherein the thermal imager is pre-selected from a plurality of sensors;

[0009] When an image selection instruction is received, the image corresponding to the selection instruction is marked as the image to be measured, and the object selected by the selection instruction in the image to be measured is marked as the wheel to be measured;

[0010] Processing the image to be measured by an edge detector to obtain a binary image including a vertical edge of the wheel to be measured;

[0011] Acquire the coordinates of the cursor in the image to be measured, and extract the horizontal coordinates of the left edge white point and the right edge white point in the vertical edge of the wheel to be measured from the binary image through an iterative algorithm, wherein the left edge white point is the edge point in the vertical left edge of the wheel to be measured with the shortest distance to the coordinates; the right edge white point is the edge point in the vertical right edge of the wheel to be measured with the shortest distance to the coordinates;

[0012] Calculating the arc angles between the horizontal center of the binary image and the left edge white point, and the right edge white point, respectively, to determine the width of the wheel to be tested;

[0013] Comparing the width with the template edge images of the wheel and rail to obtain an offset;

[0014] If the offset is not greater than the preset value, it is determined that the wheel to be tested has not shifted.

[0015] Preferably, the process of selecting a thermal imager from a plurality of sensors includes:

[0016] Obtain the weight ranking results of each expert on each lateral wheel position measurement factor to construct a ranking matrix and calculate the Kendall consistency coefficient;

[0017] Chi-square values ​​were calculated based on the Kendall coefficient of concordance described;

[0018] If the chi-square value is greater than a preset threshold, it is determined that the weight ranking results of the experts for each lateral wheel position measurement factor are consistent;

[0019] Analyzing the ranking matrix using the analytic hierarchy process to obtain an analytic hierarchy level matrix;

[0020] A thermal imager is obtained by screening from a plurality of sensors according to the hierarchical analysis level matrix.

[0021] Preferably, respectively calculating the arc angle between the horizontal center of the binary image and the left edge white point, and the right edge white point to determine the width of the wheel to be measured includes:

[0022] Calculate the left distance between the left edge white point and the horizontal center of the image, and the right distance between the right edge white point and the horizontal center of the image;

[0023] According to the left distance and the right distance, respectively calculate the left arc angle between the horizontal center of the image and the white point on the left edge, and the right arc angle between the horizontal center of the image and the white point on the right edge;

[0024] The width of the wheel to be measured is calculated based on the left arc angle and the right arc angle.

[0025] Preferably, the calculating, according to the left distance and the right distance, respectively the left arc angle between the horizontal center of the image and the white point on the left edge, and the right arc angle between the horizontal center of the image and the white point on the right edge, comprises:

[0026] Obtaining the horizontal field of view angle of the stereo camera;

[0027] Calculate the angle of each pixel according to the horizontal field of view angle and the horizontal center of the image;

[0028] Calculate the product of the angle of each pixel and the left distance to obtain the left arc angle between the horizontal center of the image and the left edge white point;

[0029] The product of the angle of each pixel and the right distance is calculated to obtain the right arc angle between the horizontal center of the image and the right edge white point.

[0030] Preferably, the calculating the width of the wheel to be measured based on the left arc angle and the right arc angle includes:

[0031] Calculating a left end distance between a left end of the wheel to be measured and a horizontal center of the image, and a right end distance between a right end of the wheel to be measured and a horizontal center of the image based on the left arc angle and the right arc angle;

[0032] The modulus of the distance difference between the left end distance and the right end distance is calculated to obtain the width of the wheel to be measured.

[0033] A second aspect of the present invention discloses a train wheel position monitoring device, the device comprising:

[0034] A measuring unit, used for periodically measuring the position of the train wheels using a stereo camera and a thermal imager, and displaying the measured images; wherein the thermal imager is pre-selected from a plurality of sensors;

[0035] a marking unit, configured to, when receiving an image selection instruction, mark the image corresponding to the selection instruction as the image to be measured, and mark the object selected by the selection instruction in the image to be measured as the wheel to be measured;

[0036] A processing unit, configured to process the image to be measured by an edge detector to obtain a binary image including a vertical edge of the wheel to be measured;

[0037] an extraction unit, used to obtain the coordinates of the cursor in the image to be measured, and to extract the horizontal coordinates of the left edge white point and the right edge white point in the vertical edge of the wheel to be measured from the binary image through an iterative algorithm, wherein the left edge white point is the edge point in the vertical left edge of the wheel to be measured with the shortest distance to the coordinates; and the right edge white point is the edge point in the vertical right edge of the wheel to be measured with the shortest distance to the coordinates;

[0038] A first calculation unit is used to calculate the arc angle between the horizontal center of the binary image and the left edge white point, and the right edge white point, respectively, to determine the width of the wheel to be measured;

[0039] A comparison unit, used for comparing the width with the template edge image of the wheel and the guide rail to obtain an offset;

[0040] The first determination unit is used to determine that the wheel to be tested has not shifted if the offset is not greater than a preset value.

[0041] Preferably, the device further comprises:

[0042] An acquisition unit is used to obtain the weight ranking results of each expert on each lateral wheel position measurement factor to construct a ranking matrix and calculate the Kendall consistency coefficient;

[0043] A second calculation unit is used to calculate a chi-square value according to the Kendall consistency coefficient;

[0044] A second determination unit is configured to determine whether the weight ranking results of the experts for each lateral wheel position measurement factor are consistent if the chi-square value is greater than a preset threshold;

[0045] An analysis unit, used to analyze the ranking matrix using a hierarchical analysis method to obtain a hierarchical analysis grade matrix;

[0046] The screening unit is used to screen the thermal imager from a plurality of sensors according to the hierarchical analysis level matrix.

