Pipeline displacement real-time measurement system and method based on three-dimensional vision

Abstract

The invention belongs to the technical field of pipeline displacement measurement, and particularly relates to a pipeline displacement real-time measurement system and method based on three-dimensional vision. Comprising a reference plate, the reference plate is connected with a base, the base is fixed to a measured pipeline, a thermoluminescence mark point is arranged on the surface of the reference plate, when the reference plate is installed on the measured pipeline, heat of the measured pipeline is transmitted to the mark point on the reference plate through a metal screw, the mark point emits radiation light, and a binocular camera is fixed to a building through a support. The binocular camera obtains a reference plate mark point image, image data are transmitted to the PoE switch through a POE signal line, and the PoE switch transmits the image data to the upper computer. The method has the advantage that displacement in three directions can be measured at the same time. A mark point is arranged through the heat-conducting luminescent material, heat of the pipeline is transmitted to the mark point through the metal screw to enable the mark point to emit light, and the binocular camera captures light of the mark point.

Description

Technical Field

[0001] This invention belongs to the field of pipeline displacement measurement technology, specifically relating to a real-time pipeline displacement measurement system and method based on three-dimensional stereo vision. Background Technology

[0002] In the nuclear power industry, the condition of pipelines is of great significance to the safe and stable operation of nuclear power plants. When nuclear power plants are under commissioning or in operation, they are affected by changing operating conditions and installation errors, which can cause pipelines to displace in different directions. In order to accurately assess the impact of pipeline displacement on system safety, it is necessary to measure and monitor pipeline displacement values ​​in real time.

[0003] Currently, nuclear power plants typically determine pipeline displacement by manually reading the scale of the pipeline supports and hangers. This method is not only time-consuming but also poses certain dangers in the relevant areas of the nuclear power plant, potentially causing burns or abnormal radiation exposure.

[0004] Simultaneously, the use of laser displacement sensors for pipeline displacement monitoring and measurement has certain application scenarios. A reference plate is placed on the pipeline using clamps, and a laser displacement sensor is positioned directly in front of the reference plate. The pipeline displacement value is recorded in real time through laser reflection. This method can measure pipeline displacement values ​​in real time to a certain extent, but it has certain limitations: when the reference plate is tilted by external forces, the laser emitted by the laser displacement sensor cannot be reflected, affecting the measurement results. Furthermore, this method only monitors pipeline displacement in one dimension; when the pipeline moves simultaneously in three directions, it cannot accurately obtain the pipeline's status. Summary of the Invention

[0005] The purpose of this invention is to provide a real-time pipeline displacement measurement system and method based on three-dimensional stereo vision, which can overcome the problems existing in current pipeline displacement monitoring and be used for real-time three-dimensional thermal displacement measurement of nuclear power pressure pipelines.

[0006] The technical solution of the present invention is as follows: A real-time pipeline displacement measurement system based on three-dimensional stereo vision includes a reference plate, which is connected to a base. The base is fixed on the pipeline under test. Thermoluminescent markers are set on the surface of the reference plate. When the reference plate is installed on the pipeline under test, the heat of the pipeline under test is transferred to the markers on the reference plate through a metal screw. The markers emit radiation light. A binocular camera is fixed on the structure through a bracket. The binocular camera acquires images of the markers on the reference plate. The image data is transmitted to a PoE switch using a PoE signal line. The PoE switch transmits the image data to a host computer.

[0007] The reference plate is connected to the base via screws.

[0008] The base is fixed to the pipe being tested by clamps.

[0009] The binocular camera is fixed to the structure by a bracket.

[0010] The binocular camera is placed on the front of the reference plate.

[0011] A method for real-time measurement of pipeline displacement based on three-dimensional stereo vision includes the following steps:

[0012] Step 1: Fix the reference plate to the base using screws, and fix the base to the pipe being tested by clamps;

[0013] Step 2: Set multiple heat-emitting markers on the reference plate. When the heat from the pipe is transferred to the reference plate through the base and screw, the markers will emit light when heated.

[0014] Step 3: The dual-sided camera acquires images of the luminous points on the reference board, and the images are then subjected to dynamic thresholding segmentation.

[0015] Step 4: Calibrate the double-sided camera using a standard checkerboard pattern;

[0016] The standard checkerboard pattern is photographed from different angles. The vertices of the checkerboard pattern are extracted by the image processing program of the host computer. The correspondence between the vertices of the checkerboard pattern and the corresponding points on the image is established. The distortion coefficients and intrinsic and extrinsic parameters of the camera are analyzed, and the relationship between the camera matrix and the world coordinate system, the camera coordinate system and the intrinsic parameter matrix is ​​obtained.

