CN118115561A - Deformation detection method and medium for tunnel structure - Google Patents
Deformation detection method and medium for tunnel structure Download PDFInfo
- Publication number
- CN118115561A CN118115561A CN202311360798.6A CN202311360798A CN118115561A CN 118115561 A CN118115561 A CN 118115561A CN 202311360798 A CN202311360798 A CN 202311360798A CN 118115561 A CN118115561 A CN 118115561A
- Authority
- CN
- China
- Prior art keywords
- point
- data
- data points
- point cloud
- cloud map
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 26
- 238000000034 method Methods 0.000 claims abstract description 47
- 239000011159 matrix material Substances 0.000 claims description 40
- 230000009466 transformation Effects 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 abstract description 31
- 230000008569 process Effects 0.000 abstract description 17
- 238000005259 measurement Methods 0.000 abstract description 8
- 238000013519 translation Methods 0.000 description 5
- 238000013459 approach Methods 0.000 description 3
- 239000003245 coal Substances 0.000 description 3
- 238000010276 construction Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000012216 screening Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 239000002689 soil Substances 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 239000011435 rock Substances 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a deformation detection method of a tunnel structure, which comprises the following steps: and carrying out point cloud matching on the point cloud maps of the front and rear time points of the tunnel, calculating the distance between the corresponding point clouds in the point cloud maps of the front and rear time points, carrying out probability density statistics on the distance to obtain rough deformation point clouds, and carrying out density clustering treatment on the rough deformation point clouds to obtain a plurality of clusters. The number of clusters represents the deformation number, and the distance between the mass center of each cluster and the corresponding first data point is the deformation amount. The method avoids the process of fitting data, avoids fitting errors, clusters the point clouds by using a density clustering algorithm, further calculates the data in one cluster, can simplify calculation steps, reduces calculation complexity and improves measurement accuracy.
Description
Technical Field
The invention relates to the technical field of tunnel deformation detection, in particular to a deformation detection method and medium of a tunnel structure.
Background
The soft soil layer lying under the tunnel can be continuously softened under the action of long-term vibration load, so that the whole settlement and the partial section deformation of the inner wall of the tunnel along the axial direction are caused, and the result of the deformation seriously affects the safe operation of the subway tunnel. Therefore, development of a deformation detection device for a subway tunnel, which adopts an effective detection technology and a data processing method to detect deformation state information of the subway tunnel, becomes urgent and important.
The intelligent safety production level of the coal mine and the construction of a safety monitoring system are provided with higher requirements. The tunnel constructed by coal mining needs to pass through a mountain body, the stress balance state of surrounding soil layers can be influenced by engineering implementation process and mining activities, and rock mass and coal around the tunnel are slowly displaced, deformed and even destroyed, so that the stability of the tunnel structure is seriously influenced, and the safety of constructors is endangered.
The current deformation detection method of tunnel type structures based on the three-dimensional laser scanning technology mostly adopts a method for fitting the tunnel cut-off surface or the central axis of the tunnel, and the method at least has the following defects:
(1) Fitting errors are introduced in the roadway tunnel deformation detection analysis in the fitting process of the method, systematic errors in the data acquisition process of the sensor are unavoidable, and interference of the two errors tends to reduce the deformation detection accuracy.
(2) The processes of data acquisition, tunnel cross section fitting curve and tunnel central axis fitting curve are generally carried out step by step, and the step by step process can complicate the deformation detection process, so that the detection efficiency is low.
Disclosure of Invention
In view of the above-described drawbacks or shortcomings of the prior art, it is desirable to provide a deformation detection method of a tunnel structure.
In one aspect, the present invention provides a method for detecting deformation of a tunnel structure, including:
acquiring initial three-dimensional point cloud map data of a tunnel to obtain a first point cloud map; the tunnel in the first point cloud map is free from deformation; the first point cloud map comprises a plurality of first data points;
Acquiring current three-dimensional point cloud map data of a tunnel to obtain a second point cloud map; the first point cloud map and the second point cloud map correspond to the same section of tunnel; the second point cloud map comprises a plurality of second data points;
Performing point cloud matching on the first point cloud map and the second point cloud map by using a nearest point iterative algorithm, and converting the first point cloud map and the second point cloud map into the same coordinate system to obtain a third point cloud map; the third point cloud map comprises a plurality of first data points and a plurality of second data points which are matched by point clouds and correspond to each other;
Calculating the distance between each group of first data points and second data points which correspond to each other in the third point cloud map to obtain a first distance set;
performing probability density statistics on elements in the first interval set, and taking second data points corresponding to the first interval in a first set interval as rough deformation point clouds;
performing density clustering processing on the rough deformation point cloud by using a density clustering algorithm to obtain an accurate deformation point cloud; the accurate deformation point cloud comprises a plurality of second cluster clusters;
counting the number of the second cluster, and taking the number of the second cluster as the deformation number of the tunnel;
And calculating the mass center of the second cluster, and taking the distance between the mass center of the second cluster and the corresponding first data point as the deformation size.
According to the technical scheme provided by the invention, the step of performing point cloud matching on the plurality of first local maps and the plurality of second local maps by using a nearest point iterative algorithm and converting the plurality of first local maps and the plurality of second local maps into the same coordinate system comprises the following steps:
Respectively calculating Euclidean distance between each second data point and each first data point;
Respectively finding a second data point of which each first data point is closest to the first data point, and respectively taking the first data point and the second data point of which the Euclidean distance is closest to the second data point as two points which are mutually corresponding;
Calculating a rotation matrix between the first point cloud map and the second point cloud map according to the first data point and the second data point which are corresponding to each other, and obtaining a second rotation matrix;
calculating coordinates of a plurality of second data points in the first point cloud map according to the second rotation matrix to obtain a plurality of transformation coordinates;
And adding a second data point into the first point cloud map according to the transformation coordinates, and calculating to obtain the third point cloud map.
