CN109373921B - Tunnel monitoring method and device - Google Patents

Abstract

The application provides a tunnel monitoring and analyzing method and a device, wherein the method comprises the following steps: acquiring original section point cloud data of a tunnel; respectively performing point cloud splicing operation and reflectivity image generation operation according to the acquired data to generate a tunnel integral point cloud model and a tunnel pipe wall expansion diagram; carrying out binarization processing on the tunnel tube wall development image, carrying out parameter mapping on a binary image by using Hough transform, and extracting a straight line segment of the binary image within a preset angle range; extracting a single segment image in a tunnel tube wall expansion diagram according to the straight line segment, and respectively extracting sub-images of the top sealing block and the connecting blocks at two sides in the single segment image according to the single segment image; dividing the tunnel integral point cloud model into point cloud small block models by using the capping block, the connection block subimages at two sides and the pixel positions of the single segment images; and calculating the slab staggering height according to the fitting planes of the point cloud small block models and the distance between the fitting planes, and analyzing the deformation condition of the tunnel segment.

Description

A kind of tunnel monitoring method and device

technical field

The present application relates to the field of computer data processing, and in particular, to a tunnel monitoring method and device.

Background technique

With the continuous increase of urban rail transit mileage, rail transit is becoming more and more important for people's travel. When the rail transit system faces the constant deformation and aging of the tunnel, the daily maintenance of the tunnel system is very important. By maintaining the tunnel system, not only can the rail transit system support operation, but also safety accidents caused by human negligence can be greatly reduced.

The daily maintenance items of general tunnels include deformation detection, misplacement, block drop, water leakage, etc. Due to the partial and overall dislocation of the segment, the overall deformation of the segment is the direct cause of block loss, water leakage and other diseases. Therefore, the monitoring of the health of the tunnel is usually achieved by monitoring the geometric deformation index.

Most of the tunnel monitoring methods used in traditional tunnel image-based disease monitoring use macroscopic geometric deformation, local disease characteristics, etc. to judge the health status of the tunnel, which can only be detected after the occurrence or aggravation of the disease, which is a passive method. The monitoring results are not precise enough, so a more refined and rational analysis method is urgently needed.

SUMMARY OF THE INVENTION

Embodiments of the present application provide a tunnel monitoring method and device, which can improve the fineness of tunnel monitoring results.

In one embodiment, a tunnel monitoring method includes:

Collect the original section point cloud data of the tunnel;

According to the collected original section point cloud data, the point cloud splicing operation and the reflectivity image generation operation are respectively performed to generate the overall point cloud model of the tunnel and the expanded view of the tunnel wall;

Performing binarization processing on the expanded image of the tunnel wall to obtain a binary image, performing parameter mapping on the binary image using Hough transform, and extracting a straight line segment of the binary image within a preset angle range;

Extracting a single segment image in the expanded view of the tunnel wall according to the straight line segment, and extracting a capping block sub-image and two side-connecting block sub-images from the single segment image;

The position of the corresponding tunnel point cloud is determined by using the pixel positions of the capping block sub-image, the two-side connection block sub-image and the single segment image, and the tunnel overall point cloud model is divided into point cloud small block models;

Fitting a plane according to the point cloud small block model, calculating the height of misalignment according to the distance between the fitting planes, and determining the deformation of the segment of the tunnel according to the height of the misalignment.

In one embodiment, the collection of the original cross-section point cloud data of the tunnel is realized by a scanner, which specifically includes: determining the moving speed of the mobile platform and the scanning parameters of the scanner according to the preset monitoring accuracy; The scanner is calibrated; when the mobile platform is moved, the scanner scans to collect the original section point cloud data.

In one embodiment, according to the collected original section point cloud data, point cloud splicing operation and reflectivity image generation operation are respectively performed to generate the overall point cloud model of the tunnel and the expanded view of the tunnel wall, including: splicing a complete tunnel A point cloud that uses reflectivity data to generate an unwrapped image of the tunnel wall.

In one embodiment, extracting a single segment image in the expanded view of the tunnel wall according to the straight segment includes: removing redundant straight segments using prior information, and extracting a single segment in the expanded view of the tunnel wall image.

In one embodiment, the prior information includes: the design width of the segment and the average width of the pipe seam.

In one embodiment, the method further comprises: extracting standard block sub-images from the single segment image.

