Railway vehicle brake pad thickness detection device and method based on structured light camera

Abstract

A kind of railway vehicle brake pad thickness detection device and method based on structured light camera.The present application relates to the field of railway train detection technology.The present application is to solve the problem that existing dynamic detection equipment is difficult to realize the detection of railway vehicle brake pad thickness.The detection device of the present application includes imaging assembly, control assembly and industrial computer, the imaging assembly is fixedly connected in the middle of track group, the imaging assembly includes 3D structured light camera, the control assembly is fixedly connected on the side of track group in front of imaging assembly, the industrial computer is arranged on the outside of track group, the control assembly and imaging assembly are electrically connected with industrial computer respectively, the industrial computer receives the vehicle signal sent by control assembly, controls imaging assembly to start photographing, displays the generated picture and carries out matching to obtain spatial coordinates, and the spatial coordinates are used for operation.The present application is used for railway vehicle brake pad thickness detection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of railway train detection, in particular to a railway vehicle brake pad thickness detection device and method based on a structured light camera. BACKGROUND

[0002] At present, dynamic detection equipment in the railway industry, such as TFDS (Railway Freight Car Dynamic Detection System) and the like, can only detect non-quantifiable faults, such as fracture, bolt shedding, foreign matter attachment and the like. Some detection items that need to be quantified, such as brake pad thickness and the like, cannot be detected. The main reason for the inability to detect is that the imaging camera does not have the function of quantitative detection, and the high speed and complex environment cause the imaging quality to be not high. At present, 2D line scanning cameras are mainly installed on the line, and such cameras cannot perform data measurement; some devices also install 3D cameras, which can generate 3D point cloud data for measurement, but due to the high speed and poor shooting environment, the 3D point cloud data has too many noise points, and the data cannot be directly used. SUMMARY

[0003] The present application is to solve the problem that the existing dynamic detection equipment cannot realize the detection of the brake pad thickness of the railway vehicle, and further proposes a railway vehicle brake pad thickness detection device and method based on a structured light camera.

[0004] The technical solution adopted by the present application to solve the above technical problem is:

[0005] A railway vehicle brake pad thickness detection device based on a structured light camera comprises an imaging assembly, a control assembly and an industrial computer, the imaging assembly is fixedly connected to the middle part of a track group, the imaging assembly comprises a 3D structured light camera, the control assembly is fixedly connected to one side of the track group in front of the imaging assembly, the industrial computer is arranged on the outside of the track group, the control assembly and the imaging assembly are respectively electrically connected with the industrial computer, the industrial computer receives a vehicle signal sent by the control assembly, controls the imaging assembly to start photographing, displays and saves the generated picture, matches the picture to obtain a spatial coordinate, and performs operation by using the spatial coordinate.

[0006] Further, when the industrial computer is working, the control assembly detects that a vehicle is passing, sends a signal to the industrial computer, and the industrial computer controls the imaging assembly to start and take pictures to generate a 3D point cloud picture and a plane picture;

[0007] The industrial computer identifies a boundary line of the brake pad from the plane picture, takes a plurality of points P on the boundary line, and obtains a plane coordinate P(x, y) of the point P on the plane picture with the boundary intersection point of the plane picture as the origin, according to the one-to-one correspondence of the positions on the 3D point cloud picture and the plane picture, the spatial coordinate P(a, b, c) of the brake pad boundary line with the camera as the origin can be obtained at the position P(x, y) on the 3D point cloud picture;

[0008] Repeat the above process to get all the spatial coordinates of the boundary line of the brake pad, remove the coordinates with noise, and leave the points with clear spatial coordinates. The thickness of the brake pad is calculated using the spatial coordinates P(a, b, c).

[0009] Further, the imaging assembly further comprises a laser generator, a prism, a mounting frame and a lens, the 3D structured light camera, the laser generator and the prism are fixed on the mounting frame, the lens is arranged in front of the 3D structured light camera, the laser generator is arranged on one side of the 3D structured light camera, and the prism is arranged in front of the laser generator.

[0010] Further, the outer side of the imaging assembly is provided with a protection assembly.

