CN110849720A - Monitoring system and analysis system for steel rail longitudinal resistance test based on image recognition - Google Patents

Abstract

The invention relates to a monitoring system and an analysis system for a steel rail longitudinal resistance test based on image recognition, wherein the monitoring system comprises more than one calibration plate, a camera I (used for recording video), an image extraction system I (used for extracting pictures in the video) and an image processing system I (used for processing the pictures to obtain strain data), wherein the camera I, the image extraction system I and the image processing system I are sequentially connected; the analysis system comprises the monitoring system, a stress acquisition system (used for acquiring stress data) and a data processing system which are connected with each other, wherein the image processing system I is connected with the data processing system, and the data processing system is used for processing the strain data and the stress data and outputting an analysis result. The invention can realize monitoring of the steel rail strain in the steel rail longitudinal resistance test by using an image processing method and data analysis, and is suitable for popularization and application.

Description

Monitoring system and analysis system of rail longitudinal resistance test based on image recognition

technical field

The invention belongs to the technical field of test devices, and relates to a monitoring system and an analysis system for a longitudinal resistance test of a rail based on image recognition.

Background technique

In the rail transit road structure, due to the strong ability of the concrete ballast to maintain the geometric shape, the longitudinal resistance of the line is generally determined by the longitudinal resistance of the rail formed by the restraint of the elastic fastener. The longitudinal resistance of the rail is not only an important parameter in the design of the track structure, but also a necessary basic data in the railway engineering construction standards, railway product standards and operation and maintenance standards, and it needs to be determined by scientific and accurate methods. At present, the strain monitoring method in the rail longitudinal resistance test is mainly measured by the contact strain sensor. The contact strain sensor is characterized by simple operation and can directly reflect the results of the strain test. However, there are also defects that cannot be ignored. Errors are easily caused by the slippage of the rails.

Therefore, in order to meet the demand for the accuracy of strain monitoring in the rail longitudinal resistance test, it is necessary to improve the strain monitoring method for the rail longitudinal resistance test.

SUMMARY OF THE INVENTION

The purpose of the present invention is to solve the problem of insufficient accuracy of strain monitoring in the longitudinal resistance test of the rail in the prior art, to provide a monitoring system for the longitudinal resistance test of the rail based on image recognition, and further, to obtain a monitoring system based on the monitoring system. an analysis system.

For achieving the above object, the technical scheme adopted in the present invention is as follows:

A monitoring system for a longitudinal resistance test of a steel rail based on image recognition, comprising a calibration plate and sequentially connected cameras I, an image extraction system I and an image processing system I;

The number of calibration plates is one, which is fixed on the side of the rail waist of the rail specimen, or the number of calibration plates is two, one is fixed on the side of the rail waist of the rail specimen, and the other is fixed on the rail specimen and the longitudinal resistance. On the fixed brackets between the testing machines, the horizontal heights of the two are the same;

The calibration plate is located in the field of view of the camera 1, and the camera 1 is fixed on the side of the rail test piece, and is used for video recording in the rail longitudinal resistance test process, and the shooting angle of the camera 1 is fixed in the video recording process;

Image extraction system 1 is used for extracting the picture in the video recording of camera 1 by frame number;

The image processing system I is used to process the picture to obtain the strain data of the rail during the longitudinal resistance test of the rail.

The present invention adopts the camera I to record the position of the calibration plate during the whole process of the test, and extracts the pictures in the video according to the number of frames by the image extraction system I, and then uses the image processing system I to process the extracted pictures, and the test process can be obtained. The strain condition of the rail in ;

Among them, when the number of calibration plates is one, when the rail specimen drives the calibration plate to move, the position of the calibration plate in each frame of pictures changes, and the position of the calibration plate in the second and subsequent pictures is the same as that of the first and second pictures. The relative displacement of the calibration plate can be obtained by comparing a picture two by two; when the number of calibration plates is two, the processing process is that when the rail specimen drives the first calibration plate to move, the The second calibration plate has no displacement, so the relative positions of the two calibration plates in each frame of the picture change. Calculate the distance between the two calibration plates in each frame of the picture, and then the Relative displacement; finally, the strain data of the rail can be obtained according to the relative displacement of the calibration plate. Compared with the method in the prior art, the image recognition method is more accurate and convenient.