[0017]

[0018] Where P is the camera matrix, R and t are the rotation and translation matrices between the world coordinate system and the camera coordinate system, and K is the intrinsic parameter matrix;

[0019] Step 5: Install the binocular camera with completed internal and external parameter calibration on the structure next to the reference board, ensuring that the reference board is within the field of view of the binocular camera;

[0020] Step 6: Transmit the reference board image data to the PoE switch via the PoE signal line, and finally transmit it to the host computer;

[0021] Step 7: Complete the stereo camera image processing on the host computer;

[0022] Step 8: Based on the three-dimensional coordinates (X, Y, Z) of different marker points i ,Y i Z i The average displacement value in each direction of each marker point is calculated, and the displacement and direction of the actual pipeline are obtained.

[0023] Step 9: When the pipe displacement causes relative displacement between the insulation layer and the pipe, resulting in bending of the screw or tilting of the reference plate, repeat step 7 to re-perform positioning and displacement measurement.

[0024] Step 3 includes first separating the color channels of the light-emitting points on the reference plate, using R channel data, and after image enhancement and denoising, using 5*5 pixel grayscale units as the basic scale for image segmentation to obtain micro-region segmentation results. The local threshold segmentation results are compared with the threshold standard deviation to obtain the final segmentation results. The different segmented regions are sorted, and the coordinate information of different light-emitting points is calculated to provide input for subsequent three-dimensional coordinate calculation.

[0025] Step 7 includes removing noise from the image using median filtering or Gaussian filtering, then using a threshold segmentation algorithm to obtain the binarized regions of different marker points, calculating the two-dimensional center coordinates of different marker points, and obtaining the three-dimensional coordinate information of different marker points by combining the camera intrinsic parameters based on the differences in the two-dimensional center coordinates of marker points in images acquired by different targets in the camera.

[0026] The calculation in step 8 is as follows:

[0027] X = X i -X i+1

[0028] Y = Y i -Y i+1

[0029] Z = Z i -Z i+1

[0030] Where X, Y, and Z represent the three-dimensional displacement values ​​within the pipe space, X i ,Y i Z i The three-dimensional coordinate information of the marker point at the moment before the pipeline displacement, X i+1 ,Y i+1 Z i+1 This provides the three-dimensional coordinate information of the pipeline at a moment after it has been displaced.

[0031] The beneficial effects of this invention are as follows: A binocular camera is used to acquire the three-dimensional coordinate information of the reference plate markers. By calculating the parallax between two images, the positions of different markers are obtained. When the pipeline shifts, the actual displacement of the pipeline is obtained by calculating the distance between the reference plate and the binocular camera. This method can simultaneously measure displacement in three directions. Markers are set using heat-emitting materials. Heat from the pipeline is transferred to the markers through metal screws, causing them to emit light, which the binocular camera captures. Image processing is performed on the images acquired by the binocular camera. A dynamic threshold segmentation algorithm is used to segment the emitting points within the image, obtaining the coordinates of different markers and thus real-time coordinate information for different points. PoE network port power supply is used, with both data transmission and power supply using the same network cable, resulting in high data transmission efficiency and simple cabling. Through the binocular camera measurement method, when the pipeline reference plate shifts, the host computer can perform real-time correction and automatically capture the markers to complete the position coordinate calculation. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of a real-time pipeline displacement measurement system based on three-dimensional stereo vision provided by the present invention.

[0033] Figure 2 This is a schematic diagram of the reference board and marker points.

[0034] In the diagram: 1-structure, 2-binocular camera, 3-support, 4-POE signal line, 5-PoE switch, 6-host computer, 7-tested pipeline, 8-reference plate, 9-screw, 10-clamp. Detailed Implementation

[0035] The present invention will be further described in detail below with reference to specific embodiments.