[0011] Further, the protection assembly comprises a box body, a box cover and a driving mechanism, the upper end of the box body is open, the box cover is arranged outside the open end of the box body, the imaging assembly is arranged towards the open end of the box body, the box cover is translated on the open end of the box body through the driving mechanism, and the driving mechanism is electrically connected with the industrial computer.

[0012] Further, the driving mechanism comprises an electric push rod and two groups of rails, the electric push rod is horizontally fixed in the box body, a connecting block is fixed at the rod end of the electric push rod, the upper end surface of the connecting block is fixed with the inner side end surface of the box cover, and one group of rails is horizontally arranged on the two sides of the open end of the box body, and the two sides of the box cover are respectively slidably connected with the rails.

[0013] Further, the control assembly comprises a vehicle sensor and a sensor support, the vehicle sensor is fixed at the upper end of the sensor support, and the sensor support is fixed on one side of the rail group.

[0014] A railway vehicle brake pad thickness detection method based on a structured light camera comprises the following steps:

[0015] During the train running process, the train enters the working section of the detection device, first approaches the control assembly, the vehicle sensor in the control assembly detects the train passing, sends a signal to the industrial computer, the industrial computer controls the detection device to start working, at this time, the box cover of the protection assembly is opened, and the imaging assembly starts working, when the train passes through the imaging assembly, the laser generator and the prism in the imaging assembly supplement light to the brake pad, the 3D structured light camera takes a photo of the brake pad, and generates a 3D point cloud picture and a plane picture;

[0016] The boundary line of the brake pad is identified from the plane picture, a plurality of points P are taken on the boundary line, and the plane coordinates P(x, y) of the points P are obtained with the intersection point of the boundary of the plane picture as the origin, according to the one-to-one correspondence of the positions on the 3D point cloud picture and the plane picture, the spatial coordinates P(a, b, c) of the brake pad boundary line with the camera as the origin can be obtained at the position of P(x, y) on the 3D point cloud picture;

[0017] Repeat the above process, that is, the boundary line of the brake pad can be obtained, and the coordinates with noise are removed, and the clear points are left. The thickness of the brake pad is calculated by using the spatial coordinates P(a, b, c).

[0018] Further, when the 3D structured light camera is opposite to the measured brake pad, the thickness of the brake pad is calculated by using the spatial coordinates P(a, b, c), which includes the following steps:

[0019] When the 3D structured light camera is opposite to the measured brake pad, the contour edge of the brake pad on the 3D point cloud picture is a parallelogram, three points P1(x1, y1, z1), P2(x2, y2, z2) and P3(x3, y3, z3) are taken on the contour edge, wherein P2 and P3 are located on the same side, P1 is located on the non-same side with P2 and P3, P1, P2 and P3 are located in the same plane, the standard equation of the straight line L on which P2 and P3 are located is obtained, the horizontal and vertical coordinates of P1 are brought in, and the distance from P1 to the straight line L is obtained, that is, the thickness of the brake pad.

[0020] Further, when the 3D structured light camera is opposite to the measured brake pad, the thickness of the brake pad is calculated by using the spatial coordinates P(a, b, c), which includes the following steps:

[0021] When the 3D structured light camera is opposite to the measured brake pad, the contour edge of the brake pad on the 3D point cloud picture is a parallelogram, three points P1(x1, y1, z1), P2(x2, y2, z2) and P3(x3, y3, z3) are taken on the contour edge, wherein P2 and P3 are located on the same side, P1 is located on the non-same side with P2 and P3, P1, P2 and P3 are located in the same plane, the standard equation of the straight line L on which P2 and P3 are located is obtained, the horizontal and vertical coordinates of P1 are brought in, and the distance from P1 to the straight line L is obtained, that is, the thickness of the brake pad.

[0022] Compared with the prior art, the present application has the following beneficial effects:

[0023] 1. The present application solves the feasibility of measuring by using a 3D camera. At present, the point cloud data of the 3D camera has many noise points, which is difficult to use directly. By using the method in the present application, measurement can be realized.