As the preferred technical solution:

A kind of monitoring system of the rail longitudinal resistance test based on image recognition as mentioned above, camera I is fixed on the camera fixing device, and the camera fixing device is hingedly connected with the camera fixing platform, which makes the camera fixing device relative to the camera fixing platform in the YZ plane. It is easy to adjust the camera to keep the displacement of the calibration plate in the horizontal direction. The purpose is to facilitate the calculation and reduce the error. The camera fixing platform is movably connected with the camera bracket, and the camera bracket is used to move the camera in the direction Y and direction Z to fix the camera. The platform, the direction Y and the direction Z are perpendicular to each other, and both are perpendicular to the side of the rail waist of the rail specimen.

In the above-mentioned monitoring system for the longitudinal resistance test of steel rails based on image recognition, the bottom of the camera bracket is provided with a switch-type magnetic base, and the camera bracket is placed on an iron working platform.

A monitoring system for the longitudinal resistance test of rails based on image recognition as mentioned above, the camera bracket is composed of 14 rods, of which 12 rods are connected to the cube frame, and the other 2 rods are parallel to each other and are fixed in the cube frame. Parallel to each other with the 4 rods in the cube frame.

In the above-mentioned monitoring system for the longitudinal resistance test of rails based on image recognition, an illumination light source fixing device is installed on the top of the camera bracket. When the ambient light is insufficient, an illumination light source can be added to compensate for the illumination.

The present invention also provides an analysis system for the longitudinal resistance test of the rail based on image recognition, including the above-mentioned monitoring system for the longitudinal resistance test of the rail based on the image recognition, and also includes a stress acquisition system and a data processing system connected to each other, The stress acquisition system is used to collect the stress data that changes from time to time during the longitudinal resistance test of the rail. The image processing system I in the monitoring system is connected to the data processing system, and the data processing system is used to draw the strain data and stress data into a one-to-one correspondence table. , select the corresponding method to analyze the strain data and stress data, and output the analysis results; specifically, the data processing system integrates the obtained strain data and stress data, matches them in chronological order, and can output them as detailed data. Excel file and stress-strain curve diagram, and then use data analysis software to analyze the trend of increase and decrease of stress data, and record the time, maximum value and strain value at the same time when the stress begins to decrease, and the analysis results can be output.

On the basis of the above-mentioned monitoring system for the longitudinal resistance test of the rail based on image recognition, the present invention further adopts a stress acquisition system and a data processing system. The aforementioned strain data and the stress data collected by the stress acquisition system are processed to obtain one-to-one corresponding data, and finally the data is analyzed to output the test results;

Since when the static friction between the two objects turns into dynamic friction, the dynamic friction force is less than or equal to the static friction force, and the longitudinal resistance of the test rail is the maximum longitudinal resistance value when the test rail starts to slide. The analysis system in the present invention is the basis of the monitoring system. , the strain of the accurate rail keeps increasing, but the friction force remains the same or begins to decrease, which also means that the value is the maximum longitudinal resistance value, that is, the test result obtained by the test.

As the preferred technical solution:

As mentioned above, an analysis system of rail longitudinal resistance test based on image recognition, the stress acquisition system is mainly composed of interconnected strain gauges and stress sensor display screens, as well as sequentially connected cameras II, image extraction system II and image processing system II , the image processing system II is connected with the data processing system;

The strain gauge is installed on the rail specimen to measure the stress data that changes at any time during the longitudinal resistance test of the rail;

The stress sensor display screen is used to display the stress data that changes all the time during the longitudinal resistance test of the rail in real time;

The stress sensor display screen is located within the field of view of the camera II, and the camera II is fixed on the side of the rail specimen for recording during the longitudinal resistance test of the rail. During the recording process, the shooting angle of the camera II is fixed;

The image extraction system II is used to extract the pictures in the video of the camera II according to the number of frames;

The image processing system II is used to process the pictures to obtain the strain data of the rail during the longitudinal resistance test of the rail.

The above-mentioned analysis system for the longitudinal resistance test of rails based on image recognition, the camera I and the camera II are the same camera, the image extraction system I and the image extraction system II are the same image extraction system, and the image processing system I and the image The processing system II is the same image processing system.

As mentioned above, an analysis system for the longitudinal resistance test of rails based on image recognition, the stress acquisition system is mainly composed of sequentially connected strain gauges, a data acquisition instrument and a PC; the strain gauges are installed on the rail specimens to measure the rails The stress data that changes all the time during the longitudinal resistance test; the data acquisition instrument is used to collect the stress data that changes all the time during the longitudinal resistance test of the rail; the PC is used to record and store the stress data that changes all the time during the longitudinal resistance test of the rail in real time.