[0036] A real-time pipeline displacement measurement system based on three-dimensional stereo vision includes a reference plate 8, which is connected to a base via a screw 9. The base is fixed to the pipeline under test 7 via a clamp 10. The screw 9 passes through an insulation layer. The final pipeline displacement is determined by measuring the displacement of the reference plate 8 outside the insulation layer. The surface of the reference plate 8 is equipped with thermoluminescent markers. When the reference plate 8 is installed on the pipeline under test 7, the heat from the pipeline under test 7 is transferred to the markers on the reference plate 8 through the metal screw 9, and the markers emit radiation light. A binocular camera 2 is fixed to a structure 1 near the pipeline under test 7 via a bracket 3, and the binocular camera 2 is placed in front of the reference plate 8. The system acquires images of the markers on the reference plate 8 and transmits the image data to a PoE switch 5 via a PoE signal line 4. The PoE switch 5 transmits the image data to a host computer 6. The host computer 6 performs image processing and template matching to obtain the positions of different markers, calculates the three-dimensional coordinates of different markers on the reference plate in real time, and obtains the three-dimensional displacement value of the pipeline by averaging.

[0037] A method for real-time measurement of pipeline displacement based on three-dimensional stereo vision includes the following steps:

[0038] Step 1: Fix the reference plate 8 to the base using screws 9, and fix the base to the pipe 7 to be tested by clamps 10.

[0039] Step 2: Multiple heat-emitting markers are set on the reference plate 8. When the heat from the pipe is transferred to the reference plate 8 through the base and metal screw 9, the markers emit light when heated.

[0040] Step 3: The dual-sided camera 2 acquires images of the light-emitting points on the reference board, and performs dynamic thresholding segmentation on the images.

[0041] First, the color channels of the luminous points on the reference board are separated using the R channel data. After image enhancement and denoising, a 5*5 pixel grayscale unit is used as the basic scale for image segmentation to obtain small-area segmentation results. The local threshold segmentation results are compared with the threshold standard deviation to obtain the final segmentation result. The different segmented regions are sorted, and the coordinate information of different luminous points is calculated for each, providing input for subsequent 3D coordinate calculations.

[0042] Step 4: Calibrate the double-sided camera using a standard checkerboard.

[0043] The standard checkerboard pattern is photographed from different angles. The vertices of the checkerboard pattern are extracted by the image processing program on the host computer. The correspondence between the vertices of the checkerboard pattern and the corresponding points on the image is established. The distortion coefficients and intrinsic and extrinsic parameters of the camera are analyzed, and the relationship between the camera matrix and the world coordinate system, the camera coordinate system and the intrinsic parameter matrix is ​​obtained.

[0044]

[0045] Where P is the camera matrix, R and t are the rotation and translation matrices between the world coordinate system and the camera coordinate system, and K is the intrinsic parameter matrix.

[0046] Step 5: Install the binocular camera 2, after completing the internal and external parameter calibration, on the structure 1 next to the reference plate 8, ensuring that the reference plate is within the field of view of the binocular camera 2.

[0047] Step 6: Transmit the reference board image data to the PoE switch 5 via the PoE signal line 4, and finally transmit it to the host computer 6.

[0048] Step 7: Complete the binocular camera image processing on the host computer 6.

[0049] First, median filtering or Gaussian filtering is used to remove noise from the image. Then, a threshold segmentation algorithm is used to obtain the binarized regions of different marker points, and the two-dimensional center coordinates of different marker points are calculated. Based on the differences in the two-dimensional center coordinates of marker points in images acquired from different targets in the camera, the three-dimensional coordinate information of different marker points is obtained by combining the camera intrinsic parameters.

[0050] Step 8: Based on the three-dimensional coordinates (X, Y, Z) of different marker points i ,Y i Z i The average displacement value of each marker point in each direction is calculated using the formula, and the displacement and direction of the actual pipeline are obtained.

[0051] X = X i -X i+1

[0052] Y = Y i -Y i+1

[0053] Z = Z i -Z i+1

[0054] Where X, Y, and Z represent the three-dimensional displacement values ​​within the pipe space, X i ,Y i Z i The three-dimensional coordinate information of the marker point at the moment before the pipeline displacement, X i+1 ,Y i+1 Z i+1 This provides the three-dimensional coordinate information of the pipeline at a moment after it has been displaced.

[0055] Step 9: When the pipe displacement causes relative displacement between the insulation layer and the pipe, resulting in bending of the screw, or when personnel accidentally touch the reference plate causing it to tilt, repeat step 7 to re-perform positioning and displacement measurement.