[0024] 2. The present application reduces the measurement cost, improves the flexibility of measurement, and makes driving safer. Similar measurements of brake pad thickness need to be measured by special personnel after the train returns to the warehouse. After using the present application, the brake pad thickness can be measured online and dynamically, which reduces cost and improves safety. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 is the overall structure schematic diagram of a railway vehicle brake pad thickness detection device based on a structured light camera in the present application, wherein the direction of the arrow represents the driving direction of the train;

[0026] Figure 2 is the structure schematic diagram of the imaging assembly 1 in the present application;

[0027] Figure 3 is the top view of Figure 2 ;

[0028] Figure 4 is the structure schematic diagram of the protection assembly 3 in the present application;

[0029] Figure 5 is the top view of Figure 4 ;

[0030] Figure 6 is the side view of Figure 4 ;

[0031] Figure 7 is the structure schematic diagram of the control assembly 2 in the present application;

[0032] Figure 8 is the outline shape of the brake pad on the 3D point cloud picture when the 3D structured light camera 14 is directly opposite the brake pad to be measured in the present application;

[0033] Figure 9 is the outline shape of the brake pad on the 3D point cloud picture when the 3D structured light camera 14 is obliquely arranged with the brake pad to be measured in the present application. DETAILED DESCRIPTION

[0034] Specific implementation one: combined with Figures 1 to 7 In this embodiment, a railway vehicle brake pad thickness detection device based on a structured light camera includes an imaging assembly 1, a control assembly 2, and an industrial computer 4. The imaging assembly 1 is fixedly connected to the middle of the track group, and includes a 3D structured light camera 14. The control assembly 2 is fixedly connected to one side of the track group in front of the imaging assembly 1. The industrial computer 4 is arranged outside the track group. The control assembly 2 and the imaging assembly 1 are respectively electrically connected to the industrial computer 4. The industrial computer 4 receives the vehicle signal sent by the control assembly 2, controls the imaging assembly 1 to start photographing, displays and saves the generated picture, matches the picture to obtain the spatial coordinates, and performs operation using the spatial coordinates.

[0035] The 3D structured light camera 14 is used to scan and photograph the detection target and its surrounding environment to produce images. The 3D structured light camera 14 used in this embodiment is a line scanning camera, such as a high-speed split 3D camera Ranger3 camera.

[0036] The control assembly 2 is installed in the direction of the incoming train and receives the train before the imaging assembly 1.

[0037] The industrial computer 4 receives the vehicle signal sent by the control component 2, controls the detection device to start working, displays the generated picture, and performs operation.

[0038] The existing brake pad thickness detection method is to directly read the feature point data (3D point cloud data, i.e. the spatial coordinates of each feature point) taken by the camera for calculation. This feature point requires high-quality photos. If the photographed object moves slightly, the feature point will be lost, and the data taken will not be accurate. In addition, if there are problems such as reflection or obstruction near the feature point, it will cause noise in the data, which is also easy to cause deviation of the data. The method provided in the present application first determines the planar position of the feature point using a 2D photo, and is not afraid of the movement of the feature point. When reading data, because the planar coordinates of the straight line where the feature point is located have been determined, the data to be read is limited. Even if it is found that some feature points have noise, the coordinates of other feature points on the straight line where the feature point is located can be read, the required data is less, and the obtained data is clearer.

[0039] Specific implementation method two: combined with Figures 1 to 7 In this embodiment, when the industrial computer 4 is working, the control component 2 detects that the vehicle is driving, sends a signal to the industrial computer 4, and the industrial computer 4 controls the imaging component 1 to start and take a photo to generate a 3D point cloud picture and a planar picture.

[0040] The industrial computer 4 identifies the boundary line of the brake pad from the planar picture, takes a plurality of points P on the boundary line, and obtains the planar coordinates P(x, y) of the point P on the planar picture with the intersection point of the boundary of the planar picture as the origin. According to the one-to-one correspondence of the positions on the 3D point cloud picture and the planar picture, the space coordinates P(a, b, c) of the brake pad boundary line with the camera as the origin can be obtained at the position P(x, y) on the 3D point cloud picture.

[0041] Repeat the above process to obtain all the space coordinates of the brake pad boundary line, remove the coordinates with noise, and leave the points with clear space coordinates. The thickness of the brake pad is calculated using these space coordinates P(a, b, c).

