Railway subgrade settlement detection system and detection method

Abstract

The present application discloses a kind of railway subgrade settlement detection system and detection method, railway subgrade settlement detection system includes being arranged on the wrist arm column of one side of track high-definition image acquisition device, track gauge measuring device, embedded processing board, data transmission device and battery module, and reference point being arranged on the wrist arm column of other side of track and target point being arranged on the sleeper of track two sides, high-definition image acquisition device is used to collect reference point and target point image and transmit to embedded processing board, track gauge measuring device is used to measure the distance between target point and high-definition image acquisition device and transmit to embedded processing board, embedded processing board is connected with data transmission device, the present application carries out settlement data calculation by embedded processing board, data is transmitted to data processing center by data transmission device, data processing center detects whether settlement occurs and timely early warning, can real-time monitoring and timely discover subgrade condition, improve the detection efficiency of subgrade settlement, and detection precision is high.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of detection, in particular to a railway roadbed settlement detection system. BACKGROUND

[0002] In recent years, China's infrastructure construction has developed rapidly, and a four-way rail transportation network has gradually been built. In the rail transportation network, railway construction occupies a large part of the position. Railway safety is a key link in railway construction, and the requirement is gradually strict. In the process of railway construction and maintenance, the settlement of the railway roadbed has always been the focus of customers.

[0003] Railway roadbed settlement is a continuous accumulation process. When the accumulation reaches a certain degree, it will cause serious harm, especially in important areas. The monitoring frequency should be high, the measurement interval should be as short as possible, and the detection accuracy needs to be continuously improved. The existing railway roadbed settlement device is mostly detected by machinery or manually, which cannot realize real-time monitoring, has low automation degree and low detection efficiency, and is prone to various accidents caused by not timely discovering the settlement of the railway roadbed. SUMMARY

[0004] In order to solve the problems of low detection efficiency and low detection accuracy of the existing railway roadbed settlement device, the present application provides a railway roadbed settlement detection system.

[0005] The technical scheme adopted by the present application to achieve the above-mentioned purpose is: the railway roadbed settlement detection system of the present application comprises a high-definition image acquisition device 6 provided on a wrist-arm column 2 on one side of the track, a track gauge measuring device 8, an embedded processing board, a data transmission device and a battery module, and a reference point 3 provided on the wrist-arm column 2 on the other side of the track and a target point 4 provided on the sleeper 1 on both sides of the track. The high-definition image acquisition device 6 is used to acquire the images of the reference point 3 and the target point 4 and transmit them to the embedded processing board. The track gauge measuring device 8 is used to measure the distance between the target point 4 and the high-definition image acquisition device 6 and transmit it to the embedded processing board. The embedded processing board is connected with the data transmission device.

[0006] Preferably, it further comprises a fixing frame 5 and a box 9. The fixing frame 5 is fixed on the wrist-arm column 2. The high-definition image acquisition device 6 and the box 9 are provided on the fixing frame 5. The embedded processing board, the data transmission device and the battery module are provided inside the box 9. The track gauge measuring device 8 is provided outside the box 9.

[0007] The high-definition image acquisition device 6 is a camera. The camera is connected with the embedded processing board. A fill light 7 is further provided outside the camera. The fill light 7 is connected with the embedded processing board.

[0008] Preferably, the gauge measuring device 8 is a laser range finder, the laser range finder is connected with the embedded processing board card, the data transmission device is a 4G antenna device, and the data processed by the embedded processing board card is transmitted through a Lora wireless network.

[0009] Preferably, the battery module is a storage battery, which is used to supply power for the high-definition image acquisition device 6, the gauge measuring device 8, and the embedded processing board card.

[0010] Preferably, the target point 4 is arranged on the sleeper 1 through a support, and the reference point 3 and the target point 4 are both reflective stickers.

[0011] The detection method of the railway roadbed settlement detection system is used for detecting the railway roadbed settlement detection system 5, and includes the following steps.