In one embodiment, the extraction of the capping block sub-image and the two-side connection block sub-images from a single segment image, respectively, includes:

adopting a function to match the capping block edge in the single segment image with the capping block design template to determine the position of the capping block in the tunnel unfolded image, and extract the capping block sub-image at the position;

According to the position of the capping block and the symmetry of the segment, determine the position of the connecting block on both sides and the position of the standard block, extract the sub-images of the two-side connecting block at the position according to the position of the connecting block on both sides, The standard block sub-image at this position is extracted.

In one embodiment, the small point cloud model includes: a point cloud model of a single segment and a point cloud model of a single connection block.

In one embodiment, the position of the corresponding tunnel point cloud is determined by using the pixel positions of the capping block sub-image, the two-side connection block sub-image and the single segment image, and the overall point cloud model of the tunnel is divided into The point cloud small block model includes: using the capping block sub-image and the connecting block sub-images on both sides to divide the overall point cloud model of the tunnel into a point cloud model of a single connected small block, according to the pixels of the single segment image The location divides the overall point cloud model of the tunnel into point cloud models of individual segments.

In one embodiment, a tunnel monitoring device includes: a collection module, a data splicing module, a line and edge detection module, a segment and connection block determination module, a point cloud segmentation module, and a monitoring and analysis module;

The acquisition module is used to acquire the original section point cloud data of the tunnel;

The data splicing module is configured to perform point cloud splicing operation and reflectivity image generation operation respectively according to the collected original section point cloud data, and generate an overall point cloud model of the tunnel and an expanded view of the tunnel pipe wall;

The straight line and edge detection module is used to perform binarization processing on the expanded image of the tunnel wall to obtain a binary image, use Hough transform to perform parameter mapping on the binary image, and extract the binary image in the A straight line segment within a preset angle range;

The segment and connection block determination module is used for extracting a single segment image in the expanded view of the tunnel wall according to the straight segment, and extracting a capping block sub-image and two sides from the single segment image respectively concatenate block subimages;

The point cloud segmentation module is used to determine the position of the corresponding tunnel point cloud by using the pixel positions of the capping block sub-image, the two-side connection sub-image and the single segment image, and divides the overall point cloud model of the tunnel into point cloud patch model;

The monitoring and analysis module is used for fitting a plane according to the point cloud small block model, calculating the height of the misalignment according to the distance between the fitting planes, and determining the deformation of the segment of the tunnel according to the height of the misalignment.

In the above embodiments, by using the tunnel monitoring method and device provided by the embodiments of the present application, the laser reflectivity image data can be used to realize the point cloud segmentation of each connection block in the tunnel segment, and to calculate the local and overall errors of the tunnel segment. According to the height of the staggered platform, the overall deformation of the segment is analyzed in a refined manner, which improves the precision of the tunnel monitoring results.

Description of drawings

The accompanying drawings constituting a part of the present application are used to provide further understanding of the present application, and the schematic embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation of the present application. In the attached image:

1 is a flowchart of a tunnel monitoring method according to an embodiment of the present specification;

2 is a schematic structural diagram of a collection device in an embodiment of the present invention;

3 is a schematic diagram of a tunnel overall point cloud model generated by splicing section point cloud data in an embodiment of the present specification;

FIG. 4 is an expanded image of the tunnel wall generated by the reflectance intensity values of each point cloud in an embodiment of the present specification;

5 is a schematic diagram of the constitution of a single segment of the tunnel and each connection block in an embodiment of the specification;

6 is a schematic diagram of a segment and each connection block extracted in an embodiment of this specification;

FIG. 7 is a schematic diagram of a segmented point cloud block model according to an embodiment of this specification;

FIG. 8 shows the staggered height calculated according to the fitting plane of the point cloud patch model according to an embodiment of the present specification;

FIG. 9 is a block diagram of a tunnel monitoring device according to an embodiment of the present specification.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present application more clear, the present application will be further described in detail below with reference to the embodiments and the accompanying drawings. Here, the exemplary embodiments and descriptions in this specification are used to explain the present application, but are not intended to limit the present application.

FIG. 1 is a flowchart of a tunnel monitoring method according to an embodiment of the present specification. As shown in FIG. 1 , the tunnel monitoring method may include the following steps.

S101: Collect original section point cloud data of the tunnel.

The raw section point cloud data of the tunnel can be collected.