[0042] The technical features not disclosed in this embodiment are the same as those in specific implementation method one.

[0043] Specific implementation method three: combined with Figures 1 to 7To illustrate the embodiment, the imaging assembly 1 of the embodiment further comprises a laser generator 11, a prism 12, a mounting frame 13 and a lens 15, the 3D structured light camera 14, the laser generator 11 and the prism 12 are fixed on the mounting frame 13, the lens 14 is arranged in front of the 3D structured light camera 14, the laser generator 11 is arranged on one side of the 3D structured light camera 14, and the prism 12 is arranged in front of the laser generator 11. The technical features not disclosed in the embodiment are the same as those in the second embodiment.

[0044] The mounting frame 13 is used to fix and mount the 3D structured light camera 14, the laser generator 11 and the prism 12.

[0045] The laser generator 11 is used to generate a laser spot, and the prism 12 diffuses the spot to the required angle.

[0046] The lens 15 determines the shooting range, the laser generator 11 generates a spot, and after passing through the prism 12, the spot becomes a light ray, which has a certain distance and angle with the scanning surface of the 3D structured light camera 14, so that the imaging accuracy is more ideal. In operation, the laser generator 11 generates a spot, and after passing through the prism 12, the light ray illuminates the object to be photographed, and the 3D structured light camera 14 photographs the illuminated part and generates normal pictures and 3D point cloud pictures.

[0047] The fourth embodiment is combined with the first embodiment. Figures 1 to 7 To illustrate the embodiment, the imaging assembly 1 of the embodiment further comprises a laser generator 11, a prism 12, a mounting frame 13 and a lens 15, the 3D structured light camera 14, the laser generator 11 and the prism 12 are fixed on the mounting frame 13, the lens 14 is arranged in front of the 3D structured light camera 14, the laser generator 11 is arranged on one side of the 3D structured light camera 14, and the prism 12 is arranged in front of the laser generator 11. The technical features not disclosed in the embodiment are the same as those in the second embodiment.

[0048] The protective assembly 3 is used to protect the imaging assembly 1 and is installed on the line where the vehicle travels.

[0049] The fifth embodiment is combined with the first embodiment. Figures 1 to 7 To illustrate the embodiment, the protective assembly 3 of the embodiment comprises a box body 31, a box cover 32 and a driving mechanism 33, the upper end of the box body 31 is open, the box cover 32 is arranged outside the open end of the box body 31, the imaging assembly 1 is arranged towards the open end of the box body 31, the box cover 32 is translated on the open end of the box body 31 through the driving mechanism 33, and the driving mechanism 33 is electrically connected with the industrial computer 4. The technical features not disclosed in the embodiment are the same as those in the fourth embodiment.

[0050] In this way, when the control assembly 2 detects that the vehicle is driving, the industrial computer 4 controls the driving mechanism 33 to start, the box cover 32 is translated to the outside of the open end of the box body 31, and the imaging assembly 1 starts to work; after obtaining the pictures photographed by the 3D structured light camera 14, the industrial computer 4 controls the driving mechanism 33 to start, the box cover 32 is translated to the open end of the box body 31, and covers the upper end of the box body 31, thereby protecting the devices in the box body 31.

[0051] Specific implementation six: combined Figures 1 to 7 In this embodiment, the driving mechanism 33 includes an electric push rod 331 and two sets of tracks 332. The electric push rod 331 is horizontally fixed in the box body 31, and the rod end of the electric push rod 331 is fixed with a connecting block 333. The upper end surface of the connecting block 333 is fixed with the inner side end surface of the box cover 32. Two sets of tracks 332 are horizontally arranged on both sides of the open end of the box body 31, and the two sides of the box cover 32 are respectively connected with the tracks 332. The technical features not disclosed in this embodiment are the same as those in specific implementation five.

[0052] The electric push rod 331 is designed to extend horizontally after receiving the control instruction from the industrial computer 4. The connecting block 333 at the rod end drives the box cover 32 to move horizontally outward until the imaging assembly 1 is completely exposed. After the imaging assembly 1 completes the shooting, the industrial computer 4 controls the electric push rod 331 to retract, and the box cover 32 moves back to cover the open end of the box body 31, thereby completing the horizontal movement of the box cover 32. The tracks 332 can ensure the horizontal movement of the box cover 32.