[0012] S1: The high-definition image acquisition device 6, the gauge measuring device 8, the embedded processing board card, and the data transmission device are corrected according to the reference point 3, and initial parameters are set;

[0013] S2: The high-definition image acquisition device 6 acquires images of the target point 4 at a set time interval, and the gauge measuring device 8 measures the distance between the high-definition image acquisition device 6 and the target point 4;

[0014] S3: The high-definition image acquisition device 6 and the gauge measuring device 8 transmit the acquired data to the embedded processing board card;

[0015] S4: The embedded processing board card processes the received data, and sends the processing result to the data processing center through the data transmission device;

[0016] S5: The data processing center determines whether the alarm condition is met according to the processing result, and outputs alarm information if yes.

[0017] Preferably, in S2, the time interval for the high-definition image acquisition device 6 to acquire images of the target point 4 is 10s.

[0018] Preferably, S4 includes the following steps.

[0019] S4-1: After correction, the high-definition image acquisition device 6 locks the target point 4 in the acquired image, and obtains the position of the target point 4 according to the gauge measuring device 8 and the space recognition technology;

[0020] S4-2: Taking the initial acquired image as a reference image, a rectangular area centered on the target point 4 in the reference image is selected as a reference image subarea, and then the acquired image is a deformation image.

[0021] S4-3: Real-time receiving the data collected by the high-definition image acquisition device 6, obtaining the deformation image, calculating the position information and deformation amount of the reference image sub-area and the deformation image according to the image recognition algorithm.

[0022] Preferably, in the S4-3, the image recognition algorithm comprises the following steps:

[0023] S4-3-1: The distance from the high-definition image acquisition device 6 to the center of each target point 4 is measured by using the gauge measurement device 8, and for each point with a known distance, the three-dimensional coordinates of the point relative to the high-definition image acquisition device 6 can be calculated by a mapping function, and the mapping function expression is as follows:

[0024] (x, y, z) = f(u, v, d) (1)

[0025] In the formula, (x, y, z) is the three-dimensional coordinates of the measured point relative to the high-definition image acquisition device 6, u and v are the coordinates of the measured point in the deformation image, and d is the distance from the measured point to the high-definition image acquisition device 6; two target points 4 on the same vertical line and at a certain distance are selected, and the positions of the centers of the two target points 4 on the deformation image are respectively (u1, v1) and (u2, v2), and the distances from the high-definition image acquisition device 6 are respectively d1 and d2.

[0026] S4-3-2: The three-dimensional coordinates of the two target points 4 relative to the high-definition image acquisition device are calculated by the world coordinate calculation method as follows:

[0027] (x1, y1, z1) = f(u1, v1, d1) (2)

[0028] (x2, y2, z2) = f(u2, v2, d2) (3) Set the coordinates of the two target points in the world coordinate system as (x'1, y'1, z'1) and (x'2, y'2, z'2), and based on the installation mode of the target point (4), the following constraints are obtained:

[0029] x'1 = x'2 = 0 (4)

[0030] z'1 = z'2 (5)

[0031] S4-3-3: The coordinates in the world coordinate system are obtained by rotating the coordinates in the high-definition image acquisition device 6 coordinate system, and the angles of rotation around the x, y, and z axes in the high-definition image acquisition device 6 coordinate system are respectively α, β, and γ. The rotation matrix for converting the coordinates of the target point 4 from the high-definition image acquisition device 6 coordinate system to the world coordinate system is as follows:

[0032]

[0033] The conversion equation is as follows:

[0034]

[0035] wherein s x ,s y ,s z are sin(a), sin(b), sin(g) respectively, c x ,c y ,c z are cos(a), cos(b), cos(g) respectively;

[0036] From the equations (4)(5)(6)(7), we can get

[0037]

[0038] wherein a=x1z2-x2z1, b=y2z1-y1z2,

[0039] Since the high-definition image acquisition device 6 is far away from the target point 4, b must be greater than 0, then the denominator in equation (8) must be positive, from equations (6)(7)(8), we can get

[0040]

[0041] Then g=arctan(s z / c z );

[0042] From the equations (4)(5)(6)(7), we can get

[0043]

[0044]

[0045] Then b=arctan(s y / c y );

[0046] From the equations (4)(5)(6)(7), we can get

[0047]

[0048]

[0049] wherein a′=(x1-x2)s y c z +(y1-y2)s y s z +(z1-z2)c y

[0050] b′=(y1-y2)cz -(x1-x2)s z

[0051] then

[0052] S4-3-4: According to the three rotation angles calculated in S4-3-3, the rotation matrix R can be obtained through formula (6); for the target point that needs to be detected for settlement, after obtaining its coordinates in the deformation image, its coordinates in the world coordinate system can be calculated through formula (1) and (7); the coordinates of each target point to be detected in the world coordinate system are calculated at the initial time and recorded, then the new coordinates are calculated during the subsequent settlement detection, and compared with the initial coordinates, the deviation of the y coordinate axis can be calculated, and the offset is the settlement value.