According to the technical scheme provided by the invention, after performing the step of performing point cloud matching on the first point cloud map and the second point cloud map by using the nearest point iterative algorithm and converting the first point cloud map and the second point cloud map into the same coordinate system, and before performing the step of calculating the distance between each group of first data points and second data points corresponding to each other in the third point cloud map, the method further comprises the following steps:
dividing the third point cloud map to obtain a plurality of local point cloud maps; the local point cloud map comprises a plurality of first data points and a plurality of second data points which correspond to each other;
According to the first data points and the second data points, rotation matrixes between the first data points and the second data points which correspond to each other are calculated, and the first rotation matrixes are obtained;
Calculating determinant of each first rotation matrix; a first rotation matrix of determinant not equal to 1 indicates the presence of a deformation between the first data point and the second data point;
Taking a local point cloud map where a first data point and a second data point corresponding to a rotation matrix with determinant not equal to 1 are located as deformation areas, and obtaining a plurality of first deformation areas;
And taking the plurality of first deformation areas as the third point cloud map.
According to the technical scheme provided by the invention, the step of respectively calculating the intervals between the first data point and the second data point corresponding to the same position of the tunnel to obtain the first interval set comprises the following steps:
dividing the third point cloud map into a dense area and a sparse area;
Searching a plurality of first data points in a searching radius around each second data point in a dense area and a sparse area in an octree mode by taking a third set value as a searching radius;
calculating Euclidean distance between each second data point and the corresponding first data point in the dense area to obtain the distance between the first data point and the second data point in the dense area;
Searching a plurality of first data points in the radius according to each second data point in the sparse region, and calculating a plane where the plurality of first data points are located to obtain a reference plane of each second data point;
Calculating the distance from the second data point to the corresponding reference plane to obtain the distance between the first data point and the second data point in the sparse region;
the distance between the first data point and the second data point in the dense area and the distance between the first data point and the second data point in the sparse area are taken as elements in the first distance set together.
According to the technical scheme provided by the invention, the step of dividing the third point cloud map into a dense area and a sparse area comprises the following steps:
Searching the number of data points around the second data point by taking the third set value as a searching radius;
taking the number of data points around the second data point as the concentration degree of the second data point;
And dividing the second data points with the concentration greater than or equal to a second set value into dense areas, and dividing the second data points with the concentration less than the second set value into sparse areas.
According to the technical scheme provided by the invention, the step of searching a plurality of first data points in the searching radius of each second data point in the sparse region in an octree mode comprises the following steps of:
S61, taking a cube space which takes the second data point as a center and the Nth reference value as a side length as an Nth search space; the initial value of N is set to 1;
s62, searching the number of first data points in the N search space to obtain the N data point number;
s63, judging whether the number of the N data points is smaller than 4;
S64, when the number of the Nth data points is greater than or equal to 4, taking half of the Nth reference value as an N+1 reference value, adding one to the N value, and repeatedly executing the steps from S61 to S63;
s65, when the number of the N data points is less than 4, performing the steps from S66 to S67;
and S66, taking all the first data points in the N-1 search space as the second data points to search a plurality of first data points in the radius, and ending the cycle.
According to the technical scheme provided by the invention, according to a plurality of first data points in the searching radius of each second data point in the sparse region, the step of calculating the plane where the plurality of first data points are located and obtaining the reference plane of each second data point comprises the following steps:
acquiring a plurality of first data points in the second data point searching radius to obtain a plurality of third data points;
calculating a center point of a plurality of the third data points;
Calculating a covariance matrix of the second data point according to the third data points and the central point;
calculating a feature vector of the covariance matrix;
taking a plurality of components of the feature vector as a plurality of first plane parameter values of the reference plane;
calculating an inner product of the coordinates of the center point and taking a negative value to obtain a second plane parameter value of the reference plane;
The reference plane is constructed from a plurality of the first plane parameter values and the second plane parameter values.
According to the technical scheme provided by the invention, the density clustering algorithm is used for carrying out density clustering processing on the rough deformation point cloud, and the step of obtaining the accurate deformation point cloud comprises the following steps:
Obtaining a neighborhood radius;
Calculating the number of data points of each second data point in the rough deformation point cloud in the neighborhood radius;
taking the second data point with the number of data points in the neighborhood radius being larger than a fourth set value as a core data point;
dividing core data points with the distance smaller than the neighborhood radius into the same cluster to obtain a plurality of first clusters;
The second data points in the neighborhood radius coverage range of all the core data points in the first cluster and the first cluster are used as second clusters together to obtain a plurality of second clusters;
And taking a plurality of second cluster clusters as accurate deformation point clouds.
According to the technical scheme provided by the invention, the step of calculating the mass center of the second cluster comprises the following steps:
Acquiring all core data points contained in the second cluster;
Calculating the coordinate average value of all the core data points;
And taking the coordinate average value of all core data points as the mass center of the second clustering cluster.