[0053] In this embodiment, the imaging assembly 1 is arranged at the front side of the box body 31, and the front end of the open end of the box body 31 is fixed with a baffle 334. The front end surface of the box cover 32 is open. When the box cover 32 is closed, the baffle 334 is spliced at the open end surface of the box cover 32. The baffle 334 can cooperate with the box cover 32 to form a complete end cover, and limit the movement of the box cover 32. The connecting block 333 is fixed with the middle part of the box cover 32, and at the same time, it is ensured that the connecting block 333 moves to the rear side of the box body 31 before touching the rear end surface, and the imaging assembly 1 has been completely exposed.

[0054] Specific implementation seven: combined Figures 1 to 7 In this embodiment, the control assembly 2 includes a vehicle sensor 21 and a sensor support 22. The vehicle sensor 21 is fixed to the upper end of the sensor support 22, and the sensor support 22 is fixed to one side of the track group. The technical features not disclosed in this embodiment are the same as those in specific implementation one, two, three, four, or five.

[0055] The vehicle sensor 21 is used to detect whether a train passes and sends a signal to start the work of the detection device. The sensor support 22 is used to fix the vehicle sensor 21 to the rail.

[0056] Specific implementation eight: combined Figures 1 to 9 In this embodiment, a railway vehicle brake pad thickness detection method based on a structured light camera includes the following steps:

[0057] When the train is running in the working area of the detection device, it first approaches the control assembly 2, the vehicle sensor 21 in the control assembly 2 detects the passing of the vehicle, sends a signal to the industrial computer 4, and the industrial computer 4 controls the detection device to start working. At this time, the cover 32 of the protection assembly 3 is opened, and the imaging assembly 1 starts working. When the train passes through the imaging assembly 1, the laser generator 11 and the prism 12 in the imaging assembly 1 supplement light to the shutter, and the 3D structured light camera 14 takes a picture of the shutter to generate a 3D point cloud picture and a plane picture;

[0058] The boundary line of the shutter is identified from the plane picture, a plurality of points P are taken on the boundary line, and the plane coordinates P(x, y) of the points P are obtained with the intersection point of the boundary of the plane picture as the origin. According to the one-to-one correspondence of the positions on the 3D point cloud picture and the plane picture, the spatial coordinates P(a, b, c) of the shutter boundary line with the camera as the origin can be obtained at the position P(x, y) on the 3D point cloud picture.

[0059] The above process is repeated to obtain all the spatial coordinates of the shutter boundary line. The coordinates with noise are removed, and the points with clear spatial coordinates are left. The thickness of the shutter is calculated using the spatial coordinates P(a, b, c).

[0060] The 3D structured light camera 14 in the imaging assembly 1 takes a picture of the shutter to generate a 3D point cloud picture and a plane picture. The plane picture includes a grayscale picture and a color picture. If monochromatic light is used for light supplementation, a black and white picture will be generated. If the 3D point cloud picture is directly used for identification and calculation, the algorithm calculation will be difficult due to too many noise points in the 3D point cloud picture. The method uses a color picture to identify the boundary line of the shutter from the image.

[0061] All the spatial coordinates of the shutter boundary line are obtained, the coordinates with noise are removed, and enough clear spatial coordinates can also be obtained. The thickness of the shutter is calculated using the spatial coordinates P(a, b, c).

[0062] Specific implementation method nine: combination Figures 1 to 9 In this embodiment, when the 3D structured light camera 14 is directly opposite the shutter to be measured, the thickness of the shutter is calculated using the spatial coordinates P(a, b, c), including the following steps:

[0063] When the 3D structured light camera 14 is facing the gate to be tested, the outline edge of the gate on the 3D point cloud image is a parallelogram. Take three points on the outline edge: P1(x1,y1,z1), P2(x2,y2,z2), and P3(x3,y3,z3). P2 and P3 are on the same side, while P1 is on a different side. The three points P1, P2, and P3 are in the same plane. The standard equation of the line L containing P2 and P3 is obtained: Ax + By + C = 0. Substitute the x and y coordinates of P1 into the equation to obtain the distance from P1 to the line L, which is the thickness of the gate.