In the above-mentioned analysis system for the longitudinal resistance test of rails based on image recognition, the analysis result is a stress-strain curve diagram.

Beneficial effects:

(1) A monitoring system for a rail longitudinal resistance test based on image recognition of the present invention can effectively monitor the rail strain in the rail longitudinal resistance test with an image processing method, and the test results are accurate and efficient;

(2) An analysis system for the longitudinal resistance test of the rail based on image recognition of the present invention can perform the function of data analysis on the basis of accurately testing the rail strain in the longitudinal resistance test of the rail;

(3) An analysis system for the longitudinal resistance test of rails based on image recognition of the present invention has the advantages of light weight, simple structure, easy disassembly, low maintenance cost, low requirements for the test environment, and is suitable for popularization and use.

Description of drawings

1 is a schematic diagram of an analysis system for an image recognition-based longitudinal resistance test of a rail in Embodiment 2 of the present invention;

2 is a schematic diagram of an analysis system for an image recognition-based longitudinal resistance test of a rail in Example 3 of the present invention;

3 is a schematic structural diagram of a camera support;

4 is a schematic structural diagram of a camera fixing device;

Among them, 1-camera I, 2-rail specimen, 3-iron working platform, 4-data acquisition instrument, 5-strain gauge, 6-camera bracket, 7-camera fixed platform, 8-switch magnetic base, 9 -Lighting source fixture, 10-Camera fixture, 11-Image extraction system I, 12-Image processing system I, 13-Data processing system, 14-Calibration board, 15-Strain sensor display screen, 16-PC machine.

Detailed ways

The present invention will be further described below in conjunction with specific embodiments. It should be understood that these examples are only used to illustrate the present invention and not to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the appended claims of the present application.

Example 1

A monitoring system for a longitudinal resistance test of a steel rail based on image recognition, comprising a calibration plate and sequentially connected cameras I, an image extraction system I and an image processing system I;

The number of the calibration plate is one, which is fixed on the side of the rail waist of the rail test piece; the calibration plate is located in the field of view of the camera I, and the camera I is fixed on the side of the rail test piece for video recording during the longitudinal resistance test of the rail, During the video recording process, the shooting angle of view of the camera I is fixed;

The camera head 1 is fixed on the camera head fixing device 10 as shown in Figure 4, the camera head fixing device 10 is hingedly connected with the camera head fixing platform 7, and the camera head fixing platform 7 is movably connected with the camera head support 6, as shown in Figure 3, the camera head support 6 is used for. Move the camera fixing platform 7 along the direction Y and the direction Z, the direction Y and the direction Z are perpendicular to each other, and both are perpendicular to the side of the rail waist of the rail test piece, wherein the bottom of the camera bracket 6 is equipped with a switch-type magnetic seat 8, and the camera bracket 6 is placed on an iron working platform; the camera bracket 6 consists of 14 rods, of which 12 rods are connected to a cube frame (the cube frame is part of the camera bracket 6), and the other 2 rods are parallel to each other and are fixed in the cube frame At the same time, it is parallel to the four rods in the cube frame; the top of the camera bracket 6 is installed with an illumination light source fixing device 9;

Image extraction system 1 is used for extracting the picture in the video recording of camera 1 by frame number;

The image processing system I is used to process the picture to obtain the strain data of the rail during the longitudinal resistance test of the rail.

Example 2

A kind of analysis system of the rail longitudinal resistance test based on image recognition, as shown in Figure 1, comprises the monitoring system of a kind of image recognition-based rail longitudinal resistance test in embodiment 1 (1 is camera 1 in the figure, 2 is rail Test piece, 3 is an iron working platform, 11 is an image extraction system 1, 12 is an image processing system 1, 14 is a calibration board), an interconnected stress acquisition system and a data processing system 13;

The stress acquisition system is used to collect the stress data that changes at any time during the longitudinal resistance test of the rail;

The stress acquisition system is composed of the interconnected strain gauges 5 and the stress sensor display screen 15, as well as the sequentially connected camera II, the image extraction system II and the image processing system II, and the image processing system II is connected with the data processing system 13;

The strain gauge 5 is installed on the rail specimen to measure the stress data that changes from time to time during the longitudinal resistance test of the rail;

The stress sensor display screen 15 is used to display in real time the stress data that changes constantly during the longitudinal resistance test of the rail;