In another aspect, the present embodiment provides a storage medium having stored thereon a program for executing a deformation detection method of a tunnel structure; when the program of the deformation detection method of the tunnel structure is executed, the program is used for:
a method of detecting deformation of a tunnel structure as claimed in any one of the preceding claims.
The invention has the beneficial effects that:
A deformation detection method of a tunnel structure is designed, which comprises the following steps: and carrying out point cloud matching on the point cloud maps of the front and rear time points of the tunnel, calculating the distance between the corresponding point clouds in the point cloud maps of the front and rear time points, carrying out probability density statistics on the distance to obtain rough deformation point clouds, and carrying out density clustering treatment on the rough deformation point clouds to obtain a plurality of clusters. The number of clusters represents the deformation number, and the distance between the mass center of each cluster and the corresponding first data point is the deformation amount. The method avoids the process of data fitting, thereby avoiding fitting errors; the point cloud is clustered by using a density clustering algorithm, so that data in one cluster is calculated, the calculation step can be simplified, the calculation complexity is reduced, and the measurement accuracy is improved.
Drawings
Other features, objects and advantages of the present invention will become more apparent upon reading of the detailed description of non-limiting embodiments, made with reference to the accompanying drawings in which:
FIG. 1 is a schematic flow chart of a method for detecting deformation of a tunnel structure according to the present invention;
FIG. 2 is a flow chart for searching data points using the octree approach.
Detailed Description
The invention is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be noted that, for convenience of description, only the portions related to the invention are shown in the drawings.
It should be noted that, without conflict, the embodiments of the present invention and features of the embodiments may be combined with each other. The invention will be described in detail below with reference to the drawings in connection with embodiments.
Example 1
The implementation process of the invention can be summarized as follows: acquiring a point cloud map without deformation at the initial stage of a tunnel and a current point cloud map; performing point cloud matching on the two maps, and converting the two maps into the same coordinate system; calculating to obtain the distance between the data points in the front and rear two-point cloud maps; screening the data points, performing density clustering, and dividing the data points into a plurality of clustering clusters; and calculating the deformation quantity and the specific deformation quantity according to the cluster. In which the total amount of computation is reduced and the effect of errors is reduced in a number of ways.
Referring to fig. 1, the method for detecting deformation of a tunnel structure provided by the present invention includes:
S1, acquiring initial three-dimensional point cloud map data of a tunnel to obtain a first point cloud map; the tunnel in the first point cloud map is free from deformation; the first point cloud map comprises a plurality of first data points;
S2, acquiring current three-dimensional point cloud map data of a tunnel to obtain a second point cloud map; the first point cloud map and the second point cloud map correspond to the same section of tunnel; the second point cloud map comprises a plurality of second data points;
s3, performing point cloud matching on the first point cloud map and the second point cloud map by using a nearest point iterative algorithm, and converting the first point cloud map and the second point cloud map into the same coordinate system to obtain a third point cloud map; the third point cloud map comprises a plurality of first data points and a plurality of second data points which are matched by point clouds and correspond to each other;
s4, calculating the distance between each group of first data points and second data points which correspond to each other in the third point cloud map to obtain a first distance set;
S5, carrying out probability density statistics on elements in the first interval set, and taking second data points corresponding to the first interval in a first set interval as rough deformation point clouds;
S6, performing density clustering processing on the rough deformation point cloud by using a density clustering algorithm to obtain an accurate deformation point cloud; the accurate deformation point cloud comprises a plurality of second cluster clusters;
s7, counting the number of the second cluster, and taking the number of the second cluster as the deformation number of the tunnel;
And S8, calculating the mass center of the second cluster, and taking the distance between the mass center of the second cluster and the corresponding first data point as the deformation size.
Specifically, the first set interval is set according to a probability density statistical result, wherein the probability density statistical result has a mean value and a standard deviation. The first setting interval is set to be larger than the average value minus 2 times of standard deviation and smaller than the average value plus 4 times of standard deviation.
In this embodiment, the specific range of the first setting section is set to 2cm to 8cm.
Namely, the distances between the deformation point clouds are divided by taking the mean value as the center and taking 2cm as the dividing interval, and the points (the points with deformation of 2cm to 8 cm) with the first interval in the 2 nd interval, the 3 rd interval and the 4 th interval are output as rough deformation point clouds.
And carrying out probability density statistics on the distance between the first point cloud and the second point cloud, screening, and removing the conditions of too small distance and too large distance, so as to eliminate the original deformation-free position caused by measurement errors, display the trace deformation of the measurement results, and obtain the deformation beyond the actual possible range caused by measurement of various error factors. The method can reduce the error influence, reduce the total data quantity required to be calculated and improve the calculation efficiency.
In some embodiments, a robot carrying a laser radar is used for data acquisition, a FAST-LIO algorithm (FAST-LIO: A FAST, robot LiDAR-inertial Odometry Package by Tightly-Coupled ITERATED KALMAN FILTER, a laser radar-inertial coupling map building algorithm based on tightly Coupled iterative Kalman filtering) is adopted to build a roadway three-dimensional point cloud map, and finally a first point cloud map and a second point cloud map are obtained.
The first point cloud map acquisition process comprises the following steps: and (4) at the initial stage of tunnel construction completion, acquiring a three-dimensional point cloud map once, and storing the three-dimensional point cloud map. And after a period of time, acquiring the three-dimensional point cloud map under the current condition of the tunnel again, and carrying out the deformation detection method on the two-point cloud map. The deformation quantity and the deformation quantity of each position of the tunnel in the current state can be obtained.