[0064] The undisclosed technical features in this embodiment are the same as those in specific embodiment eight.

[0065] Specific Implementation Method Ten: Combining Figures 1 to 9 This embodiment describes how, when the 3D structured light camera 14 and the gate plate to be measured are tilted, the thickness of the gate plate is calculated using spatial coordinates P(a,b,c), including the following steps:

[0066] When the 3D structured light camera 14 is tilted relative to the gate piece under test, the outline edge of the gate piece on the 3D point cloud image is a cube. Four points are taken on the outline edge: P4(x4,y4,z4), P5(x5,y5,z5), P6(x6,y6,z6), and P7(x7,y7,z7). Among them, P5, P6, and P7 are located on different sides of the same plane, while P4 is not on the same plane as P5, P6, and P7. The standard equation of plane A can be determined by the three points P5, P6, and P7: Ax + By + Cz + D = 0. Substituting the coordinates of P4(x4,y4,z4) into the equation, the distance from P4 to plane A is obtained, which is the thickness of the gate piece.

[0067] The undisclosed technical features in this embodiment are the same as those in specific embodiment eight.

[0068] Working principle

[0069] This invention relates to a method and apparatus for detecting the thickness of brake pads on railway vehicles. When using this method to detect brake pad thickness, dynamic detection of the brake pad thickness can be performed on moving vehicles. The equipment is installed between two railway rails, with a structured light camera aligned with the brake pad. The camera's supplementary lighting device, a laser generator, is at a certain angle to the camera, taking pictures of the brake pad area of ​​the passing train. Since the structured light camera can generate 3D point cloud images and normal images, the boundary of the brake pad is identified using the normal image. By utilizing the correspondence between the 3D point cloud image and the normal image, the spatial coordinate data of the corresponding calculation points are obtained, and the brake pad thickness is calculated accordingly.

[0070] Based on the imaging characteristics of structured light, this invention proposes a novel method for calculating gate thickness. It utilizes the 2D image generated by structured light to find feature points (lines), and obtains the planar coordinates P(x,y) of the point to be detected on the image from these feature points (lines). Then, leveraging the correspondence between the 2D image and the 3D point cloud image from the structured light camera, the spatial coordinates P(a,b,c) of point P are obtained on the 3D point cloud image, thereby calculating the dimensions to be measured, such as the gate thickness.

