A registration method for overbreak and underbreak calculation of mountain tunnels based on hierarchical clustering algorithm

Abstract

The present invention relates to the technical field of mountain tunnel scanning. The present invention discloses a mountain tunnel over-break and under-break calculation registration method based on a hierarchical clustering algorithm, comprising the following steps: step S1: using a three-dimensional laser scanner to scan the point cloud data and excavation data of a rock wall to be excavated in a mountain tunnel, building a 3D tunnel single-loop model based on the tunnel size corresponding to the point cloud data and excavation data, and obtaining point cloud data entity data based on the 3D tunnel single-loop model; step S2: performing clustering and denoising processing on the point cloud data and excavation data based on a hierarchical clustering algorithm to obtain target point cloud data; step S3: using an ICP algorithm to perform registration and comparison on the target point cloud data and the point cloud data entity data; step S4: using Boolean operations to obtain the point cloud data distance between the registered target point cloud data and the point cloud data entity data; step S5: the point cloud data distance is the over-break and under-break data, and the tunnel over-break and under-break data is output.

Description

Technical Field

[0001] The invention relates to the technical field of mountain tunnel scanning, and more specifically to a mountain tunnel over-break and under-break calculation and registration method based on a hierarchical clustering algorithm. Background Art

[0002] During the excavation of mountain tunnels, whether over-excavation or under-excavation occurs during the excavation process has always been a concern of all parties. For example, patent application No. 201910223238.3 discloses a tunnel primary support and secondary lining invasion limit and thickness analysis method, which uses a 3D laser scanner to perform a 3D panoramic scan of the excavated rough hole, primary support and secondary lining to obtain massive point cloud data of the scanned object, with detailed description of the target and high sampling rate, solving the limitations, randomness and poor representativeness of the traditional method of detecting sections.

[0003] However, in tunnel construction under complex geological conditions, when the surrounding rock conditions are poor, over-excavation and under-excavation are very likely to occur. If the degree of over-excavation and under-excavation is not accurately measured, it will have a significant impact on the safety assurance, schedule planning, and cost control during the tunnel excavation and construction process. It is more likely to cause rework due to poor quality. However, due to the complex construction environment of mountain tunnel excavation, it is difficult to complete the accurate calculation of over-excavation and under-excavation.

[0004] In view of this, the present invention provides a mountain tunnel over-break and under-break calculation and registration method based on a hierarchical clustering algorithm. Summary of the invention

[0005] In order to overcome the problems in the prior art, the present invention proposes a mountain tunnel over-break and under-break calculation and registration method based on a hierarchical clustering algorithm. Based on the hierarchical clustering algorithm, the rock wall information is identified and CYCLONE is used to complete point cloud data denoising. Then, by establishing a three-dimensional model that conforms to the design size, the three-dimensional model is converted into point cloud data. Then, with the help of the ICP algorithm, the two phases of point cloud data of the design model point cloud data and the tunnel point cloud data are registered, and then a value between the two phases of point cloud data is obtained.

[0006] According to one aspect of the present invention, a mountain tunnel over-break and under-break calculation and registration method based on a hierarchical clustering algorithm is provided, comprising the following steps:

[0007] Step S1: Use a 3D laser scanner to scan the point cloud data excavation data of the rock wall to be excavated in the mountain tunnel, and propose a 3D tunnel single-loop model based on the tunnel size corresponding to the point cloud data excavation data, and obtain point cloud data entity data based on the 3D tunnel single-loop model;

[0008] Step S2: clustering and denoising the excavation data of the point cloud data based on a hierarchical clustering algorithm to obtain target point cloud data;

[0009] Step S3: Use the ICP algorithm to align and compare the target point cloud data and the point cloud data entity data;

[0010] Step S4: obtaining the point cloud data distance between the registered target point cloud data and the point cloud data entity data by means of Boolean operation;

[0011] Step S5: The point cloud data distance is over-break and under-break data, and the over-break and under-break data of the tunnel are output.