OusterOs0-64 laser radar adopts 865nm wavelength infrared as a light source, carries a built-in inertial measurement unit (Inertial Measurement Unit, IMU for short) supporting SLAM algorithm (Simultaneous Localization AND MAPPING, instant positioning and map construction) and outputs uniformly-spaced three-dimensional point cloud images and pixel-aligned two-dimensional photographic-grade images. ousterOs0 the vertical angle of view of the 0-64 type laser radar is 90 degrees, the horizontal angle of view is 360 degrees, the vertical resolution is 64 lines, and 130 ten thousand points can be output per second.
The FAST-LIO algorithm can calculate the kalman gain by using a new rule, and the calculated amount of the process is only related to the magnitude of the state quantity, but not directly proportional to the conventional observed quantity data quantity, so that the problem of the calculation amount surge caused by a large number of characteristic points is solved.
The method avoids the process of fitting data, avoids fitting errors, clusters the point clouds by using a density clustering algorithm, further calculates the data in one cluster, can simplify calculation steps, reduces calculation complexity and improves measurement accuracy.
The method provided by the invention is used for detecting the deformation of the tunnel at regular intervals, and can be used for continuously monitoring the deformation of the tunnel. When the tunnel generates deformation exceeding the dangerous value, the specific deformation position and deformation quantity can be timely checked.
Further, the step of performing point cloud matching on the plurality of first local maps and the plurality of second local maps using a closest point iterative algorithm, and converting the plurality of first local maps and the plurality of second local maps to the same coordinate system includes:
Respectively calculating Euclidean distance between each second data point and each first data point;
Respectively finding a second data point of which each first data point is closest to the first data point, and respectively taking the first data point and the second data point of which the Euclidean distance is closest to the second data point as two points which are mutually corresponding;
Calculating a rotation matrix between the first point cloud map and the second point cloud map according to the first data point and the second data point which are corresponding to each other, and obtaining a second rotation matrix;
calculating coordinates of a plurality of second data points in the first point cloud map according to the second rotation matrix to obtain a plurality of transformation coordinates;
And adding a second data point into the first point cloud map according to the transformation coordinates, and calculating to obtain the third point cloud map.
Specifically, the closest point iterative algorithm is simply referred to as the ICP algorithm, (ITERATIVE CLOSEST POINT).
Transforming the second data point into the first point cloud map according to the first data point and the second data point which correspond to each other comprises:
the initial rotation matrix is acquired first, then the Euler angle and the translation amount between the first data point and the second data point are calculated, and finally the Euler angle and the translation amount are substituted into the initial rotation matrix.
The initial rotation matrix is specifically shown in formula (one);
Wherein, alpha, beta and gamma respectively represent Euler angles among three coordinate axes. In the embodiment, euler angles and proper amounts of translation between three coordinate axes of the two-point cloud map are measured.
Bringing Euler angles into an initial rotation matrix to obtain a second rotation matrix; and finally substituting a proper amount of translation and a second rotation matrix into a coordinate transformation formula, and calculating the coordinates of the second data point in the first point cloud map to obtain a plurality of transformation coordinates.
The coordinate transformation formula is shown as a formula (II);
wherein, Representing transformed coordinates, in this embodiment representing coordinates of the second data point in the first point cloud map; /(I)Representing the coordinates before transformation, substituting the coordinates of the second data point in the second point cloud map in the embodiment; alpha, beta and gamma respectively represent Euler angles to be substituted; /(I)Representing a translation vector between two point cloud maps.
Specifically, the third point cloud map includes a plurality of first data points and a plurality of second data points, and after the point cloud matching transformation is performed to the same coordinate system, the distances between the data points corresponding to each other are conveniently calculated.
Further, after performing the step of performing point cloud matching on the first point cloud map and the second point cloud map by using the nearest point iterative algorithm and converting the first point cloud map and the second point cloud map to the same coordinate system, and before performing the step of calculating the distance between each group of first data points and second data points corresponding to each other in the third point cloud map, the method further includes:
dividing the third point cloud map to obtain a plurality of local point cloud maps; the local point cloud map comprises a plurality of first data points and a plurality of second data points which correspond to each other;
According to the first data points and the second data points, rotation matrixes between the first data points and the second data points which correspond to each other are calculated, and the first rotation matrixes are obtained;
Calculating determinant of each first rotation matrix; a first rotation matrix of determinant not equal to 1 indicates the presence of a deformation between the first data point and the second data point;
Taking a local point cloud map where a first data point and a second data point corresponding to a rotation matrix with determinant not equal to 1 are located as deformation areas, and obtaining a plurality of first deformation areas;
And taking the plurality of first deformation areas as the third point cloud map.
In some embodiments, the third point cloud map is equally partitioned, and a specific partition interval is set according to a tunnel length.
For example, the tunnel length detected is short, and the pitch is set to be 0.5 to 1 meter or any value in between, only 50 to 100 meters. When the detected tunnel length is 1-2 km, the pitch is set to be 5-10 m or any value in between.
In this embodiment, the first rotation matrix is the same as the second rotation matrix, and specific values of the second rotation matrix can be directly obtained and substituted into calculation in this step.
When the determinant of the first rotation matrix between the first data point and the second data point is equal to 1, it indicates that there is no deformation between the two mutually corresponding first data point and second data point; when the determinant of the first rotation matrix is not equal to 1, it indicates that there is a deformation between the two.