In one embodiment, a scanner may be used to collect the raw section point cloud data. FIG. 2 is a schematic structural diagram of a collection device in an embodiment of the present invention. Referring to Figure 2, the scanner may be mounted on a capture device that also includes a mobile platform.

Specifically, the moving speed of the mobile platform and the scanning parameters of the scanner can be determined according to the preset monitoring accuracy; the scanner is calibrated by using the calibration position; when the mobile platform is moved, the scanner scans to collect The original section point cloud data.

Wherein, the monitoring precision may be the precision of an image picture. For example, it can be 2 mm, then, the moving speed of the moving platform can be 0.25 m/s, the scanning resolution of the scanner can be set to 1/4, and the scanning quality can be set to: 3.

Using the calibration position to calibrate the scanner may include: pre-positioning a calibration object at the scanning starting point position. By calibrating the scanner, the point cloud data and the image data can be matched, and at the same time, it can be used to correct the compression ratio of the data in the Y-axis direction.

In one embodiment, the collected raw section point cloud data may be stored as a point cloud file of a preset size. For example, each point cloud file can be stored as one fls type file per 100 sections, that is, 426,800 points.

In one embodiment, the point cloud data includes: 3 coordinate data relative to the center of the circle and 1 reflectance intensity value. The section data can be represented as {X _t , Y _t , Z _t , I _t }, wherein X _t can represent the X-axis coordinate value, Y _t can represent the Y-axis coordinate value, and Z _t can represent the Z-axis coordinate value. It can represent the reflectance _intensity value.

S102: According to the collected point cloud data of the original section, perform point cloud splicing operation and reflectivity image generation operation respectively to generate an overall point cloud model of the tunnel and an expanded view of the tunnel wall.

According to the moving speed V _c of the mobile platform, the number of point clouds N _p generated by the scanner per second and the number of point clouds C _single included in a single section, the point cloud coordinates of the Y axis can be calculated. The calculation of the point cloud coordinates of the Y-axis can use the following formula:

Further, the sections can be spliced into the overall point cloud model of the tunnel and the expanded view of the tunnel wall generated based on the reflectivity according to the time stamp order corresponding to the sections.

Specifically, the point cloud coordinate data in the section data can be used to stitch the complete tunnel point cloud, and the reflectivity data in the section data can be used to generate an expanded image of the tunnel wall.

In one embodiment, the cross-sectional data {X _t , Z _t } can be arranged in time sequence {t ₁ , t ₂ , t ₃ , ..., t _n }, and the cross-sectional data can be spliced into a complete piece according to the calculated Y _t The tunnel point cloud of {X ₁ , Y ₁ , Z ₁ , X ₂ , Y ₂ , Z ₂ , …, X _n , Y _n , Z _n }. The Y-axis values are continuously accumulated, namely: Y ₁ =Y _t ; Y ₂ =Y ₁ +Y _t ; Y _n =Y ₁ +Y ₂ +...+Y _n-1 .

In one embodiment, the number of point clouds C _sc contained in a single section can be taken as the number of pixels in the Y-axis direction of the image, then, a single point cloud of a single section can be regarded as a pixel on the Y-axis, that is, a single section {X _t , Z _t } constitutes a column of pixels in the image. The intensity value of a single pixel is I _t , resulting in an image of C _sc ×W pixels, that is, an expanded view of the tunnel wall. Wherein, the number of pixels W in the X-axis direction can be selected according to actual needs.

FIG. 3 is a schematic diagram of an overall point cloud model of a tunnel generated by splicing section point cloud data according to an embodiment of the present specification. FIG. 4 is an unfolded image of the tunnel wall generated from the reflectance intensity values of each point cloud in an embodiment of the present specification.

In another embodiment, the tunnel monitoring method may further include: correcting the point cloud data so that one point cloud corresponds to one pixel.

For example, suppose that the total number of broken surfaces in the original data is W _s , and in the spliced point cloud data, the total length of the tunnel is M _s . Then, before the generated original image is interpolated, the size of the image should be C _sc ×W _s , that is, there are W _s pixels distributed in the interval of length M _s , that is, a single pixel in the X-axis direction needs to be stretched is the length of M _s /W pixels, so that the 3D point cloud and the 2D image can be registered with each other.

In another embodiment, the tunnel monitoring method may further include: performing interpolation processing on the corrected expanded view of the tunnel wall.