[0047] Preferably, the first computing unit comprises:

[0048] A first calculation module, used for calculating the left distance between the left edge white point and the horizontal center of the image, and the right distance between the right edge white point and the horizontal center of the image;

[0049] A second calculation module, used for respectively calculating a left arc angle between the horizontal center of the image and the white point on the left edge, and a right arc angle between the horizontal center of the image and the white point on the right edge according to the left distance and the right distance;

[0050] The third calculation module is used to calculate the width of the wheel to be measured based on the left arc angle and the right arc angle.

[0051] Preferably, the second calculation module includes:

[0052] An acquisition submodule, used to acquire the horizontal field of view angle of the stereo camera;

[0053] A first calculation submodule, used for calculating the angle of each pixel according to the horizontal field of view angle and the horizontal center of the image;

[0054] A second calculation submodule is used to calculate the product of the angle of each pixel and the left distance to obtain the left arc angle between the horizontal center of the image and the left edge white point;

[0055] The third calculation submodule is used to calculate the product of the angle of each pixel and the right distance to obtain the right arc angle between the horizontal center of the image and the right edge white point.

[0056] A third aspect of the present invention discloses a train wheel position monitoring system, the system comprising a controller and a measuring device;

[0057] The measuring device is arranged at a preset measuring position of the train; the outer surface of the measuring device is provided with a shell;

[0058] The measuring device comprises a stereo camera and a thermal imager; the stereo camera and the thermal imager are respectively connected to the controller;

[0059] The stereo camera is used to capture wheel position images of the train and send the images to the controller;

[0060] The thermal imager is used to measure the wheel position of the train and send it to the controller;

[0061] The controller is used to execute the train wheel position monitoring method disclosed in the first aspect of the present invention.

[0062] Based on the above-mentioned embodiments of the present invention, a train wheel position monitoring method, device and system are provided. The train wheel position is accurately measured by periodically using a stereo camera and a thermal imager, and the measurement image is automatically displayed, which significantly improves the accuracy and robustness of the measurement. When receiving an image selection instruction, the system automatically marks the image to be measured and the wheel to be measured therein, and then processes the image through an edge detector to extract the vertical edge of the wheel, and further uses an iterative algorithm to accurately extract the horizontal coordinates of the edge white point. The wheel width is determined by calculating the arc angle between the horizontal center of the image and the edge white point, and compared with the preset template to obtain the offset, so as to determine whether the wheel is displaced. This method not only reduces the dependence on camera settings, improves operational efficiency, but also reduces implementation and operating costs, while meeting the strict requirements of the railway transportation system for high safety and high efficiency, showing great promotion value. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0064] Figure 1 A flow chart of a train wheel position monitoring method provided by an embodiment of the present invention;

[0065] Figure 2 A flow chart of screening and obtaining a thermal imager provided in an embodiment of the present invention;

[0066] Figure 3 A schematic diagram of the motion characteristics of a wheelset on a train track provided by an embodiment of the present invention;

[0067] Figure 4 A schematic diagram of the movement of a wheelset on a tangent track provided by an embodiment of the present invention;

[0068] Figure 5 A schematic diagram of the wavelength formed on the rail by the periodic lateral motion (oscillation) of the wheel provided in an embodiment of the present invention;

[0069] Figure 6 An example image provided by an embodiment of the present invention;

[0070] Figure 7 An example diagram of a binary image containing a vertical edge of a wheel to be tested provided by an embodiment of the present invention;

[0071] Figure 8 A schematic diagram of determining the width of a wheel to be tested provided by an embodiment of the present invention;

[0072] Fig. 9 A flow chart for determining the width of a wheel to be tested provided by an embodiment of the present invention;

[0073] Fig.10 A flowchart for calculating left radian angle and right radian angle provided by an embodiment of the present invention;

[0074] Fig.11 A schematic diagram of a train wheel position monitoring system provided by an embodiment of the present invention;

[0075] Fig.12 An example diagram of the setting of a measuring device provided in an embodiment of the present invention;

[0076] Fig.13 An example diagram of a housing of a measuring device provided by an embodiment of the present invention;

[0077] Fig.14 A structural block diagram of a train wheel position monitoring device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0078] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0079] In this application, the terms "comprises", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element.

[0080] As can be seen from the background technology, the current method of monitoring the position of train wheels is mainly based on the oscillation principle of accelerometers. In this process, factors such as rail quality, rail gaps, curves, and track geometry will affect the characteristics of the vibration signal, resulting in a decrease in measurement accuracy. In addition, the acceleration sensor needs to be calibrated regularly, which increases the cost and complexity of operation. And the measurement range is limited by factors such as wheel size and running speed.

[0081] Therefore, the embodiments of the present invention provide a method, device and system for monitoring the position of train wheels, which periodically use stereo cameras and thermal imagers to accurately measure the position of train wheels and automatically display the measured images, thereby significantly improving the accuracy and robustness of the measurement. When receiving an image selection instruction, the system automatically marks the image to be measured and the wheel to be measured therein, and then processes the image through an edge detector to extract the vertical edge of the wheel, and further uses an iterative algorithm to accurately extract the horizontal coordinates of the edge white point. The wheel width is determined by calculating the arc angle between the horizontal center of the image and the edge white point, and compared with a preset template to obtain the offset, thereby determining whether the wheel is displaced. This method not only reduces the dependence on camera settings, improves operational efficiency, but also reduces implementation and operating costs, while meeting the strict requirements of the railway transportation system for high safety and high efficiency, showing great promotion value.

[0082] See also Figure 1 , shows a flow chart of a train wheel position monitoring method provided by an embodiment of the present invention, the monitoring method comprising:

[0083] Step S101: periodically measure the position of the train wheels using a stereo camera and a thermal imager, and display the measured images.

[0084] It is understandable that the thermal imager is pre-screened from multiple sensors. The specific screening process is detailed in the embodiment of the present invention. Figures 2 to 5 The content in.