And judging whether deformation exists in the segmented region according to whether the determinant of the rotation matrix is equal to 1. And further screening out the area with deformation, and taking the area with deformation as a third point cloud map again.
The method can filter the deformation-free area, and in the subsequent calculation process, the deformation-free area is not calculated any more, so that the total data amount of calculation can be greatly reduced, and the calculation efficiency is improved.
For example, a tunnel 1 km long is detected, and the division interval is set to 10 meters. For ease of understanding, it is assumed that the entire tunnel is divided into 100 partial point cloud maps only along the tunnel extending direction. If twenty deformation positions exist in the tunnel actually, ten regions with deformation positions distributed in the tunnel are deformed, 90 regions without deformation can be filtered through the mode, and only 10 regions needing further calculation remain. Since the point cloud is not completely evenly distributed, it can be estimated approximately that: the total amount of data was reduced by about 90%. This approach thus achieves increased computational efficiency by greatly reducing the total amount of data.
According to the set segmentation mode, the input point cloud map is divided into cubes, and the mass center in each cube approximates to a plurality of points in the cube. The method can carry out downsampling processing on the premise of not damaging the whole structure information of the point cloud, retain deformation information and improve operation efficiency.
In this embodiment, the tunnel deformation information is obtained by calculating and analyzing the positional relationship between the corresponding points of the point cloud. The deformation detection result in the mode has good precision, but the calculation process has high requirements on the operation speed and accuracy. ousterOs0-64 laser radars have scanning points of up to 130 ten thousand per second, and the acquired data density is rich. However, the excessive amount of the point cloud data can affect the calculation efficiency of deformation detection, and the repeated calculation can be performed due to the existence of the point cloud with excessive density in a small space in the deformation detection process, so that the calculation force is wasted. Therefore, the data needs to be downsampled without affecting the accuracy of the lidar data.
Further, the step of calculating the distances between the first data point and the second data point corresponding to the same position of the tunnel respectively to obtain the first distance set includes:
dividing the third point cloud map into a dense area and a sparse area;
Searching a plurality of first data points in a searching radius around each second data point in a dense area and a sparse area in an octree mode by taking a third set value as a searching radius;
calculating Euclidean distance between each second data point and the corresponding first data point in the dense area to obtain the distance between the first data point and the second data point in the dense area;
Searching a plurality of first data points in the radius according to each second data point in the sparse region, and calculating a plane where the plurality of first data points are located to obtain a reference plane of each second data point;
Calculating the distance from the second data point to the corresponding reference plane to obtain the distance between the first data point and the second data point in the sparse region;
the distance between the first data point and the second data point in the dense area and the distance between the first data point and the second data point in the sparse area are taken as elements in the first distance set together.
In some embodiments, the third set point is set to 1 centimeter.
Specifically, the dense areas have more data points and are relatively close to each other, so that the error is smaller. The sparse region has fewer data points, the distance between the two data points is larger, the direct calculation of the Euclidean distance is easily influenced by various factors, and errors are generated in the calculation of the distance.
For the reasons described above, the present embodiment directly calculates the Euclidean distance between the dense region data points. And (3) calculating the reference plane for the points between the sparse areas, calculating the distance between the data points and the reference plane, and taking the distance between the second data point and the reference plane as the distance between the second data point and the first data point.
The reference plane is constructed by utilizing a plurality of data points, so that the influence caused by errors can be reduced, and the accuracy of a calculation result is ensured. Therefore, the sparse region and the dense region are calculated in different modes, so that the influence of errors on the calculation of the data point distance of the sparse region can be reduced under the condition that the calculated amount of the data point distance in the dense region is not increased.
Further, the step of dividing the third point cloud map into a dense area and a sparse area includes:
Searching the number of data points around the second data point by taking the third set value as a searching radius;
taking the number of data points around the second data point as the concentration degree of the second data point;
And dividing the second data points with the concentration greater than or equal to a second set value into dense areas, and dividing the second data points with the concentration less than the second set value into sparse areas.
In some embodiments, the third set point is set to 1 centimeter, the second set point is set to 24, and the concentration is 24 or more indicating dense and less than 24 indicating sparse.
In some implementations, the search process includes:
Acquiring coordinates of a second data point calculated currently;
Traversing all data points, and calculating the distance from other second data points to the second data points;
Listing a second data point with a distance smaller than the search radius into a set of adjacent points;
and after the calculation process is finished, counting the number of data points in the adjacent point set, and obtaining the number of data points around the second data point.
Further, referring to fig. 2, the step of searching a plurality of first data points within each of the second data point searching radii within the sparse region using an octree manner includes:
S61, taking a cube space which takes the second data point as a center and the Nth reference value as a side length as an Nth search space; the initial value of N is set to 1;
s62, searching the number of first data points in the N search space to obtain the N data point number;
s63, judging whether the number of the N data points is smaller than 4;
S64, when the number of the Nth data points is greater than or equal to 4, taking half of the Nth reference value as an N+1 reference value, adding one to the N value, and repeatedly executing the steps from S61 to S63;
s65, when the number of the N data points is less than 4, performing the steps from S66 to S67;
and S66, taking all the first data points in the N-1 search space as the second data points to search a plurality of first data points in the radius, and ending the cycle.
In some embodiments, the first reference value is set to 1 centimeter.
In combination with the above-mentioned division manner of the dense region, the number of data points in the space of 1 cubic centimeter in the dense region is greater than 24 (including the second data point itself), and after one cycle, the search space is reduced to 1/8 cubic centimeter.