S103: Binarize the expanded image of the tunnel wall to obtain a binary image, use Hough transform to perform parameter mapping on the binary image, and extract a straight line segment of the binary image within a preset angle range .

In one embodiment, the Canny edge detection operator may be used to perform binarization processing on the expanded image of the tunnel wall. For example, the Gaussian radius of the Canny edge detection operator can be set to 2, the low threshold can be set to 40, and the high threshold can be set to 80.

The binary image after the binarization process can be saved as:

_E d=((x ₁ ,y ₁ ),(x ₂ ,y ₂ ),…,(x _n ,y _n ))

Since there are vertical or nearly vertical straight line segments in the binary image, here we need to replace the traditional straight line equation with the polar coordinate parametric equation. The parametric equation in the technique can use the following formula:

ρ=x cosθ+y sinθ;

In the above formula, ρ represents the vertical distance from the origin to the line, and θ represents the angle between the x-axis and the line. θ can be used as a preset angle, and the value range of θ can be [+10°, -10°] and [+80°, -80°]. Through the above formula, horizontal straight lines (or approaching horizontal straight lines), vertical straight lines and edge parts of trapezoidal capping blocks (abbreviated as SF) in the binary image can be detected respectively.

By limiting the range of polar coordinates within the above-mentioned preset angle range, the influence of noise data can be greatly reduced, and the monitoring efficiency can be improved.

S104: Extract a single segment image in the expanded view of the tunnel wall according to the straight line segment, and extract a capping block sub-image and a two-side connection block sub-image respectively from the single segment image.

A priori information can be used to remove redundant straight line segments to determine a single segment image in the expanded view of the tunnel wall.

In one embodiment, the design width of the segment (abbreviated as d _cs ) and the average width of the pipe seam (abbreviated as g) may be used as prior information.

Determining a single segment image in the expanded view of the tunnel wall by removing redundant straight line segments through prior information may specifically include: filtering out [d _cs -g,d _cs + on the expanded view of the tunnel wall The vertical line segment within the range of g] is the position of a single segment, and the image at the position of the single segment on the expanded view of the tunnel wall is the image of the single segment.

In another embodiment, the single segment image may also be stored. For example, the single segment image may be stored as _Is .

The capping block sub-image and the two-side connecting block sub-images may be extracted from the single segment image, respectively.

In one embodiment, sub-images of standard blocks in the single segment image may also be extracted according to the single segment image.

An optimization function can be used to match the capping block in a single segment image to a capping block (SF) design template to determine the location of the capping block in the tunnel deployment image, and extract the capping block sub-image at that location.

Specifically, the capping block design template can be used as an n×n template operator K _design , an n×n pixel block K _h is taken in the picture, and a convolution operation is performed with the template operator to traverse the entire tunnel The expanded view of the pipe wall, the position of the capping block is determined by the following formula:

The capping block template operator K _design may be a matrix of size n×n, including [0,1] elements, where the edge part may be represented by 1, and the rest part may be filled with 0.

Further, the positions of the connecting blocks (SL1, SL2) and the standard blocks (SB1, SB2) on both sides can be determined according to the position of the capping block (SF) and the symmetry of the segment, and extracted according to the positions of the connecting blocks on both sides. The two sides of the position are connected to the block sub-image, and the standard block sub-image at the position is extracted according to the position of the standard block. FIG. 5 is a schematic diagram of the constitution of a single segment of a tunnel and each connection block in an embodiment of the specification. Figure 5 includes: capping block (SF), connecting block (SL1, SL2), standard block (SB1, SB2, GD). FIG. 6 is a schematic diagram of a segment and each connection block extracted in an embodiment of the present specification. Figure 6 includes: capping block (SF), connection blocks (SL1, SL2), and standard blocks (SB1, SB2).

S105: Determine the position of the corresponding tunnel point cloud by using the pixel positions of the capping block sub-image, the two-side connection block sub-image and the single segment image, and divide the overall point cloud model of the tunnel into small point cloud models.

The point cloud small block model may include: a point cloud model of a single segment and a point cloud model of a single connected small block.

Specifically, the overall point cloud model of the tunnel can be divided into a single connected small block point cloud model by using the capping block sub-image and the connecting block sub-images on both sides, and the The overall point cloud model of the tunnel is divided into point cloud models of individual segments.

FIG. 7 is a schematic diagram of a segmented point cloud small block model according to an embodiment of the present specification.