[0053] The railway subgrade settlement detection system and the detection method of the present application install the railway subgrade settlement detection system on the wrist arm column, set up the target point and the reference point, collect the image and the data within the set time after correction, calculate the settlement data through the embedded processing board and detect, the data is transmitted to the data processing center through the data transmission device, the data processing center detects whether the settlement occurs and timely warns, can real-time monitor and timely find the subgrade condition, improves the detection efficiency of the subgrade settlement, the measurement interval time is short, and the detection precision is high. BRIEF DESCRIPTION OF DRAWINGS

[0054] Fig. 1 It is the installation schematic diagram of the railway settlement detection system of the present application.

[0055] Fig. 2 It is the principle diagram of the railway subgrade settlement detection system of the present application.

[0056] Fig. 3 It is the installation schematic diagram of the railway subgrade settlement detection device of the present application.

[0057] In the figure: 1, sleeper; 2, wrist arm column; 3, reference point; 4, target point; 5, fixed frame; 6, high-definition image acquisition device; 7, light supplementing lamp; 8, track gauge measuring device; 9, box body. DETAILED DESCRIPTION

[0058] The railway subgrade settlement detection system of the present application comprises Figs. 1 to 3As shown, the railway subgrade settlement detection system comprises a high-definition image acquisition device 6 provided on a wrist-arm column 2 on one side of the track, a track gauge measuring device 8, an embedded processing board, a data transmission device and a battery module, a reference point 3 provided on the wrist-arm column 2 on the other side of the track, and a target point 4 provided on the sleeper 1 on both sides of the track. The high-definition image acquisition device 6 is used to acquire images of the reference point 3 and the target point 4 and transmit them to the embedded processing board. The track gauge measuring device 8 is used to measure the distance between the target point 4 and the high-definition image acquisition device 6 and transmit it to the embedded processing board. The embedded processing board is connected to the data transmission device and the data processing center. The data processing center is used to detect whether there is settlement and give early warning.

[0059] The railway subgrade settlement detection system further comprises a fixing frame 5 and a box 9. The fixing frame 5 is fixed on the wrist-arm column 2 by a hoop. The box 9 is fixedly connected below the high-definition image acquisition device 6. The high-definition image acquisition device 6 and the box 9 are fixedly connected with the wrist-arm column 2 through the fixing frame 5. The embedded processing board, the data transmission device and the battery module are arranged inside the box 9. The track gauge measuring device 8 is arranged outside the box 9.

[0060] The high-definition image acquisition device 6 is a camera. The camera is connected with the embedded processing board through a gigabit network port. The camera is further provided with a fill light 7 outside. The fill light 7 is connected with the embedded processing board through an RS-232 serial port. The camera is electrically connected with the fill light 7. Fig. 3 The light in the camera is the fill light 7.

[0061] The track gauge measuring device 8 is a laser range finder. The laser range finder is connected with the embedded processing board through an RS-232 serial port.

[0062] The data transmission device is a 4G antenna device. The data processed by the embedded processing board is transmitted through a Lora wireless network. Specifically, the data is transmitted from the embedded processing board to the Lora wireless network. The Lora wireless network is connected to the 4G antenna through a serial port. The 4G antenna transmits the data to the data processing center.

[0063] The battery module is a storage battery. The system further comprises a voltage reduction module and a voltage stabilization module. The modules are used to supply power to the high-definition image acquisition device, the track gauge measuring device and the embedded processing board. The embedded processing board is an ARM board.