If the data points are evenly distributed in space, the number of data points in the space around the second data point should be around 3 (4 points together with the second data point). While the actual data points are unevenly distributed, this may result in some cases in a number of data points greater than 3, and in other cases in less than 3, of the space around the second data point.
The above-described loop may be performed such that the number of other data points around the second data point searched is greater than 3 and less. The number of data points required for constructing the reference plane can be searched enough, and the total calculated data amount can be reduced.
Specifically, when the nth data point number is equal to 4, it means that the second data point itself and the other three second data points are included in the nth search space. The resolution of the reference plane can be calculated from the coordinates of the other three data points.
In order to avoid search errors and reduce calculation errors caused by deviation of individual points, when the number of data points is smaller than 4, all data points in the search space obtained in the previous step are used as a plurality of nearest data points, and more data points are combined for calculation so as to obtain a reference plane.
The method can further reduce the calculation complexity and ensure the calculation accuracy.
Further, the step of calculating a plane in which the plurality of first data points are located according to the plurality of first data points in the searching radius of each second data point in the sparse region, and obtaining a reference plane of each second data point includes:
acquiring a plurality of first data points in the second data point searching radius to obtain a plurality of third data points;
calculating a center point of a plurality of the third data points;
Calculating a covariance matrix of the second data point according to the third data points and the central point;
calculating a feature vector of the covariance matrix;
taking a plurality of components of the feature vector as a plurality of first plane parameter values of the reference plane;
calculating an inner product of the coordinates of the center point and taking a negative value to obtain a second plane parameter value of the reference plane;
The reference plane is constructed from a plurality of the first plane parameter values and the second plane parameter values.
In this embodiment, the centroid and centroid remain identical, i.e., the center point of the plurality of third data points and the centroid of the plurality of third data points are the same point, since no quality factor is considered.
The covariance matrix is shown in a formula (III);
where M 3×3 represents the third order covariance matrix, n represents the total number of third data points, p i represents the ith third data point, Representing the center point, (. Cndot.) T represents the transpose of the matrix.
The center point is calculated by the formula (IV):
calculating the minimum eigenvalue and eigenvector of the covariance matrix according to a formula (five);
wherein, Representing the eigenvector and V representing the eigenvalue. And calculating to obtain a plurality of characteristic values, wherein the minimum characteristic value is taken in the embodiment.
Calculating the corresponding feature vector when the minimum feature value is obtainedThe components of (a) represent a plurality of first plane parameter values a, b, c, respectively; the second plane parameter value d is equal to the coordinates of the center point and takes the negative value.
The reference plane thus constructed is represented by formula (six);
ax+by+cz+d=0 formula (six).
Further, the step of performing density clustering processing on the rough deformed point cloud by using a density clustering algorithm to obtain an accurate deformed point cloud comprises the following steps:
Obtaining a neighborhood radius;
Calculating the number of data points of each second data point in the rough deformation point cloud in the neighborhood radius;
taking the second data point with the number of data points in the neighborhood radius being larger than a fourth set value as a core data point;
dividing core data points with the distance smaller than the neighborhood radius into the same cluster to obtain a plurality of first clusters;
The second data points in the neighborhood radius coverage range of all the core data points in the first cluster and the first cluster are used as second clusters together to obtain a plurality of second clusters;
And taking a plurality of second cluster clusters as accurate deformation point clouds.
In this embodiment, the neighborhood radius is set to 5cm, and the fourth set value is set to 10.
The neighborhood radius needs to be set according to the acquisition precision of the laser radar sensor, and the fourth set value is set according to the data density acquired by the laser radar sensor.
Specifically, if there are three core data points, they are respectively named: a first core data point, a second core data point, and a third core data point. Assuming that the distance between the first core data point and the second core data point is less than the fourth set point, the distance between the second core data point and the third core data point is less than the fourth set point, but the distance between the first core data point and the third core data point is greater than the fourth set point. In this case, the first core data point, the second core data point, and the third core data point all belong to the same cluster.
This approach can remove relatively scattered data points, further reducing the overall amount of data. Data points are simultaneously divided into a plurality of clusters. During calculation, only the data points in each cluster are calculated, the relation of the data points among the clusters is not needed to be considered, the calculation complexity can be reduced, and the calculation efficiency is further improved.
Further, the step of calculating the centroid of the second cluster includes:
Acquiring all core data points contained in the second cluster;
Calculating the coordinate average value of all the core data points;
And taking the coordinate average value of all core data points as the mass center of the second clustering cluster.
In some embodiments, only the average coordinate value of the core data points in the cluster is calculated, so that the influence of the non-core data point coordinate deviation in the cluster on the average coordinate value can be eliminated, the total calculation amount can be reduced, and the calculation efficiency is improved.
Example 2
The present embodiment provides a storage medium having stored thereon a program that executes a deformation detection method of a tunnel structure; when the program of the deformation detection method of the tunnel structure is executed, the program is used for:
a deformation detection method of a tunnel structure according to any one of the above embodiments is performed.
Specifically, the present invention also provides a storage medium that may be included in the electronic device described in the above embodiment; or may exist alone without being incorporated into the electronic device. The storage medium carries one or more programs which, when executed by an electronic device, cause the electronic device to implement the method according to the above embodiment, and in particular, the method for detecting deformation of a tunnel structure according to any one of the above embodiments.