[0085] In the specific implementation of step S101, the lateral position of the train wheel-rail, that is, the position of the train wheel, is periodically measured by a stereo camera (such as a ZED Stereo Camera, or other stereo cameras) and a thermal imager (such as a FLIR Infrared Thermal Imager (FIR)), and the measured image is displayed on a computer display so that the user can select the wheel to be measured.

[0086] It should be noted that the measured image is a visual image with stereo depth information. Figure 6 An example of an image is shown in which a user can select an object to measure the width of, i.e., a wheel. Figure 6 In the displayed content, the cross mark indicates the position of the mouse cursor.

[0087] In some specific embodiments, if the depth image measured by the stereo camera and the thermal imager is available, the distance between the cursor selected position and the stereo camera lens is displayed next to the mouse cursor.

[0088] Step S102: when an image selection instruction is received, the image corresponding to the selection instruction is marked as the image to be measured, and the object selected by the selection instruction in the image to be measured is marked as the wheel to be measured.

[0089] In the specific implementation of step S102, after receiving an image selection instruction triggered by a user from a computer display, the image corresponding to the selection instruction is used as the image to be measured; and the object selected by the user in the image to be measured is marked as the wheel to be measured; so as to subsequently measure the width of the wheel to be measured in the image to be measured, thereby determining whether the wheel to be measured is offset.

[0090] Step S103: Processing the image to be measured by an edge detector to obtain a binary image containing the vertical edge of the wheel to be measured.

[0091] In the specific implementation of step S103, the image to be measured selected by the user is processed using an edge detector method (Canny edge detector) to obtain a binary image containing the vertical edge of the wheel to be measured, wherein the binary image is, for example, Figure 7 shown.

[0092] It should be noted that the Canny edge detector is an edge detection algorithm that can identify points in an image where the intensity changes significantly, i.e., edges. These edges usually correspond to the outlines of objects in the image or the boundaries between different areas. A binary image is an edge image, which contains only edge pixels (usually white) and non-edge pixels (usually black).

[0093] Step S104: obtaining the coordinates of the cursor in the image to be measured, and extracting the horizontal coordinates of the left edge white point and the right edge white point in the vertical edge of the wheel to be measured from the binary image through an iterative algorithm.

[0094] It should be noted that the left edge white point is the edge point with the shortest distance coordinate among the vertical left edge of the wheel to be tested; the right edge white point is the edge point with the shortest distance coordinate among the vertical right edge of the wheel to be tested.

[0095] In the specific implementation of step S104, the point selected by the user in the image to be measured, that is, the coordinates of the mouse cursor, is obtained; then, through an iterative algorithm, the edges in the binary image are searched on both sides of the coordinates, thereby extracting the horizontal coordinates of the left edge white point and the right edge white point in the vertical edge of the wheel to be measured.

[0096] It can be understood that the pixels on both sides of the cursor coordinates on the binary image are checked one by one through an iterative algorithm until an edge pixel (white pixel) is found, thereby obtaining the horizontal coordinates of the left edge white point and the right edge white point.

[0097] Step S105: respectively calculating the arc angle between the horizontal center of the binary image and the white point on the left edge, and the white point on the right edge, to determine the width of the wheel to be measured.

[0098] In the specific implementation of step S105, Figure 8 A schematic diagram for determining the width of the wheel to be measured is shown, in which the left distance between the left edge white point and the horizontal center of the image is calculated according to the horizontal coordinate of the left edge white point; the right distance between the right edge white point and the horizontal center of the image is calculated according to the horizontal coordinate of the right edge white point; the left arc angle and the right arc angle are calculated based on the left distance and the right distance, and thus the width of the wheel to be measured is calculated based on the left arc angle and the right arc angle.

[0099] It should be noted that the specific process of determining the width of the wheel to be tested is detailed in the embodiment of the present invention. Fig. 9 The content in Fig. 9 include:

[0100] Step S901: Calculate the left distance between the left edge white point and the horizontal center of the image, and the right distance between the right edge white point and the horizontal center of the image.

[0101] In the specific implementation process of step S901, the left distance between the left edge white point and the horizontal center of the image is calculated according to the horizontal coordinate of the left edge white point, as shown in formula (1); the right distance between the right edge white point and the horizontal center of the image is calculated according to the horizontal coordinate of the right edge white point, as shown in formula (2).

[0102]

[0103] Among them, d xL Indicates the left distance between the left edge white point and the horizontal center of the image; x L Indicates the horizontal coordinate of the white point on the left edge; w I Indicates the width of the horizontal center of the image (specifically, the image width in points).

[0104]

[0105] Among them, d xR Indicates the right distance between the right edge white point and the horizontal center of the image; x R Indicates the horizontal coordinate of the white point on the right edge; w I Indicates the width of the horizontal center of the image (specifically, the image width in points).

[0106] Step S902: Calculate the left arc angle between the horizontal center of the image and the white point on the left edge, and the right arc angle between the horizontal center of the image and the white point on the right edge, respectively, based on the left distance and the right distance.

[0107] In the process of implementing step S902, the left arc angle between the horizontal center of the image and the white point on the left edge is calculated based on the left distance between the white point on the left edge and the horizontal center of the image; the right arc angle between the horizontal center of the image and the white point on the right edge is calculated based on the right distance between the white point on the right edge and the horizontal center of the image.

[0108] It should be noted that the specific process of calculating the left arc angle and the right arc angle is detailed in the embodiment of the present invention. Fig.10 The content in Fig.10 include:

[0109] Step S1001: Obtain the horizontal field of view angle of the stereo camera.

[0110] Step S1002: Calculate the angle of each pixel according to the horizontal field of view angle and the horizontal center of the image.