[0071] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for detecting the thickness of railway vehicle brake pads based on a structured light camera, which utilizes a railway vehicle brake pad thickness detection device based on a structured light camera. The detection device includes an imaging component (1), a control component (2), and an industrial control computer (4). The imaging component (1) is fixedly connected to the middle of the track group. The imaging component (1) includes a 3D structured light camera (14). The control component (2) is fixedly connected to one side of the track group in front of the imaging component (1). The industrial control computer (4) is set on the outside of the track group. The control component (2) and the imaging component (1) are electrically connected to the industrial control computer (4). The industrial control computer (4) receives the vehicle signal sent by the control component (2), controls the imaging component (1) to start taking pictures, displays and saves the generated pictures, matches the pictures to obtain spatial coordinates, and uses the spatial coordinates for calculation. Its features are: The method includes the following steps: During the train's journey, it enters the working area of ​​the detection device and first approaches the control component (2). The vehicle sensor (21) in the control component (2) detects the passing of the vehicle and sends a signal to the industrial control computer (4). The industrial control computer (4) controls the detection device to start working. At this time, the cover (32) of the protective component (3) is opened, and the imaging component (1) starts working. When the train passes the imaging component (1), the laser generator (11) and prism (12) in the imaging component (1) provide supplementary lighting for the brake pads, and the 3D structured light camera (14) takes pictures of the brake pads to generate 3D point cloud images and planar images. Identify the boundary line of the gate plate from the planar image, select several points P on the boundary line, and obtain the planar coordinates P(x,y) of point P on the planar image with the intersection of the planar image boundary as the origin. Based on the one-to-one correspondence between the positions on the 3D point cloud image and the planar image, the spatial coordinates P(a,b,c) of the gate plate boundary line with the camera as the origin can be obtained from the position of P(x,y) on the 3D point cloud image. Repeat the above process to obtain all the spatial coordinates of the gate plate boundary line. Remove the coordinates with noisy points and leave the points with clear spatial coordinates. Use these spatial coordinates P(a,b,c) to calculate the thickness of the gate plate. When the 3D structured light camera (14) is facing the gate plate to be tested, the thickness of the gate plate is calculated using spatial coordinates P(a,b,c) using the following steps: When the 3D structured light camera (14) is facing the gate to be tested, the outline edge of the gate on the 3D point cloud image is a parallelogram. Take three points P1 (x1, y1, z1), P2 (x2, y2, z2) and P3 (x3, y3, z3) on the outline edge. Among them, P2 and P3 are on the same side, and P1 is on a different side from P2 and P3. The three points P1, P2 and P3 are in the same plane. The standard equation of the line L where P2 and P3 are located is obtained, Ax + By + C = 0. Substitute the horizontal and vertical coordinates of P1 into the equation to obtain the distance from P1 to the line L, which is the thickness of the gate. When the 3D structured light camera (14) is tilted relative to the gate plate to be measured, the thickness of the gate plate is calculated using spatial coordinates P(a,b,c) using the following steps: When the 3D structured light camera (14) is tilted relative to the gate plate to be tested, the outline edge of the gate plate on the 3D point cloud image is a cube. Four points are taken on the outline edge: P4 (x4,y4,z4), P5 (x5,y5,z5), P6 (x6,y6,z6) and P7 (x7,y7,z7). Among them, P5, P6 and P7 are located on different sides of the same plane, while P4 is not on the same plane as P5, P6 and P7. The standard equation of plane A can be determined by the three points P5, P6 and P7: Ax+By+Cz+D=0. Substituting the coordinates of P4 (x4,y4,z4) into the equation, the distance from P4 to plane A is obtained, which is the thickness of the gate plate.

2. The method for detecting the thickness of railway vehicle brake pads based on a structured light camera according to claim 1, characterized in that: The imaging assembly (1) also includes a laser generator (11), a prism (12), a mounting bracket (13), and a lens (15). The 3D structured light camera (14), the laser generator (11), and the prism (12) are all fixed on the mounting bracket (13). The lens (15) is located in front of the 3D structured light camera (14), the laser generator (11) is located on one side of the 3D structured light camera (14), and the prism (12) is located in front of the laser generator (11).

3. The method for detecting the thickness of railway vehicle brake pads based on a structured light camera according to claim 2, characterized in that: The imaging component (1) is provided with a protective component (3) on its outer side.

4. The method for detecting the thickness of railway vehicle brake pads based on a structured light camera according to claim 3, characterized in that: The protective component (3) includes a housing (31), a cover (32) and a drive mechanism (33). The upper end of the housing (31) is open. The cover (32) is located on the outside of the open end of the housing (31). The imaging component (1) is located facing the open end of the housing (31). The cover (32) moves horizontally on the open end of the housing (31) through the drive mechanism (33). The drive mechanism (33) is electrically connected to the industrial control computer (4).

5. The method for detecting the thickness of railway vehicle brake pads based on a structured light camera according to claim 4, characterized in that: The drive mechanism (33) includes an electric push rod (331) and two sets of rails (332). The electric push rod (331) is horizontally fixed inside the housing (31). A connecting block (333) is fixed to the end of the electric push rod (331). The upper end face of the connecting block (333) is fixed to the inner end face of the housing cover (32). A set of rails (332) is horizontally provided on both sides of the open end of the housing (31). The two sides of the housing cover (32) are slidably connected to the rails (332).

6. A method for detecting the thickness of railway vehicle brake pads based on a structured light camera according to claim 1, 2, 3, 4 or 5, characterized in that: The control component (2) includes a vehicle sensor (21) and a sensor bracket (22). The vehicle sensor (21) is fixed to the upper end of the sensor bracket (22), and the sensor bracket (22) is fixed to one side of the track assembly.

Images