[0064] The detection method of the railway subgrade settlement detection system is used to detect the railway subgrade settlement detection system. The method comprises the following steps:

[0065] S1: The railway subgrade settlement detection system is corrected and initial parameters are set according to the reference point 3. The correction process is as follows: the notebook computer is connected to the embedded processing board card through the network, and the computer correction program is run. In the correction program, the image collected by the camera can be seen. First, the operator frames the approximate position of each target point on the image in the correction program, and measures the distance from the retroreflector to the camera by a laser range finder and other devices, and fills the value into the program. Then, the correction program sends the framed information and the corresponding retroreflector distance to the device board card. The board card accurately identifies the target points framed and calculates the three-dimensional position. This position is the standard position;

[0066] S2: The high-definition image acquisition device 6 acquires images of the target points 4 at a set time interval. The time interval for the high-definition image acquisition device 6 to acquire images of the target points 4 is 10s. At the same time, the track gauge measuring device 8 measures the distance between the high-definition image acquisition device 6 and the target points 4;

[0067] S3: The high-definition image acquisition device 6 and the track gauge measuring device 8 transmit the collected data to the embedded processing board card;

[0068] S4: The embedded processing board card processes the received data and transmits the processing results to the data processing center. The specific steps include the following:

[0069] S4-1: After correction, the high-definition image acquisition device 6 locks the target points 4 in the collected images, and obtains the positions of the target points 4 according to the track gauge measuring device 8 and the space recognition technology. The space recognition technology is to install a radar on the railway subgrade settlement detection system;

[0070] S4-2: Take the initial collected image as the reference image, and frame a rectangular area centered on the target point 4 in the reference image as the reference image sub-area. The collected image is a deformed image;

[0071] S4-3: Real-time receive the data collected by the high-definition image acquisition device 6, obtain the deformed image, and calculate the position information and deformation of the reference image sub-area and the deformed image according to the image recognition algorithm. The image recognition algorithm includes the following steps:

[0072] S4-3-1: The track gauge measuring device 8 measures the distance from the high-definition image acquisition device 6 to the center of each target point 4. For each known distance point, the three-dimensional coordinates of the point relative to the high-definition image acquisition device 6 can be calculated through the mapping function. The mapping function expression is as follows:

[0073] (x, y, z) = f(u, v, d) (1)

[0074] In the formula, (x, y, z) is the three-dimensional coordinates of the measurement point relative to the high-definition image acquisition device 6, u, v are the coordinates of the measurement point in the deformed image, and d is the distance from the measurement point to the high-definition image acquisition device 6; two target points 4 on the same vertical line and at a certain distance apart are selected, and the positions of the centers of the two target points 4 on the deformed image are respectively (u1, v1) and (u2, v2), and the distances from the high-definition image acquisition device 6 are respectively d1 and d2;

[0075] S4-3-2: The three-dimensional coordinates of the two target points 4 relative to the high-definition image acquisition device are calculated by the world coordinate calculation method as follows:

[0076] (x1, y1, z1) = f(u1, v1, d1) (2)

[0077] (x2, y2, z2) = f(u2, v2, d2) (3) The coordinates of the two target points in the world coordinate system are set as (x'1, y'1, z'1) and (x'2, y'2, z'2), and based on the installation mode of the target points 4, the following constraints are obtained:

[0078] x'1 = x'2 = 0 (4)

[0079] z'1 = z'2 (5)

[0080] S4-3-3: The coordinates in the world coordinate system are obtained by rotating the coordinates in the high-definition image acquisition device 6 coordinate system, and the angles of rotation around the x, y, and z axes in the high-definition image acquisition device 6 coordinate system are set as α, β, and γ, respectively. The rotation matrix for converting the coordinates of the target points 4 from the high-definition image acquisition device 6 coordinate system to the world coordinate system is as follows:

[0081]

[0082] The conversion equation is as follows:

[0083]

[0084] where s x ,s y ,s z are sin(α), sin(β), and sin(γ), respectively, and c x ,c y ,c z are cos(α), cos(β), and cos(γ), respectively.