S106 : Fitting a plane according to the point cloud small block model, calculating a height of a misalignment according to the distance between the fitting planes, and determining the deformation of the segment of the tunnel according to the height of the misalignment.

Specifically, according to the fitting plane of the point cloud small block model, the distance between the point cloud small block model and the point cloud plane around the point cloud small block model can be calculated, and the tunnel pipe can be judged according to the height of the staggered platform. sheet deformation. So as to realize the deformation monitoring of the tunnel segment.

FIG. 8 shows the height of the staggered stage calculated according to the fitting plane of the point cloud patch model according to an embodiment of the present specification.

From the above description, it can be seen that the embodiments of the present application achieve the following technical effects: the laser reflectivity image data can be used to realize the point cloud segmentation of each connection block in the tunnel segment, and the local and overall existence of the tunnel segment can be calculated. The height of the staggered platform, and the overall deformation of the segment is analyzed according to the height of the staggered platform, which improves the precision of the tunnel monitoring results.

Based on the same inventive concept, the embodiments of the present application also provide a tunnel monitoring device, as described in the following embodiments. Since the principle of solving the problem of the tunnel monitoring device is similar to that of the tunnel monitoring method, the implementation of the tunnel monitoring device may refer to the implementation of the tunnel monitoring method, and the repetition will not be repeated. As used below, the term "unit" or "module" may be a combination of software and/or hardware that implements a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, implementations in hardware, or a combination of software and hardware, are also possible and contemplated. FIG. 9 is a block diagram of a tunnel monitoring device according to an embodiment of the present specification. As shown in FIG. 9 , the tunnel monitoring device may include: a collection module 901, a data splicing module 902, a line and edge detection module 903, segment and Connect the block determination module 904 , the point cloud segmentation module 905 and the monitoring analysis module 906 .

The collection module 901 can be used to collect the original cross-section point cloud data of the tunnel.

The data splicing module 902 can be configured to perform point cloud splicing operation and reflectivity image generation operation respectively according to the collected original section point cloud data, and generate an overall point cloud model of the tunnel and an expanded view of the tunnel wall.

The straight line and edge detection module 903 can be used to perform binary processing on the expanded image of the tunnel wall to obtain a binary image, use Hough transform to perform parameter mapping on the binary image, and extract the binary image. A straight line segment of the image within a preset angle range.

The segment and connection block determination module 904 can be configured to extract a single segment image in the expanded view of the tunnel wall according to the straight line segment, and extract a capping block sub-image and a capping block sub-image respectively from the single segment image. Connect block subimages on both sides.

The point cloud segmentation module 905 can be used to determine the position of the corresponding tunnel point cloud by using the pixel positions of the capping block sub-image, the two-side connection sub-image and the single segment image, and divide the overall point cloud model of the tunnel. Segment the model into small pieces of point cloud.

The monitoring and analysis module 906 can be used to fit a plane according to the point cloud small block model, calculate the height of the staggered platform according to the distance between the fitted planes, and determine the segment of the tunnel according to the height of the staggered platform. Deformation situation. So as to realize the deformation monitoring of the tunnel segment.

Obviously, those skilled in the art should understand that each module or each step of the above-mentioned embodiments in this specification can be implemented by a general-purpose computing device, and they can be centralized on a single computing device, or distributed in multiple computing devices. Alternatively, they can be implemented with program code executable by a computing device on a network composed of The steps shown or described may be performed in the order shown or described, either by fabricating them separately into individual integrated circuit modules, or by fabricating multiple modules or steps of them into a single integrated circuit module. As such, embodiments of this specification are not limited to any particular combination of hardware and software.

The above descriptions are only preferred embodiments of the present application, and are not intended to limit the present application. For those skilled in the art, various modifications and changes may be made to the embodiments of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this application shall be included within the protection scope of this application.