[0111] In the specific implementation of step S1002, the angle of each pixel is calculated according to the horizontal field angle of the stereo camera and the horizontal center of the image, as shown in formula (3).

[0112] aPIX=aHFOV / w I (3)

[0113] Among them, aPIX represents the angle of each pixel, and aHFOV represents the horizontal field of view angle of the stereo camera.

[0114] Step S1003: Calculate the product of the angle of each pixel and the left distance to obtain the left arc angle between the horizontal center of the image and the white point on the left edge.

[0115] It should be noted that the process of calculating the left arc angle is shown in formula (4).

[0116] ∠XLC=d xL ·aPIX (4)

[0117] Where ∠XLC represents the left arc angle between the horizontal center of the image and the white point on the left edge; d xL It indicates the left distance between the left edge white point and the horizontal center of the image; aPIX indicates the angle of each pixel.

[0118] Step S1004: Calculate the product of the angle of each pixel and the right distance to obtain the right arc angle between the horizontal center of the image and the white point on the right edge.

[0119] It should be noted that the process of calculating the right arc angle is shown in formula (5).

[0120] ∠XRC=d xR ·aPIX (5)

[0121] Where ∠XRC represents the right arc angle between the horizontal center of the image and the white point on the right edge; d xR It indicates the right distance between the right edge white point and the horizontal center of the image; aPIX indicates the angle of each pixel.

[0122] Step S903: Calculate the width of the wheel to be measured based on the left arc angle and the right arc angle.

[0123] In the specific implementation of step S903, the left end distance between the left end of the wheel to be measured and the horizontal center of the image is calculated based on the left arc angle and the right arc angle, as shown in formula (6), and the right end distance between the right end of the wheel to be measured and the horizontal center of the image is calculated, as shown in formula (7); then the modulus of the distance difference between the left end distance and the right end distance is calculated to obtain the width of the wheel to be measured, as shown in formula (8).

[0124] d xLmm =d C2OX ·sin(∠XLC) (6)

[0125] d xRmm =d C2OX ·sin(∠XRC) (7)

[0126] d xRxLmm =|d xRmm -d xLmm | (8)

[0127] Among them, d xRmm Indicates the left end distance; d xRmm Indicates the right end distance; d C2OX Indicates the distance from the stereo camera to the x-plane where the object is located; d xRxLmm Indicates the width of the wheel to be tested.

[0128] Step S106: Compare the width with the template edge image of the wheel and the guide rail to obtain the offset.

[0129] It can be understood that the template edge image of the wheel and the guide rail is a reference image, and the offset of the wheel to be tested is obtained by comparing the width of the wheel to be tested with the reference image.

[0130] Step S107: If the offset is not greater than the preset value, it is determined that the wheel to be tested is not displaced.

[0131] It can be understood that if the offset of the wheel to be tested is greater than a preset value, it is determined that the wheel to be tested is displaced.

[0132] In the embodiment of the present invention, the position of the train wheels is accurately measured by periodically using a stereo camera and a thermal imager, and the measured images are automatically displayed, which significantly improves the accuracy and robustness of the measurement. When receiving an image selection instruction, the system automatically marks the image to be measured and the wheel to be measured therein, and then processes the image through an edge detector to extract the vertical edge of the wheel, and further uses an iterative algorithm to accurately extract the horizontal coordinates of the edge white point. The wheel width is determined by calculating the arc angle between the horizontal center of the image and the edge white point, and compared with the preset template to obtain the offset, thereby determining whether the wheel is displaced. This method not only reduces the dependence on camera settings, improves operational efficiency, but also reduces implementation and operating costs, while meeting the strict requirements of the railway transportation system for high safety and high efficiency, showing great promotion value.

[0133] The above embodiments of the present invention Figure 1 For the specific implementation method of pre-screening a thermal imager from multiple sensors, see Figure 2 , shows a flow chart of screening and obtaining thermal imagers provided by an embodiment of the present invention, including:

[0134] It is understandable that before determining the method of monitoring the position of train wheels, the motion characteristics of the wheelset on the train track need to be studied and evaluated in depth. Figure 3 As shown in Figure 1, when the train runs along the track, the wheelset not only moves linearly along the longitudinal axis of the track, but also oscillates along the transverse axis. This phenomenon is called Klingel wheelset motion, which manifests as a sinusoidal lateral oscillation.

[0135] During operation, the wheelset will continue to move closer to the center of the track. Figure 4 The schematic diagram of the motion of the wheelset on the tangent track shows that the wheels not only undergo lateral displacement, but also rotate around the vertical axis (Z axis) and the horizontal axis (Y axis). During the oscillation of the wheelset, longitudinal and lateral creep forces are generated between the wheels and the track, and a creep moment around the Z axis is also formed.

[0136] Based on this motion characteristic, Figure 5 A schematic diagram showing the wavelengths formed on the rail by the periodic lateral motion (oscillation) of the wheel. These oscillations form ripples on the rail surface, which have a significant impact on the wear and stability of the rail and wheels.

[0137] It can be seen that the wheel movement is correlated with multiple factors. Therefore, the embodiment of the present invention adopts a multi-criteria decision-making technology to ultimately select the most suitable thermal imager as the sensor used in the embodiment of the present invention.

[0138] Step S201: Obtain the weight ranking results of each expert for each lateral wheel position measurement factor to construct a ranking matrix, and calculate the Kendall consistency coefficient.

[0139] It is understandable that the embodiment of the present invention collects the weight ranking results of the importance of the following lateral wheel position measurement factors from multiple experts in advance through a questionnaire survey method. The lateral wheel position measurement factors include but are not limited to:

[0140] Size and weight: including the system's geometric parameters, weight, installation method, etc.

[0141] Energy consumption: involves aspects such as the system's energy efficiency and input voltage.