[0085] From equations (4), (5), (6), and (7),

[0086]

[0087] wherein a=x1z2-x2z1, b=y2z1-y1z2,

[0088] Since the high-definition image acquisition device 6 is far away from the target point 4, b is certainly greater than 0, and the denominator in formula (8) is certainly positive, and formula (6)(7)(8) can be obtained:

[0089]

[0090] Then γ=arctan(s z / c z );

[0091] From formula (4)(5)(6)(7), we can obtain:

[0092]

[0093]

[0094] Then β=arctan(s y / c y );

[0095] From formula (4)(5)(6)(7), we can obtain:

[0096]

[0097]

[0098] wherein a′=(x1-x2)s y c z +(y1-y2)s y s z +(z1-z2)c y

[0099] b′=(y1-y2)c z -(x + -x2)s z

[0100] Then

[0101] S4-3-4: According to the three rotation angles calculated in S4-3-3, the rotation matrix R can be obtained through formula (6); for the target point that needs to be detected for settlement, after obtaining its coordinates in the deformed image, its coordinates in the world coordinate system can be calculated through formula (1) and (7); the coordinates of each target point to be measured in the world coordinate system are calculated at the initial time and recorded, and then the new coordinates are calculated during the subsequent settlement detection, and compared with the initial coordinates, the deviation of the y coordinate axis can be calculated, and the offset is the settlement value.

[0102] S5: The data processing center judges whether the alarm condition is met according to the processing result, and outputs alarm information if yes.

[0103] The railway roadbed settlement detection system of the present application is installed on a wrist-arm stand, target points and reference points are set, images and data are collected within a set time after correction, settlement data calculation is performed through an embedded processing board, data is transmitted to a data processing center through a data transmission device, the data processing center detects whether settlement occurs and timely warns, the roadbed condition can be monitored in real time and timely discovered, the detection efficiency of roadbed settlement is improved, the measurement interval time is short, and the detection precision is high.

[0104] The present application is described through examples, and those skilled in the art know that various changes or equivalent replacements can be made to these features and examples without departing from the spirit and scope of the present application. In addition, these features and examples can be modified to adapt to specific conditions and materials under the guidance of the present application without departing from the spirit and scope of the present application. Therefore, the present application is not limited by the specific examples disclosed herein, and all examples falling within the scope of the claims of the present application belong to the protection scope of the present application.