The above description is only illustrative of the preferred embodiments of the present invention and of the principles of the technology employed. It will be appreciated by persons skilled in the art that the scope of the invention referred to in the present invention is not limited to the specific combinations of the technical features described above, but also covers other technical features formed by any combination of the technical features described above or their equivalents without departing from the inventive concept. Such as the above-mentioned features and the technical features disclosed in the present invention (but not limited to) having similar functions are replaced with each other.
Claims (10)
1. A deformation detection method for a tunnel structure, comprising:
acquiring initial three-dimensional point cloud map data of a tunnel to obtain a first point cloud map; the tunnel in the first point cloud map is free from deformation; the first point cloud map comprises a plurality of first data points;
Acquiring current three-dimensional point cloud map data of a tunnel to obtain a second point cloud map; the first point cloud map and the second point cloud map correspond to the same section of tunnel; the second point cloud map comprises a plurality of second data points;
Performing point cloud matching on the first point cloud map and the second point cloud map by using a nearest point iterative algorithm, and converting the first point cloud map and the second point cloud map into the same coordinate system to obtain a third point cloud map; the third point cloud map comprises a plurality of first data points and a plurality of second data points which are matched by point clouds and correspond to each other;
Calculating the distance between each group of first data points and second data points which correspond to each other in the third point cloud map to obtain a first distance set;
performing probability density statistics on elements in the first interval set, and taking second data points corresponding to the first interval in a first set interval as rough deformation point clouds;
performing density clustering processing on the rough deformation point cloud by using a density clustering algorithm to obtain an accurate deformation point cloud; the accurate deformation point cloud comprises a plurality of second cluster clusters;
counting the number of the second cluster, and taking the number of the second cluster as the deformation number of the tunnel;
And calculating the mass center of the second cluster, and taking the distance between the mass center of the second cluster and the corresponding first data point as the deformation size.
2. The method of claim 1, wherein the step of performing point cloud matching on the plurality of first partial maps and the plurality of second partial maps using a nearest point iterative algorithm, and converting the plurality of first partial maps and the plurality of second partial maps to the same coordinate system comprises:
Respectively calculating Euclidean distance between each second data point and each first data point;
Respectively finding a second data point of which each first data point is closest to the first data point, and respectively taking the first data point and the second data point of which the Euclidean distance is closest to the second data point as two points which are mutually corresponding;
Calculating a rotation matrix between the first point cloud map and the second point cloud map according to the first data point and the second data point which are corresponding to each other, and obtaining a second rotation matrix;
calculating coordinates of a plurality of second data points in the first point cloud map according to the second rotation matrix to obtain a plurality of transformation coordinates;
And adding a second data point into the first point cloud map according to the transformation coordinates, and calculating to obtain the third point cloud map.
3. The method for detecting deformation of a tunnel structure according to claim 1, wherein after performing the step of performing point cloud matching on the first point cloud map and the second point cloud map using the nearest point iterative algorithm and converting the first point cloud map and the second point cloud map to the same coordinate system, and before performing the step of calculating a distance between each group of first data points and second data points corresponding to each other in the third point cloud map, the method further comprises:
dividing the third point cloud map to obtain a plurality of local point cloud maps; the local point cloud map comprises a plurality of first data points and a plurality of second data points which correspond to each other;
According to the first data points and the second data points, rotation matrixes between the first data points and the second data points which correspond to each other are calculated, and the first rotation matrixes are obtained;
Calculating determinant of each first rotation matrix; a first rotation matrix of determinant not equal to 1 indicates the presence of a deformation between the first data point and the second data point;
Taking a local point cloud map where a first data point and a second data point corresponding to a rotation matrix with determinant not equal to 1 are located as deformation areas, and obtaining a plurality of first deformation areas;
And taking the plurality of first deformation areas as the third point cloud map.
4. The method for detecting deformation of a tunnel structure according to claim 1, wherein the step of calculating the distances between the first data point and the second data point corresponding to the same position of the tunnel, respectively, to obtain the first distance set comprises:
dividing the third point cloud map into a dense area and a sparse area;
Searching a plurality of first data points in a searching radius around each second data point in a dense area and a sparse area in an octree mode by taking a third set value as a searching radius;
Calculating Euclidean distance between each second data point in the intensive area and a plurality of first data points in the surrounding searching radius, and taking the Euclidean distance minimum value as the distance between the first data points and the second data points;
Searching a plurality of first data points in the radius according to each second data point in the sparse region, and calculating a plane where the plurality of first data points are located to obtain a reference plane of each second data point;
Calculating the distance from the second data point to the corresponding reference plane to obtain the distance between the first data point and the second data point in the sparse region;
the distance between the first data point and the second data point in the dense area and the distance between the first data point and the second data point in the sparse area are taken as elements in the first distance set together.
5. The method of claim 4, wherein the step of dividing the third point cloud map into a dense area and a sparse area comprises:
Searching the number of data points around the second data point by taking the third set value as a searching radius;
taking the number of data points around the second data point as the concentration degree of the second data point;
And dividing the second data points with the concentration greater than or equal to a second set value into dense areas, and dividing the second data points with the concentration less than the second set value into sparse areas.
6. The method of claim 4, wherein searching for a plurality of first data points within each of the second data point search radii within the sparse zone using an octree method comprises:
S61, taking a cube space which takes the second data point as a center and the Nth reference value as a side length as an Nth search space; the initial value of N is set to 1;
s62, searching the number of first data points in the N search space to obtain the N data point number;
s63, judging whether the number of the N data points is smaller than 4;
S64, when the number of the Nth data points is greater than or equal to 4, taking half of the Nth reference value as an N+1 reference value, adding one to the N value, and repeatedly executing the steps from S61 to S63;
s65, when the number of the N data points is less than 4, performing the steps from S66 to S67;
and S66, taking all the first data points in the N-1 search space as the second data points to search a plurality of first data points in the radius, and ending the cycle.