Claims

1. A real-time pipeline displacement measurement system based on three-dimensional stereo vision, characterized in that: The system includes a reference plate connected to a base, which is fixed to the pipe under test. The reference plate has thermoluminescent markers on its surface. When the reference plate is installed on the pipe under test, the heat from the pipe is transferred to the markers on the reference plate through a metal screw, causing the markers to emit radiation. A binocular camera is fixed to the structure by a bracket. The binocular camera acquires images of the markers on the reference plate and transmits the image data to a PoE switch via a PoE signal line. The PoE switch then transmits the image data to a host computer.

2. The real-time pipeline displacement measurement system based on three-dimensional stereo vision as described in claim 1, characterized in that: The reference plate is connected to the base via screws.

3. The real-time pipeline displacement measurement system based on three-dimensional stereo vision as described in claim 1, characterized in that: The base is fixed to the pipe being tested by clamps.

4. The real-time pipeline displacement measurement system based on three-dimensional stereo vision as described in claim 1, characterized in that: The binocular camera is fixed to the structure by a bracket.

5. The real-time pipeline displacement measurement system based on three-dimensional stereo vision as described in claim 1, characterized in that: The binocular camera is placed on the front of the reference plate.

6. A method for real-time measurement of pipeline displacement based on three-dimensional stereo vision, characterized in that, Includes the following steps: Step 1: Fix the reference plate to the base using screws, and fix the base to the pipe being tested by clamps; Step 2: Set multiple heat-emitting markers on the reference plate. When the heat from the pipe is transferred to the reference plate through the base and screw, the markers will emit light when heated. Step 3: The dual-sided camera acquires images of the luminous points on the reference board, and the images are then subjected to dynamic thresholding segmentation. Step 4: Calibrate the double-sided camera using a standard checkerboard pattern; The standard checkerboard pattern is photographed from different angles. The vertices of the checkerboard pattern are extracted by the image processing program of the host computer. The correspondence between the vertices of the checkerboard pattern and the corresponding points on the image is established. The distortion coefficients and intrinsic and extrinsic parameters of the camera are analyzed, and the relationship between the camera matrix and the world coordinate system, the camera coordinate system and the intrinsic parameter matrix is ​​obtained. Where P is the camera matrix, R and t are the rotation and translation matrices between the world coordinate system and the camera coordinate system, and K is the intrinsic parameter matrix; Step 5: Install the binocular camera with completed internal and external parameter calibration on the structure next to the reference board, ensuring that the reference board is within the field of view of the binocular camera; Step 6: Transmit the reference board image data to the PoE switch via the PoE signal line, and finally transmit it to the host computer; Step 7: Complete the stereo camera image processing on the host computer; Step 8: Based on the three-dimensional coordinates (X, Y, Z) of different marker points i ,Y i Z i The average displacement value in each direction of each marker point is calculated, and the displacement and direction of the actual pipeline are obtained. Step 9: When the pipe displacement causes relative displacement between the insulation layer and the pipe, resulting in bending of the screw or tilting of the reference plate, repeat step 7 to re-perform positioning and displacement measurement.

7. The method for real-time measurement of pipeline displacement based on three-dimensional stereo vision as described in claim 6, characterized in that: Step 3 includes first separating the color channels of the light-emitting points on the reference plate, using R channel data, and after image enhancement and denoising, using 5*5 pixel grayscale units as the basic scale for image segmentation to obtain micro-region segmentation results. The local threshold segmentation results are compared with the threshold standard deviation to obtain the final segmentation results. The different segmented regions are sorted, and the coordinate information of different light-emitting points is calculated to provide input for subsequent three-dimensional coordinate calculation.

8. The method for real-time measurement of pipeline displacement based on three-dimensional stereo vision as described in claim 6, characterized in that: Step 7 includes removing noise from the image using median filtering or Gaussian filtering, then using a threshold segmentation algorithm to obtain the binarized regions of different marker points, calculating the two-dimensional center coordinates of different marker points, and obtaining the three-dimensional coordinate information of different marker points by combining the camera intrinsic parameters based on the differences in the two-dimensional center coordinates of marker points in images acquired by different targets in the camera.

9. The method for real-time measurement of pipeline displacement based on three-dimensional stereo vision as described in claim 6, characterized in that, The calculation in step 8 is as follows: X=X i -X i+1 Y / Y i -AND i+1 Z=Z i -WITH i+1 Where X, Y, and Z represent the three-dimensional displacement values ​​within the pipe space, X i, Y i Z i The three-dimensional coordinate information of the marker point at the moment before the pipeline displacement, X i+1 ,Y i+1 Z i+1 This provides the three-dimensional coordinate information of the pipeline at a moment after it has been displaced.

Images