The stress sensor display screen 15 is located in the field of view of the camera II, and the camera II is fixed on the side of the rail test piece for video recording during the longitudinal resistance test of the rail, and the shooting angle of the camera II is fixed during the video recording process;

The image extraction system II is used to extract the pictures in the video of the camera II according to the number of frames;

The image processing system II is used to process the pictures to obtain the strain data of the rail during the longitudinal resistance test of the rail;

Camera I and camera II are the same camera, image extraction system I and image extraction system II are the same image extraction system, and image processing system I and image processing system II are the same image processing system; data processing system 13, image extraction system The I11 and the image processing system I12 are installed on the same PC 16;

The image processing system 1 in the monitoring system is connected with the data processing system 13, and the data processing system 13 is used to draw the strain data and the stress data into a one-to-one correspondence table, select the corresponding method to analyze the strain data and the stress data, and output the stress. --Strain curve graph.

Example 3

An analysis system for the longitudinal resistance test of the rail based on image recognition, as shown in Figure 2, includes the monitoring system for the longitudinal resistance test of the rail based on the image recognition in Embodiment 1, the interconnected stress acquisition system and the data processing system 13;

The stress acquisition system is used to collect the stress data that changes from time to time during the longitudinal resistance test of the rail. The image processing system 112 in the monitoring system is connected with the data processing system 13. The stress acquisition system consists of the sequentially connected strain gauges 5, data acquisition instruments 4 and PC machine is composed of 16; the strain gauge 5 is installed on the rail test piece, which is used to measure the stress data that changes from time to time during the longitudinal resistance test of the rail; the data acquisition instrument is used to collect the stress data that changes from time to time during the longitudinal resistance test of the rail; PC machine 16 It is used for real-time recording and storage of stress data that changes at any time during the longitudinal resistance test of the rail;

The data processing system 13 is used to draw the strain data and the stress data into a one-to-one correspondence table, select a corresponding method to analyze the strain data and the stress data, and output a stress-strain curve graph.

Example 4

A kind of analysis system of the rail longitudinal resistance test based on image recognition in embodiment 2 is used for testing, and the steps of testing are as follows:

(1) the calibration plate is pasted on the side of the rail waist of the rail test piece, adjust the camera 1 so that the calibration plate is positioned at the camera 1 in the field of view and fix the camera 1;

(2) Install the strain gauge on the rail specimen, connect the strain gauge to the stress sensor display screen, adjust the camera II so that the stress sensor display screen is within the field of view of the camera II, and fix the camera II;

(3) Turn on the video recording function of camera I and camera II, and start the rail longitudinal resistance test;

(4) end the test, after the picture in the video of the camera 1 is extracted by the image extraction system 1 by the frame number, the picture is processed by the image processing system 1 to obtain the strain data of the rail in the longitudinal resistance test process of the rail and sent to the data processing system, At the same time, after the image extraction system II is used to extract the pictures in the video of the camera II according to the number of frames, the image processing system II processes the pictures to obtain the strain data of the rail during the longitudinal resistance test of the rail and sends it to the data processing system;

(5) The data processing system draws the strain data and stress data into a one-to-one correspondence table, outputs the stress-strain curve graph, and selects the corresponding method to analyze the strain data and stress data. When the stress remains unchanged or begins to decrease, If the strain continues to increase, it can be judged that the rail slips at this time, and the test result can be obtained by calculating the maximum stress value.

Comparative Example 1

A traditional test device is used for the test. The test process is as follows: connect the contact strain acquisition system and the stress acquisition system to the rail specimen respectively, and adjust the data of the two acquisition systems to zero. After the test starts, the strain acquisition system The system records and stores the strain data of the test process, and the stress acquisition system records and stores the stress data of the test process until the end of the test.

Use the data processing system to draw the strain data and stress data into a one-to-one correspondence table, according to the relevant test standards, manually compare the table data, and judge the test results.

It can be seen from the comparison between Example 1 and Example 4 that the slip of the rail (the slip situation that produces errors) was recorded during the test in Example 4, and according to the record, it can be more accurately judged The corresponding relationship between stress and strain is convenient and quick.