[0142] Life Cycle Cost (LCC): includes factors such as system price, maintenance costs, long life, durability and reliability.

[0143] System robustness: including resistance to vibration, shock, high temperature, dust, and interference from railway signaling and automation systems, and adaptability to various environmental conditions (such as rain, snow, mud, sleet, fog, etc.).

[0144] Measurement accuracy: including sampling rate, etc., and the accuracy of the system depends on the train speed.

[0145] Technical compatibility (interoperability): includes the possibility of installing the system under different operating conditions, as well as compatibility with different instruments, signal / automation systems and interfaces.

[0146] Output data: This includes the type, format, amount of data and additional data (such as location, time, date, visualization data, etc.) of the output data, as well as the complexity of the final parameter estimate derived from it.

[0147] Repeatability of measurement: refers to the reliability of the results, ensuring the consistency of the measurement results.

[0148] It should be noted that if certain lateral wheel position measurement factors are of equal importance in the opinion of the experts, they will be given the same weight, which is the arithmetic mean of the respective weights of these lateral wheel position measurement factors.

[0149] It can be understood that the weight ranking results of these factors will provide an important basis for the determination of sensors.

[0150] It should be noted that in order to evaluate whether the evaluations of various experts on various lateral wheel position measurement factors are consistent, the consistency coefficient is used for evaluation.

[0151] In the process of implementing step S201, the weight ranking results of each expert for each lateral wheel position measurement factor are obtained, that is, each expert ranks each lateral wheel position measurement factor, wherein the most important factor has a rank of 1, and the second most important factor has a rank of 2, decreasing step by step; then a ranking matrix is ​​constructed based on these ranking results, and the Kendall consistency coefficient is calculated, and the consistency level of the expert evaluation is quantified by the Kendall consistency coefficient.

[0152] It should be noted that the Kendall coefficient of consistency is related to the sum of the levels of various lateral wheel position measurement factors, and it can help determine whether the experts' estimates remain consistent.

[0153] It can be understood that the ranking matrix constructed based on the ranking results is shown in Table 1, and the ranking matrix contains the ranking information of each expert on each lateral wheel position measurement factor.

[0154] Table 1

[0155]

[0156] It should be noted that when the number of lateral wheel position measurement factors is greater than the preset value (such as 7), the chi-square distribution is used to determine the significance of the Kendall consistency coefficient.

[0157] Step S202: Calculate the chi-square value according to the Kendall coefficient of consistency.

[0158] It should be noted that in order to determine the significance of the Kendall consistency coefficient, a specific significance level (such as 0.05) is first set, and the boundary value of the Kendall consistency coefficient is determined based on this level. When the number of lateral wheel position measurement factors is greater than the preset value (such as 7), the chi-square (x 2 ) distribution to test the significance of Kendall's coefficient of consistency.

[0159] In the specific implementation of step S202, the chi-square value (x 2 ), as shown in formula (9).

[0160]

[0161] Among them, x 2 represents the chi-square value, W represents the Kendall consistency coefficient; m represents the total number of lateral wheel position measurement factors; r represents the number of experts; S represents the sum of the levels of lateral wheel position measurement factors.

[0162] Step S203: If the chi-square value is greater than the preset threshold, it is determined that the weight ranking results of the experts for each lateral wheel position measurement factor are consistent.

[0163] Specifically, if the calculated chi-square value x 2 If it is greater than the critical value under the corresponding degrees of freedom and significance level, it can be considered that the expert's judgment is consistent at the 95% confidence level.

[0164] Step S204: Analyze the ranking matrix using the hierarchical analysis method to obtain a hierarchical analysis grade matrix.

[0165] It can be understood that the analytic hierarchy process (AHP) is a decision-making method based on a pairwise comparison matrix. In this matrix, each element pij (where i and j range from 1 to m) represents the relative importance between two lateral wheel position measurement factors Ri and Rj. Here m is the total number of lateral wheel position measurement factors. Experts determine the relative importance between these lateral wheel position measurement factors by comparing all possible pairs of lateral wheel position measurement factors. Ideally, the elements of this pairwise comparison matrix represent the relationship between the unknown criterion weights, as shown in formula (10):

[0166]

[0167] Where ω represents the importance of different lateral wheel position measurement factors, p ij Indicates the importance of factor 'i' to factor 'j'.

[0168] It should be noted that this comparison is qualitative and easy to perform, and it shows whether one lateral wheel position measurement factor is more important than another lateral wheel position measurement factor and at what level of priority it belongs. This technique allows the qualitative estimate obtained from the expert to be converted into a quantitative estimate.

[0169] It can be understood that the pairwise comparison matrix is ​​an inverse symmetric (reciprocal) matrix, for example:

[0170]

[0171] The main principle for filling the matrix is ​​simple, as the expert should indicate how much more important a particular lateral wheel position measurement factor is than another lateral wheel position measurement factor.

[0172] It can be understood that the normalized subjective weight or relative importance is calculated as shown in formula (11):

[0173]

[0174] Among them, W i Represents supervisor weight.

[0175] Multiply the comparison matrix and the weight vector pairwise to calculate the eigenvalue: Pw = λMCDMwi.

[0176] Among them, λMCDM is the comparison matrix.

[0177] The weight vector ω in the method is the normalized component of the eigenvector, corresponding to the maximum computable eigenvalue, as shown in formula (12):

[0178] det A=P-λMCDME=0 (12)

[0179] Wherein, det A represents the maximum eigenvalue; P represents the inverse symmetric (reciprocal) matrix mentioned above; λMCDME represents the comparison matrix.

[0180] The AHP method uses the consistency index to evaluate the consistency of each expert's estimate (as shown in formula (13)):

[0181]

[0182] Where CI represents the consistency result; λMCDM,max represents the maximum eigenvalue of the matrix; and m represents the order of the matrix.

[0183] The relationship between the calculated consistency index is: I. A specific matrix and the random consistency index RI (Table 3) is called the consistency relationship. It determines the consistency degree of the matrix (as shown in formula (14)):

[0184]

[0185] Among them, CR represents the consistency degree of the matrix.

[0186] A value of matrix consistency R. less than or equal to 0.1 is acceptable, which means that the matrix is ​​consistent.

[0187] The above AHP method was used to rank the experts and perform calculations. The consistency coefficient was calculated and the estimated value was W = 0.805. The value of parameter 2 was 67.61, which was greater than the critical value 2kr = 14.067. The degrees of freedom ν = 8-1 = 7, and the importance α = 0.05. Using the ranking of one expert as a reference, the hierarchical analysis matrix was developed (as shown in Table 2).

[0188] Table 2

[0189] 1 2 3 4 5 6 7 8 1 1 05 0.14 0.11 0.13 0.17 2 0.2 2 2 1 0.2 0.11 0.14 0.33 3 0.5 3 7 5 1 0.2 0.33 2 5 3 4 9 9 5 1 2 3 7 5 5 8 7 3 0.5 1 3 7 5 6 6 3 0.5 0.33 0.33 1 5 2 7 0.5 0.33 0.2 0.14 0.14 0.2 1 0.2 8 5 2 0.33 0.2 0.2 0.5 5 1

[0190] Step S205: Filter and obtain a thermal imager from a plurality of sensors according to the hierarchical analysis level matrix.

[0191] In the specific implementation of step S205, the most suitable sensor is comprehensively considered from multiple angles according to the hierarchical analysis level matrix, so as to select a thermal imager from multiple sensors.

[0192] In the embodiment of the present invention, by constructing a ranking matrix and calculating the Kendall consistency coefficient, the method ensures the consistency of the expert's ranking results of the weights of the lateral wheel position measurement factors, and improves the objectivity and reliability of the decision. The ranking matrix is ​​further analyzed using the hierarchical analysis method to scientifically select thermal imagers as the optimal sensors for monitoring the lateral wheel position, which not only improves the monitoring efficiency and accuracy, but also reduces costs and enhances the adaptability and flexibility of the system. In addition, the transparency and fairness of the method are also enhanced because each step of the decision-making process is based on clear calculations and evaluations, thereby promoting the scientific and rational decision-making.

[0193] See also Fig.11 An embodiment of the present invention further provides a train wheel position monitoring system, which includes a controller 1101 and a measuring device 1102.

[0194] Specifically, the measuring device 1102 is set at a preset measuring position on the train, for example Fig.12 The rectangular frame shown.

[0195] It should be noted that the specific location of the measuring device 1102 needs to be determined according to the actual situation. Fig.12 The positions shown are for illustrative purposes only and are not intended to limit specific setting positions.

[0196] Specifically, Fig.13 As shown, the outer surface of the measuring device 1102 is provided with a housing.

[0197] It is understood that the housing can effectively protect the measuring device 1102 from being affected by environmental conditions and particle emissions. If the measuring device 1102 is covered with dust, water, mud or snow, measurement may not be performed.

[0198] It should be noted that this shell structure requires the introduction of additional airflow to increase its internal pressure and thus enhance the durability of the entire system. Fig.13 As shown. During the design, the smallest size was chosen in order to reduce the overall volume of the measuring device and minimize air consumption. This design takes into account both the protection needs of the equipment and the optimization of energy efficiency and space.

[0199] However, this housing design may limit the camera's field of view. To address this issue, the present invention selected aluminum as the housing material because aluminum is not only suitable for the production of the final product, but also has other advantages. Considering that the reflection on the surface of the orifice plate may have a negative impact on the system performance, it was determined after experiments that as long as the distance between the back wall of the housing and the camera surface is kept within 17 mm, the system can work normally and the performance will not be significantly affected. This design not only ensures the camera's field of view, but also takes into account the stability and reliability of the system.

[0200] Specifically, the measuring device 1102 includes a stereo camera and a thermal imager; wherein the stereo camera and the thermal imager are respectively connected to the controller 1101 .

[0201] It can be understood that the stereo camera is used to take images of the wheel positions of the train and send them to the controller 1101; the thermal imager is used to measure the wheel positions of the train and send them to the controller 1101. The controller 1101 is used to execute the embodiment of the present invention. Figure 1 A train wheel position monitoring method is provided.

[0202] In the embodiment of the present invention, a contactless monitoring method is adopted, which can meet the high requirements of the railway transportation system for wheel position monitoring. The design scheme has strong portability and modifiability, is low-cost, easy to implement, and can well meet the actual needs of the site. By improving the accuracy and robustness of the measurement, the present invention reduces the dependence on camera settings, thereby improving the safety and operational efficiency of railway transportation. Therefore, the present invention is not only close to the actual needs of the site, but also has great promotion value, which is of great significance for ensuring the safety of railway transportation and improving operational efficiency.

[0203] Corresponding to the train wheel position monitoring method provided by the above embodiment of the present invention, see Fig.14 , shows a structural block diagram of a train wheel position monitoring device provided by an embodiment of the present invention.

[0204] Specifically, the train wheel position monitoring device includes: a measuring unit 1401 , a marking unit 1402 , a processing unit 1403 , an extracting unit 1404 , a first calculating unit 1405 , a comparing unit 1406 and a first determining unit 1407 .

[0205] The measuring unit 1401 is used to periodically measure the position of the train wheels using a stereo camera and a thermal imager, and display the measured images; wherein the thermal imager is pre-selected from multiple sensors.

[0206] The marking unit 1402 is used for marking the image corresponding to the selection instruction as the image to be measured, and marking the object selected by the selection instruction in the image to be measured as the wheel to be measured when receiving the image selection instruction.