Claims

1. A detection method for a railway subgrade settlement detection system, characterized in that, A railway subgrade settlement detection system is used, including a high-definition image acquisition device (6), a track gauge measuring device (8), an embedded processing board, a data transmission device, and a battery module installed on a cantilever column (2) on one side of the track, as well as a reference point (3) installed on a cantilever column (2) on the other side of the track and a target point (4) installed on sleepers (1) on both sides of the track. The high-definition image acquisition device (6) is used to acquire images of the reference point (3) and the target point (4) and transmit them to the embedded processing board. The track gauge measuring device (8) is used to measure the distance between the target point (4) and the high-definition image acquisition device (6) and transmit it to the embedded processing board. The embedded processing board is connected to the data transmission device. The detection process using the aforementioned railway subgrade settlement detection system includes the following steps: S1: Based on the reference point (3), the high-definition image acquisition device (6), track gauge measurement device (8), embedded processing board and data transmission device are calibrated and the initial parameters are set; S2: The high-definition image acquisition device (6) acquires images of the target point (4) at a set time interval, while the track gauge measuring device (8) measures the distance between the high-definition image acquisition device (6) and the target point (4); S3: The high-definition image acquisition device (6) and the track gauge measurement device (8) transmit the acquired data to the embedded processing board; S4: The embedded processing board processes the received data and sends the processing results to the data processing center through the data transmission device; S4-1: After correction, the high-definition image acquisition device (6) locks the target point (4) in the acquired image and obtains the location of the target point (4) according to the track gauge measurement device (8) and spatial recognition technology; S4-2: Using the initial acquired image as the reference image, a rectangular area is selected in the reference image with the target point (4) as the center as the reference image sub-area, and the image acquired afterward is a deformed image; S4-3: Receive data acquired by the high-definition image acquisition device (6) in real time, obtain a deformed image, and calculate the position information and deformation amount of the reference image sub-region and the deformed image according to the image recognition algorithm. The image recognition algorithm includes the following steps: S4-3-1: The distance from the high-definition image acquisition device (6) to the center of each target point (4) is measured using the track gauge measuring device (8). For each point with a known distance, the three-dimensional coordinates of the point relative to the high-definition image acquisition device (6) can be calculated using the mapping function. The mapping function expression is as follows: (1) In the formula, (x,y,z) are the three-dimensional coordinates of the measurement point relative to the high-definition image acquisition device (6), u,v are the coordinates of the measurement point in the deformed image, and d is the distance from the measurement point to the high-definition image acquisition device (6); select two target points (4) on the same vertical line and at a certain distance from each other. Let the positions of the centers of the two target points (4) on the deformed image be (u1,v1) and (u2,v2), respectively, and the distances to the high-definition image acquisition device (6) be d1 and d2, respectively. S4-3-2: The three-dimensional coordinates of the two target points (4) relative to the high-definition image acquisition device are calculated using the world coordinate calculation method as follows: (x1,y1,z1)=f(u1,v1,d1) (2) (x2,y2,z2)=f(u2,v2,d2) (3) The coordinates of the two target points in the world coordinate system are set as (x1', y1', z1') and (x'2, y'2, z'2). Based on the installation method of target point (4), the following constraints apply: x1'=x'2=0 (4) z′1=z′2 (5) S4-3-3: The coordinates in the world coordinate system are obtained by rotating the coordinates in the coordinate system of the high-definition image acquisition device (6). Let the rotation angles around the x, y, and z axes in the coordinate system of the high-definition image acquisition device (6) be α, β, and γ, respectively. Then the rotation matrix for the target point (4) from the coordinate system of the high-definition image acquisition device (6) to the world coordinate system is as follows: The transformation equation is as follows: Where s x ,s y ,s z They are sin(α), sin(β), sin(γ), c respectively. x ,c y ,c z They are cos(α), cos(β), cos(γ) respectively; From equations (4)(5)(6)(7), we can obtain that Where, a = x1z2 - x2z1, b = y2z1 - y1z2. Since the high-definition image acquisition device (6) is far from the target point (4), b must be greater than 0. Therefore, the denominator in equation (8) must be a positive number. From equations (6), (7), and (8), we can obtain: γ=arctan(s z / c z ); From equations (4)(5)(6)(7), we can obtain: Then β = arctan(s) y / c y ); From equations (4)(5)(6)(7), we can obtain: Where a'=(x1-x2)s y c z +(y1-y2)s y s z +(z1-z2)c y b′=(y1-y2)c z -(x1-x2)s z but S4-3-4: The rotation matrix R can be obtained by equation (6) based on the three rotation angles calculated in S4-3-3. For the target point that needs to be detected for settlement, after obtaining its coordinates in the deformed image, its coordinates in the world coordinate system can be calculated by equations (1) and (7). The coordinates of each target point to be tested in the world coordinate system are calculated and recorded at the beginning. Then, the new coordinates are calculated during the subsequent settlement detection and compared with the initial coordinates. The deviation of the y-coordinate axis can be calculated, and the deviation is the settlement value. S5: The data processing center determines whether the alarm conditions are met based on the processing results. If so, it outputs alarm information.

2. The detection method of the railway subgrade settlement detection system according to claim 1, characterized in that, The railway subgrade settlement detection system also includes a fixed frame (5) and a box (9). The fixed frame (5) is fixed on the cantilever column (2). The high-definition image acquisition device (6) and the box (9) are located on the fixed frame (5). The embedded processing board, data transmission device and battery module are located inside the box (9). The track gauge measuring device (8) is located outside the box (9).

3. The detection method of the railway subgrade settlement detection system according to claim 1, characterized in that, The high-definition image acquisition device (6) is a camera, which is connected to an embedded processing board. A fill light (7) is also provided outside the camera, and the fill light (7) is connected to the embedded processing board.

4. The detection method of the railway subgrade settlement detection system according to claim 1, characterized in that, The track gauge measuring device (8) is a laser rangefinder, which is connected to the embedded processing board. The data transmission device is a 4G antenna device, and transmits the data processed by the embedded processing board through the LoRa wireless network.

5. The detection method of the railway subgrade settlement detection system according to claim 1, characterized in that, The battery module is a storage battery used to power the high-definition image acquisition device (6), the track gauge measurement device (8), and the embedded processing board.

6. The detection method of the railway subgrade settlement detection system according to claim 1, characterized in that, The target point (4) is set on the sleeper (1) by a bracket, and both the reference point (3) and the target point (4) are reflective stickers.

7. The detection method of the railway subgrade settlement detection system according to claim 1, characterized in that, In S2, the time interval for the high-definition image acquisition device (6) to acquire the image of the target point (4) is 10s.

Images