Claims (10)

1. A monitoring system for a steel rail longitudinal resistance test based on image recognition is characterized in that: the system comprises a calibration board, a camera I, an image extraction system I and an image processing system I which are sequentially connected;

the number of the calibration plates is one, and the calibration plates are fixed on the side surface of the rail web of the steel rail test piece, or the number of the calibration plates is two, one calibration plate is fixed on the side surface of the rail web of the steel rail test piece, the other calibration plate is fixed on a fixing support between the steel rail test piece and the longitudinal resistance testing machine, and the horizontal heights of the calibration plates and the fixing support are the same;

the calibration plate is positioned in the visual field range of the camera I, the camera I is fixed on the side of the steel rail test piece and used for recording video in the longitudinal resistance test process of the steel rail, and the shooting visual angle of the camera I is fixed in the video recording process;

the image extraction system I is used for extracting pictures in the video of the camera I according to the number of frames;

and the image processing system I is used for processing the pictures to obtain the strain data of the steel rail in the longitudinal resistance test process of the steel rail.

2. The monitoring system for the rail longitudinal resistance test based on the image recognition as claimed in claim 1, wherein the camera I is fixed on a camera fixing device, the camera fixing device is hinged with a camera fixing platform, the camera fixing platform is movably connected with a camera support, the camera support is used for moving the camera fixing platform along a direction Y and a direction Z, the direction Y and the direction Z are perpendicular to each other, and the direction Y and the direction Z are perpendicular to the side surface of the rail web of the rail test piece.

3. The monitoring system for the steel rail longitudinal resistance test based on the image recognition is characterized in that the switch type magnetic seat is arranged at the bottom of the camera support, and the camera support is placed on an iron working platform.

4. The monitoring system for the rail longitudinal resistance test based on the image recognition is characterized in that the camera support is composed of 14 rods, wherein 12 rods are connected into a cubic frame, and the other 2 rods are parallel to each other, fixed in the cubic frame and parallel to 4 rods in the cubic frame.

5. The monitoring system for the rail longitudinal resistance test based on the image recognition is characterized in that an illumination light source fixing device is installed at the top of the camera support.

6. An analysis system for a steel rail longitudinal resistance test based on image recognition is characterized in that: the monitoring system comprises the monitoring system for the steel rail longitudinal resistance test based on the image recognition as claimed in any one of claims 1 to 5, and further comprises a stress acquisition system and a data processing system which are connected with each other, wherein the stress acquisition system is used for acquiring stress data which changes at any time in the steel rail longitudinal resistance test process, an image processing system I in the monitoring system is connected with the data processing system, the data processing system is used for drawing the strain data and the stress data into a one-to-one corresponding table, a corresponding method is selected for analyzing the strain data and the stress data, and an analysis result is output.

7. The analysis system for the steel rail longitudinal resistance test based on the image recognition is characterized in that the stress acquisition system mainly comprises a strain gauge and a stress sensor display screen which are connected with each other, and a camera II, an image extraction system II and an image processing system II which are connected in sequence, wherein the image processing system II is connected with the data processing system;

the strain gauge is arranged on the steel rail test piece and used for measuring stress data which changes constantly in the process of the steel rail longitudinal resistance test;

the stress sensor display screen is used for displaying stress data which changes constantly in the process of the longitudinal resistance test of the steel rail in real time;

the stress sensor display screen is positioned in the visual field range of the camera II, the camera II is fixed on the side of the steel rail test piece and used for recording video in the longitudinal resistance test process of the steel rail, and the shooting visual angle of the camera II is fixed in the video recording process;

the image extraction system II is used for extracting pictures in the video of the camera II according to the number of frames;

and the image processing system II is used for processing the pictures to obtain the strain data of the steel rail in the longitudinal resistance test process of the steel rail.

8. The analysis system for the steel rail longitudinal resistance test based on the image recognition as claimed in claim 7, wherein the camera I and the camera II are the same camera, the image extraction system I and the image extraction system II are the same image extraction system, and the image processing system I and the image processing system II are the same image processing system.

9. The analysis system for the steel rail longitudinal resistance test based on the image recognition is characterized in that the stress acquisition system mainly comprises a strain gauge, a data acquisition instrument and a PC (personal computer) which are sequentially connected; the strain gauge is arranged on the steel rail test piece and used for measuring stress data which changes constantly in the process of the steel rail longitudinal resistance test; the data acquisition instrument is used for acquiring stress data which changes constantly in the process of the longitudinal resistance test of the steel rail; the PC is used for recording and storing stress data of the steel rail longitudinal resistance change at any moment in the test process in real time.

10. The analysis system for the steel rail longitudinal resistance test based on the image recognition is characterized in that the analysis result is a stress-strain curve graph.

Images