Claims (10)

1. a tunnel monitoring method, is characterized in that, comprises: Collect the original section point cloud data of the tunnel; According to the collected point cloud data of the original section, the overall point cloud model of the tunnel is generated by performing the point cloud splicing operation, and the expanded view of the tunnel pipe wall is generated by the reflectivity image generation operation; Performing binarization processing on the expanded image of the tunnel wall to obtain a binary image, performing parameter mapping on the binary image using Hough transform, and extracting a straight line segment of the binary image within a preset angle range; Extracting a single segment image in the expanded view of the tunnel wall according to the straight line segment, and extracting a capping block sub-image and two side-connecting block sub-images from the single segment image; The position of the corresponding tunnel point cloud is determined by using the pixel positions of the capping block sub-image, the two-side connection block sub-image and the single segment image, and the tunnel overall point cloud model is divided into point cloud small block models; Fitting a plane according to the point cloud small block model, calculating the height of misalignment according to the distance between the fitting planes, and determining the deformation of the segment of the tunnel according to the height of the misalignment. 2 . A tunnel monitoring method according to claim 1 , wherein the acquisition of the original cross-section point cloud data of the tunnel is realized by a scanner, which specifically includes: determining the moving speed of the mobile platform according to a preset monitoring accuracy. 3 . and scanning parameters of the scanner; the scanner is calibrated by using the calibration position; when the mobile platform is moved, the scanner scans to collect the point cloud data of the original section. 3. A tunnel monitoring method according to claim 1, characterized in that, according to the collected original section point cloud data, point cloud splicing operation and reflectivity image generation operation are respectively performed to generate the overall point cloud model of the tunnel And the unfolded map of the tunnel wall, including: splicing the complete tunnel point cloud, and using the reflectivity data to generate the unfolded image of the tunnel wall. 4. A tunnel monitoring method according to claim 1, wherein extracting a single segment image in the expanded view of the tunnel wall according to the straight line segment comprises: removing redundant straight line segments by using prior information, Extract images of individual segments in the expanded view of the tunnel wall. 5 . The tunnel monitoring method according to claim 4 , wherein the prior information includes: the design width of the segment and the average width of the pipe seam. 6 . 6 . The tunnel monitoring method according to claim 1 , wherein the method further comprises: extracting standard block sub-images from the single segment image. 7 . 7. A tunnel monitoring method according to claim 1 or 6, wherein the extracting the capping block sub-image and the two-side connection block sub-image respectively from the single segment image, comprising: adopting a function to match the capping block edge in the single segment image with the capping block design template to determine the position of the capping block in the tunnel unfolded image, and extract the capping block sub-image at the position; According to the position of the capping block and the symmetry of the segment, determine the position of the connecting block on both sides and the position of the standard block, extract the sub-images of the two-side connecting block at the position according to the position of the connecting block on both sides, The standard block sub-image at this position is extracted. 8 . The tunnel monitoring method according to claim 1 , wherein the point cloud small block model comprises: a point cloud model of a single segment and a point cloud model of a single connection block small block. 9 . 9 . The tunnel monitoring method according to claim 8 , wherein the corresponding tunnel point cloud is determined by using the pixel positions of the capping block sub-image, the two-side connection block sub-image and the single segment image. 10 . position, dividing the overall point cloud model of the tunnel into small point cloud models, including: dividing the overall point cloud model of the tunnel into a single connected small block by using the capping block sub-image and the connecting block sub-images on both sides A point cloud model, which divides the overall point cloud model of the tunnel into point cloud models of a single segment according to the pixel positions of the single segment image. 10. A tunnel monitoring device, comprising: a collection module, a data splicing module, a straight line and edge detection module, a segment and connection block determination module, a point cloud segmentation module and a monitoring and analysis module; The acquisition module is used to acquire the original section point cloud data of the tunnel; The data splicing module is configured to generate an overall point cloud model of the tunnel by performing a point cloud splicing operation according to the collected original section point cloud data, and generate an expanded view of the tunnel wall through a reflectivity image generating operation; The straight line and edge detection module is used to perform binarization processing on the expanded image of the tunnel wall to obtain a binary image, use Hough transform to perform parameter mapping on the binary image, and extract the binary image in the A straight line segment within a preset angle range; The segment and connection block determination module is used for extracting a single segment image in the expanded view of the tunnel wall according to the straight segment, and extracting a capping block sub-image and two sides from the single segment image respectively concatenate block subimages; The point cloud segmentation module is used to determine the position of the corresponding tunnel point cloud by using the pixel positions of the capping block sub-image, the two-side connection sub-image and the single segment image, and divides the overall point cloud model of the tunnel into point cloud patch model; The monitoring and analysis module is used for fitting a plane according to the point cloud small block model, calculating the height of the misalignment according to the distance between the fitting planes, and determining the deformation of the segment of the tunnel according to the height of the misalignment.

Images