[0207] The processing unit 1403 is used to process the image to be measured by an edge detector to obtain a binary image containing the vertical edge of the wheel to be measured.

[0208] The extraction unit 1404 is used to obtain the coordinates of the cursor in the image to be measured, and extract the horizontal coordinates of the left edge white point and the right edge white point in the vertical edge of the wheel to be measured from the binary image through an iterative algorithm, wherein the left edge white point is the edge point with the shortest distance coordinate in the vertical left edge of the wheel to be measured; and the right edge white point is the edge point with the shortest distance coordinate in the vertical right edge of the wheel to be measured.

[0209] The first calculation unit 1405 is used to calculate the arc angle between the horizontal center and the left edge white point of the binary image, and the right edge white point, to determine the width of the wheel to be measured.

[0210] The comparison unit 1406 is used to compare the width with the template edge image of the wheel and the guide rail to obtain the offset.

[0211] The first determining unit 1407 is configured to determine that the wheel to be tested has not shifted if the offset is not greater than a preset value.

[0212] In the embodiment of the present invention, the position of the train wheels is accurately measured by periodically using a stereo camera and a thermal imager, and the measured images are automatically displayed, which significantly improves the accuracy and robustness of the measurement. When receiving an image selection instruction, the system automatically marks the image to be measured and the wheel to be measured therein, and then processes the image through an edge detector to extract the vertical edge of the wheel, and further uses an iterative algorithm to accurately extract the horizontal coordinates of the edge white point. The wheel width is determined by calculating the arc angle between the horizontal center of the image and the edge white point, and compared with the preset template to obtain the offset, thereby determining whether the wheel is displaced. This method not only reduces the dependence on camera settings, improves operational efficiency, but also reduces implementation and operating costs, while meeting the strict requirements of the railway transportation system for high safety and high efficiency, showing great promotion value.

[0213] Combination Fig.14 According to the content shown, the device also includes: an acquisition unit, a second calculation unit, a second determination unit, an analysis unit and a screening unit.

[0214] The acquisition unit is used to obtain the weight ranking results of each expert on each lateral wheel position measurement factor to construct a ranking matrix and calculate the Kendall consistency coefficient.

[0215] The second calculation unit is used to calculate the chi-square value according to the Kendall consistency coefficient.

[0216] The second determining unit is used to determine whether the weight ranking results of each expert on each lateral wheel position measurement factor are consistent if the chi-square value is greater than a preset threshold.

[0217] The analysis unit is used to analyze the ranking matrix using the hierarchical analysis method to obtain a hierarchical analysis grade matrix.

[0218] The screening unit is used for screening thermal imagers from a plurality of sensors according to a hierarchical analysis level matrix.

[0219] Combination Fig.14 As shown in the figure, the first computing unit 1405 includes: a first computing module, a second computing module and a third computing module.

[0220] The first calculation module is used to calculate the left distance between the left edge white point and the horizontal center of the image, and the right distance between the right edge white point and the horizontal center of the image.

[0221] The second calculation module is used to calculate the left arc angle between the horizontal center of the image and the white point on the left edge, and the right arc angle between the horizontal center of the image and the white point on the right edge according to the left distance and the right distance.

[0222] The third calculation module is used to calculate the width of the wheel to be measured based on the left arc angle and the right arc angle.

[0223] The third calculation module is specifically used to calculate the left end distance between the left end of the wheel to be measured and the horizontal center of the image, and the right end distance between the right end of the wheel to be measured and the horizontal center of the image based on the left arc angle and the right arc angle; calculate the modulus of the distance difference between the left end distance and the right end distance to obtain the width of the wheel to be measured.

[0224] Combination Fig.14 The content shown, the second calculation module, includes: an acquisition submodule, a first calculation submodule, a second calculation submodule and a third calculation submodule.

[0225] The acquisition submodule is used to obtain the horizontal field of view angle of the stereo camera.

[0226] The first calculation submodule is used to calculate the angle of each pixel according to the horizontal field of view angle and the horizontal center of the image.

[0227] The second calculation submodule is used to calculate the product of the angle of each pixel and the left distance to obtain the left arc angle between the horizontal center of the image and the white point on the left edge.

[0228] The third calculation submodule is used to calculate the product of the angle of each pixel and the right distance to obtain the right arc angle between the horizontal center of the image and the right edge white point.

[0229] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can refer to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiment. The system and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without creative work.

[0230] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0231] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for monitoring the position of train wheels, characterized in that: The method comprises: Periodically measuring the position of the train wheels using a stereo camera and a thermal imager, and displaying the measured images; wherein the thermal imager is pre-selected from a plurality of sensors; When an image selection instruction is received, the image corresponding to the selection instruction is marked as the image to be measured, and the object selected by the selection instruction in the image to be measured is marked as the wheel to be measured; Processing the image to be measured by an edge detector to obtain a binary image including a vertical edge of the wheel to be measured; Acquire the coordinates of the cursor in the image to be measured, and extract the horizontal coordinates of the left edge white point and the right edge white point in the vertical edge of the wheel to be measured from the binary image through an iterative algorithm, wherein the left edge white point is the edge point in the vertical left edge of the wheel to be measured with the shortest distance to the coordinates; the right edge white point is the edge point in the vertical right edge of the wheel to be measured with the shortest distance to the coordinates; Calculating the arc angles between the horizontal center of the binary image and the left edge white point, and the right edge white point, respectively, to determine the width of the wheel to be tested; Comparing the width with the template edge images of the wheel and rail to obtain an offset; If the offset is not greater than the preset value, it is determined that the wheel to be tested has not shifted.