7. The method for detecting deformation of a tunnel structure according to claim 4, wherein the step of calculating a plane in which the plurality of first data points are located according to the plurality of first data points in the search radius of each of the second data points in the sparse region, and obtaining a reference plane of each of the second data points comprises:
acquiring a plurality of first data points in the second data point searching radius to obtain a plurality of third data points;
calculating a center point of a plurality of the third data points;
Calculating a covariance matrix of the second data point according to the third data points and the central point;
calculating a feature vector of the covariance matrix;
taking a plurality of components of the feature vector as a plurality of first plane parameter values of the reference plane;
calculating an inner product of the coordinates of the center point and taking a negative value to obtain a second plane parameter value of the reference plane;
The reference plane is constructed from a plurality of the first plane parameter values and the second plane parameter values.
8. The method for detecting deformation of a tunnel structure according to claim 1, wherein the step of performing density clustering on the rough deformed point cloud by using a density clustering algorithm to obtain an accurate deformed point cloud comprises:
Obtaining a neighborhood radius;
Calculating the number of data points of each second data point in the rough deformation point cloud in the neighborhood radius;
taking the second data point with the number of data points in the neighborhood radius being larger than a fourth set value as a core data point;
dividing core data points with the distance smaller than the neighborhood radius into the same cluster to obtain a plurality of first clusters;
The second data points in the neighborhood radius coverage range of all the core data points in the first cluster and the first cluster are used as second clusters together to obtain a plurality of second clusters;
And taking a plurality of second cluster clusters as accurate deformation point clouds.
9. The method of claim 8, wherein the step of calculating the centroid of the second cluster comprises:
Acquiring all core data points contained in the second cluster;
Calculating the coordinate average value of all the core data points;
And taking the coordinate average value of all core data points as the mass center of the second clustering cluster.
10. A storage medium, wherein a program for executing a deformation detection method of a tunnel structure is stored on the storage medium; when the program of the deformation detection method of the tunnel structure is executed, the program is used for:
a method of detecting deformation of a tunnel structure according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311360798.6A CN118115561A (en) | 2023-10-20 | 2023-10-20 | Deformation detection method and medium for tunnel structure |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311360798.6A CN118115561A (en) | 2023-10-20 | 2023-10-20 | Deformation detection method and medium for tunnel structure |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118115561A true CN118115561A (en) | 2024-05-31 |
Family
ID=91214939
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311360798.6A Pending CN118115561A (en) | 2023-10-20 | 2023-10-20 | Deformation detection method and medium for tunnel structure |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118115561A (en) |
-
2023
- 2023-10-20 CN CN202311360798.6A patent/CN118115561A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110969624B (en) | Laser radar three-dimensional point cloud segmentation method | |
| CN114526745B (en) | Drawing construction method and system for tightly coupled laser radar and inertial odometer | |
| CN113432600A (en) | Robot instant positioning and map construction method and system based on multiple information sources | |
| CN103278115B (en) | A kind of method and system calculating silt arrester siltation volume based on DEM | |
| CN112215958B (en) | Laser radar point cloud data projection method based on distributed computation | |
| CN112711034B (en) | Object detection method, device and equipment | |
| CN114488194A (en) | Method for detecting and identifying targets under structured road of intelligent driving vehicle | |
| Xie et al. | A comparison and review of surface detection methods using MBL, MABEL, and ICESat-2 photon-counting laser altimetry data | |
| CN112034431A (en) | Radar and RTK external reference calibration method and device | |
| CN113255677B (en) | Method, equipment and medium for rapidly extracting rock mass structural plane and occurrence information | |
| CN111797836A (en) | Extraterrestrial celestial body patrolling device obstacle segmentation method based on deep learning | |
| CN114387506A (en) | Transmission tower monitoring method and device, computer equipment and storage medium | |
| CN103256914A (en) | Method and system for calculating inundated areas of warp land dam based on DEM | |
| CN114998395A (en) | Effective embankment three-dimensional data change detection method and system | |
| CN111208495A (en) | Ground extraction method based on laser radar point cloud characteristic line and plane calibration | |
| CN111738214B (en) | Unmanned aerial vehicle target detection method in laser point cloud | |
| WO2022099620A1 (en) | Three-dimensional point cloud segmentation method and apparatus, and mobile platform | |
| CN114022760A (en) | Railway tunnel barrier monitoring and early warning method, system, equipment and storage medium | |
| Lin et al. | Noise point detection from airborne lidar point cloud based on spatial hierarchical directional relationship | |
| CN112015199A (en) | Flight path planning method and device applied to underground coal mine intelligent inspection unmanned aerial vehicle | |
| CN109785388B (en) | Short-distance accurate relative positioning method based on binocular camera | |
| CN118115561A (en) | Deformation detection method and medium for tunnel structure | |
| CN116929336A (en) | Minimum error-based laser reflection column SLAM (selective laser absorption) mapping method | |
| CN115792805A (en) | Passive auxiliary positioning method and system based on azimuth information | |
| CN112907567B (en) | SAR image ordered artificial structure extraction method based on spatial reasoning method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |