Search

CN-121977481-A - Tunnel lining surface flatness detection method based on three-dimensional laser scanning

CN121977481ACN 121977481 ACN121977481 ACN 121977481ACN-121977481-A

Abstract

The invention relates to the technical field of tunnel engineering construction quality detection, and discloses a tunnel lining surface flatness detection method based on three-dimensional laser scanning, which is used for acquiring three-dimensional point cloud data of a tunnel, extracting a central axis of the tunnel based on a bilinear projection method and combining a RANSAC algorithm, converting the three-dimensional point cloud data from a world coordinate system XYZ to a local coordinate system X ́ Y ́ Z along the trend of the tunnel, carrying out equidistant slicing along the Z axis direction of the local coordinate system, then carrying out left-right separation operation, and equally dividing along the axial direction of the tunnel to generate a series of independent point cloud fragments, traversing all point pair combinations of any two points in the point cloud fragments for each point cloud fragment to obtain a final stable posture after rotation-translation transformation, and calculating the vertical distance from each point in each point cloud fragment to a Y ́ =0 datum plane under the final stable posture so as to represent the flatness condition corresponding to the point cloud fragment.

Inventors

  • MENG FANZHI
  • WANG LEI
  • HUANG JINKUN
  • CHENG ANCHUN
  • CHEN YANG

Assignees

  • 上海工程技术大学

Dates

Publication Date
20260505
Application Date
20260407

Claims (8)

  1. 1. A tunnel lining surface flatness detection method based on three-dimensional laser scanning is characterized by comprising the following steps: Step one, three-dimensional point cloud data of a tunnel are obtained; Step two, extracting a central axis of a tunnel based on a bilinear projection method and combining a RANSAC algorithm, and converting the three-dimensional point cloud data from a world coordinate system XYZ to a local coordinate system X ́ Y ́ Z along the trend of the tunnel; step three, equidistant slicing is carried out along the Z-axis direction of the local coordinate system, then left-right separation operation is carried out, equal-length division is carried out along the tunnel axis direction, and a series of independent point cloud fragments are generated; Step four, traversing all point pair combinations of any two points in each point cloud segment to obtain a final stable gesture after rotation translation transformation; and fifthly, calculating the vertical distance between each point in each point cloud segment and a y ́ =0 reference plane under the final stable gesture, so as to represent the flatness condition corresponding to the point cloud segment.
  2. 2. The method for detecting the flatness of a tunnel liner surface based on three-dimensional laser scanning of claim 1, wherein in the fourth step, a point cloud segment is projected to an X ́ OY ́ plane, all combinations of point pairs are traversed, and a rotation angle required for rotating a two-point connecting line of each point pair to a horizontal state is calculated And carrying out rotation translation transformation on the cloud segment of the point so that after the two points fall on a y ́ =0 reference plane, calculating a corresponding stability score under the posture at the moment, and selecting the posture with the highest stability score as the final stable posture.
  3. 3. The method for detecting the flatness of a tunnel liner surface based on three-dimensional laser scanning of claim 2, wherein when performing rotational-translational transformation, first, the point cloud segment is rotated integrally by an angle θ in a horizontal direction to make the line connecting the point pairs in a horizontal state, and then the point cloud segment is translated in a Y ́ axis direction to make the two points fall on a Y ́ =0 reference plane, i.e. the ordinate thereof is zeroed.
  4. 4. The method for detecting the flatness of a tunnel lining surface based on three-dimensional laser scanning of claim 2, wherein the two points of any pair of points are respectively And The rotation angle is calculated using the following formula , Calculating a stability score using the following formula , Wherein, the Representing the geometric center point of the point cloud segment to a support line segment The difference between the abscissa X ́ of the midpoints; Representing support line segments Is a length of (2); Representing a numerical tolerance.
  5. 5. The method for detecting the flatness of a tunnel lining surface based on three-dimensional laser scanning according to claim 1, wherein the local coordinate system X ́ Y ́ Z is set to be an origin of a world coordinate system, a tunnel central axis trend is an X ́ axis, a vertical upward direction, namely a gravity reverse direction is a Z axis, a Y ́ axis is determined by the X ́ axis and the Z axis according to a right-hand rule, In the third step, a three-dimensional point cloud in a local coordinate system X ́ Y ́ Z is sliced along the Z axis direction by using a slice with a preset thickness, then left-right separation operation is performed by using the Y ́ axis as a boundary, equal-length division is performed along the tunnel axis according to the standard length of l=2m, a series of independent point cloud fragments are generated, finally, point cloud fragments with the number of points smaller than a set threshold are removed, and the rest point cloud fragments are subjected to data density optimization one by adopting a voxel downsampling algorithm, so that a final point cloud fragment is obtained.
  6. 6. The method for detecting the flatness of a tunnel lining surface based on three-dimensional laser scanning of claim 1, wherein in the second step, three-dimensional point cloud data of a tunnel are orthogonally projected to an XOZ plane and a YOZ plane, then a central line of two side contour boundaries of the tunnel is extracted in the XOZ plane as a horizontal candidate line, a central line of upper and lower contour boundaries of the tunnel is extracted in the YOZ plane as a vertical candidate line, three-dimensional space discrete points (X, Y, Z) corresponding to the central axis of the tunnel are reconstructed based on the horizontal candidate line and the vertical candidate line, and finally a RANSAC algorithm is adopted to perform space straight line fitting on the discrete points to obtain a straight line where the central axis is located.
  7. 7. The method for detecting the flatness of a tunnel lining surface based on three-dimensional laser scanning of claim 6, wherein any point on a central axis is selected as an intercept point, a direction vector of the intercept point is an extending direction of a tunnel, three-dimensional point cloud data under a world coordinate system is converted to a local coordinate system through rigid transformation operation, namely an intercept point B of the three-dimensional point cloud is moved to an origin of the local coordinate system through translation operation, and rigid rotation is performed on the translated three-dimensional point cloud through rotation operation, so that the direction vector of the central axis of the three-dimensional point cloud is completely overlapped with an X ́ axis of the local coordinate system.
  8. 8. The method for detecting the flatness of the surface of a tunnel lining based on three-dimensional laser scanning of claim 1, wherein in the first step, the number of neighborhood points is set to be 20, the standard deviation multiple is set to be 1.5, dust and splash noise points are effectively filtered, in the data acquisition process, a multi-section stepping scanning method is adopted, the overlapping area is required to be more than or equal to 30%, multi-section three-dimensional point cloud data of the tunnel are obtained, then point cloud denoising is carried out based on a statistical outlier removal algorithm SOR, and finally the multi-section three-dimensional point cloud data are spliced based on an ICP algorithm, so that complete three-dimensional point cloud data of the tunnel are obtained.