2. The method according to claim 1, characterized in that The process of selecting a thermal imager from multiple sensors includes: Obtain the weight ranking results of each expert on each lateral wheel position measurement factor to construct a ranking matrix and calculate the Kendall consistency coefficient; Chi-square values ​​were calculated based on the Kendall coefficient of concordance described; If the chi-square value is greater than a preset threshold, it is determined that the weight ranking results of the experts for each lateral wheel position measurement factor are consistent; Analyzing the ranking matrix using the analytic hierarchy process to obtain an analytic hierarchy level matrix; A thermal imager is obtained by screening from a plurality of sensors according to the hierarchical analysis level matrix.

3. The method according to claim 1, characterized in that The step of respectively calculating the arc angle between the horizontal center of the binary image and the left edge white point, and the right edge white point to determine the width of the wheel to be measured comprises: Calculate the left distance between the left edge white point and the horizontal center of the image, and the right distance between the right edge white point and the horizontal center of the image; According to the left distance and the right distance, respectively calculate the left arc angle between the horizontal center of the image and the white point on the left edge, and the right arc angle between the horizontal center of the image and the white point on the right edge; The width of the wheel to be measured is calculated based on the left arc angle and the right arc angle.

4. The method according to claim 3, characterized in that The step of calculating the left arc angle between the horizontal center of the image and the white point on the left edge, and the right arc angle between the horizontal center of the image and the white point on the right edge, respectively, according to the left distance and the right distance comprises: Obtaining the horizontal field of view angle of the stereo camera; Calculate the angle of each pixel according to the horizontal field of view angle and the horizontal center of the image; Calculate the product of the angle of each pixel and the left distance to obtain the left arc angle between the horizontal center of the image and the left edge white point; The product of the angle of each pixel and the right distance is calculated to obtain the right arc angle between the horizontal center of the image and the right edge white point.

5. The method according to claim 3, characterized in that: The calculating the width of the wheel to be measured based on the left arc angle and the right arc angle includes: Calculating a left end distance between a left end of the wheel to be measured and a horizontal center of the image, and a right end distance between a right end of the wheel to be measured and a horizontal center of the image based on the left arc angle and the right arc angle; The modulus of the distance difference between the left end distance and the right end distance is calculated to obtain the width of the wheel to be measured.

6. A train wheel position monitoring device, characterized in that: The device comprises: A measuring unit, used for periodically measuring the position of the train wheels using a stereo camera and a thermal imager, and displaying the measured images; wherein the thermal imager is pre-selected from a plurality of sensors; a marking unit, configured to, when receiving an image selection instruction, mark the image corresponding to the selection instruction as the image to be measured, and mark the object selected by the selection instruction in the image to be measured as the wheel to be measured; A processing unit, configured to process the image to be measured by an edge detector to obtain a binary image including a vertical edge of the wheel to be measured; an extraction unit, used to obtain the coordinates of the cursor in the image to be measured, and to extract the horizontal coordinates of the left edge white point and the right edge white point in the vertical edge of the wheel to be measured from the binary image through an iterative algorithm, wherein the left edge white point is the edge point in the vertical left edge of the wheel to be measured with the shortest distance to the coordinates; and the right edge white point is the edge point in the vertical right edge of the wheel to be measured with the shortest distance to the coordinates; A first calculation unit is used to calculate the arc angle between the horizontal center of the binary image and the left edge white point, and the right edge white point, respectively, to determine the width of the wheel to be measured; A comparison unit, used for comparing the width with the template edge image of the wheel and the guide rail to obtain an offset; The first determination unit is used to determine that the wheel to be tested has not shifted if the offset is not greater than a preset value.

7. The device according to claim 6, characterized in that The device also includes: An acquisition unit is used to obtain the weight ranking results of each expert on each lateral wheel position measurement factor to construct a ranking matrix and calculate the Kendall consistency coefficient; A second calculation unit is used to calculate a chi-square value according to the Kendall consistency coefficient; A second determination unit is configured to determine whether the weight ranking results of the experts for each lateral wheel position measurement factor are consistent if the chi-square value is greater than a preset threshold; An analysis unit, used to analyze the ranking matrix using a hierarchical analysis method to obtain a hierarchical analysis grade matrix; The screening unit is used to screen the thermal imager from a plurality of sensors according to the hierarchical analysis level matrix.

8. The device according to claim 6, characterized in that The first computing unit comprises: A first calculation module, used for calculating the left distance between the left edge white point and the horizontal center of the image, and the right distance between the right edge white point and the horizontal center of the image; A second calculation module, used for respectively calculating a left arc angle between the horizontal center of the image and the white point on the left edge, and a right arc angle between the horizontal center of the image and the white point on the right edge according to the left distance and the right distance; The third calculation module is used to calculate the width of the wheel to be measured based on the left arc angle and the right arc angle.

9. The device according to claim 8, characterized in that The second calculation module includes: An acquisition submodule, used to acquire the horizontal field of view angle of the stereo camera; A first calculation submodule, used for calculating the angle of each pixel according to the horizontal field of view angle and the horizontal center of the image; A second calculation submodule is used to calculate the product of the angle of each pixel and the left distance to obtain the left arc angle between the horizontal center of the image and the left edge white point; The third calculation submodule is used to calculate the product of the angle of each pixel and the right distance to obtain the right arc angle between the horizontal center of the image and the right edge white point.

10. A train wheel position monitoring system, characterized in that: The system includes a controller and a measuring device; The measuring device is arranged at a preset measuring position of the train; the outer surface of the measuring device is provided with a shell; The measuring device comprises a stereo camera and a thermal imager; the stereo camera and the thermal imager are respectively connected to the controller; The stereo camera is used to capture wheel position images of the train and send the images to the controller; The thermal imager is used to measure the wheel position of the train and send it to the controller; The controller is used to execute the train wheel position monitoring method as described in any one of claims 1-5.

Images