Description

Tunnel lining surface flatness detection method based on three-dimensional laser scanning Technical Field The invention belongs to the technical field of tunnel engineering construction quality detection, and particularly relates to a tunnel lining surface flatness detection method based on three-dimensional laser scanning. Background In tunnel engineering construction, the flatness of the surface of the lining is a key quality index affecting the safety and waterproof performance of the structure. According to the specification of acceptance inspection of construction quality of concrete structure engineering (GB 50204), a combined measurement method of a 2-meter guiding rule and a wedge-shaped feeler gauge is adopted for flatness detection, and the physical principle is that after the rigid guiding rule is contacted with a salient point of a measured surface, the maximum clearance value between the guiding rule and the surface is measured. With the development of measurement technology, although the flatness detection method based on three-dimensional laser scanning is gradually popularized, in the prior art, the flatness is characterized by adopting a least square method to fit a tunnel curved surface and then calculating the vertical distance from a point cloud to a fitting surface, the method essentially only reflects the overall fluctuation condition of the surface, not only can the physical mechanism of a guiding rule specified in the national standard for supporting and contacting with a convex point be not reproduced, but also the influence of the local inclination of the tunnel on the posture of the guiding rule is not fully considered to cause the mixing of a false inclination component in a calculated gap value, and finally, systematic deviation of +/-8% - +/-15% exists between a detection result and a manual actual measurement value, so that the requirement of high-precision acceptance is difficult to meet. There is a need for a digital inspection method that can strictly follow the principle of "bump support, gravity direction measurement" to bridge the gap between three-dimensional scanning techniques and manual measurement standards. Disclosure of Invention The invention provides a tunnel lining surface flatness detection method based on three-dimensional laser scanning, which solves the problems that the existing method cannot reproduce the physical mechanism of a guiding rule regulated in the national standard for 'bump supporting contact', has low measurement precision, is difficult to meet actual demands and the like. In order to achieve the above purpose, the present invention provides the following technical solutions: a tunnel lining surface flatness detection method based on three-dimensional laser scanning comprises the following steps: Step one, three-dimensional point cloud data of a tunnel are obtained; Step two, extracting a central axis of a tunnel based on a bilinear projection method and combining a RANSAC algorithm, and converting the three-dimensional point cloud data from a world coordinate system XYZ to a local coordinate system X ́ Y ́ Z along the trend of the tunnel; step three, equidistant slicing is carried out along the Z-axis direction of the local coordinate system, then left-right separation operation is carried out, equal-length division is carried out along the tunnel axis direction, and a series of independent point cloud fragments are generated; Step four, traversing all point pair combinations of any two points in each point cloud segment to obtain a final stable gesture after rotation translation transformation; and fifthly, calculating the vertical distance between each point in each point cloud segment and a y ́ =0 reference plane under the final stable gesture, so as to represent the flatness condition corresponding to the point cloud segment. Further, in the fourth step, the point cloud segment is projected to the X ́ OY ́ plane, all the combinations of the point pairs are traversed, and the rotation angle required for rotating the two-point connection line of each point pair to a horizontal state is calculated firstAnd carrying out rotation translation transformation on the cloud segment of the point so that after the two points fall on a y ́ =0 reference plane, calculating a corresponding stability score under the posture at the moment, and selecting the posture with the highest stability score as the final stable posture. Further, when the rotation translation conversion is performed, firstly, the point cloud segment is integrally rotated by an angle θ in the horizontal direction so that the connecting line of the point pairs is in a horizontal state, and then the point cloud segment is translated along the Y ́ axis direction so that the two points fall on the Y ́ =0 reference plane, namely, the ordinate thereof is zeroed. Further, the two points of any point pair are respectivelyAndThe rotation angle is calculated using the following formula, Calcula