[0012] As a preferred technical solution of the present invention, the acquisition logic of the target point cloud data is:

[0013] The point cloud data mining data is used as sample data, and the sample data is gradually merged into hierarchical clustered point cloud data mining cluster data through a hierarchical clustering algorithm, wherein the point cloud data mining cluster data is cluster data or a data subset of the sample data;

[0014] Remove the redundant noise points in the point cloud data excavation cluster data, and intercept the single-ring point cloud data with a single-ring length of 50 cm;

[0015] The single-loop point cloud data is imported into the 3D tunnel single-loop model and converted into target point cloud data based on the 3D tunnel single-loop model.

[0016] As a preferred technical solution of the present invention, the specific acquisition logic of the point cloud data excavation cluster data is:

[0017] Taking the point cloud data excavation data as an independent sample cluster, the point cloud data excavation data includes a plurality of point cloud data points;

[0018] Create a Euclidean distance matrix, use the Euclidean distance to calculate the distance between different point cloud data points, mark them as distance results, and compare the calculated distance results with the preset distance results.

[0019] The point cloud data points whose distance results meet the preset distance results are marked as the same cluster of point cloud data points, and the point cloud data points in the same cluster are combined and iteratively processed through the Euclidean distance matrix until clustering generates a cluster of data;

[0020] Based on the above steps, tree-shaped cluster data is generated by clustering, wherein the tree-shaped cluster data includes a root node and a data node, the root node represents the cloud excavation data corresponding to the entire data set, and the data node represents the point cloud data excavation cluster data corresponding to each cluster data.

[0021] As a preferred technical solution of the present invention, another acquisition logic of the distance results between different point cloud data points is:

[0022] Calculate the point cloud data centroid based on each of the point cloud data points, mark the distance between the centroids corresponding to the two point cloud data points as a distance result,

[0023] Merge the distance results of two point cloud data points whose centroid distance is less than the preset merging distance to generate point cloud data excavation cluster data;

[0024] Merge two point cloud data points that are less than the preset merge distance, and then update the centroid distance. When the number of point cloud data clusters in the point cloud data excavation cluster data is 1, the loop terminates.

[0025] As a preferred technical solution of the present invention, the specific logic of point cloud data registration based on ICP algorithm is:

[0026] First, two point cloud data sets are given, representing the target point cloud data and the point cloud data entity data respectively; specifically:

[0027] Target point cloud data:

[0028] Point cloud data entity data:

[0029] Among them: a i and p i Represents the point cloud data coordinates of the target point cloud data and the point cloud data entity data, N A and N P Respectively represent the number of point cloud data of target point cloud data and point cloud data entity data;

[0030] The point cloud data coordinates corresponding to the target point cloud data and the point cloud data entity data are stored in xyz format and marked as rock wall point cloud data coordinate information and model point cloud data coordinate information;

[0031] Calculate the rotation matrix and translation matrix from the rock wall point cloud data coordinate information to the model point cloud data coordinate information;

[0032] The calculated rotation matrix and translation vector are used for the next generation of iterative calculations until the error value is less than the set threshold, and the rotation matrix and translation vector are output.

[0033] As a preferred technical solution of the present invention, the logic for obtaining the point cloud data distance is:

[0034] The translation vector t and the rotation matrix R are used to calculate the point cloud data distance from the rock wall point cloud data coordinate information to the model point cloud data coordinate information through the Boolean operation formula;

[0035] The Boolean operation formula is:

[0036]

[0037] Among them, E(R,t) represents the point cloud data distance from the rock wall point cloud data coordinate information to the model point cloud data coordinate information.

[0038] According to another aspect of the present invention, a computer program product stored on a computer-readable medium is provided, characterized in that it includes a computer-readable program, which, when executed on an electronic device, provides a user input interface to implement a mountain tunnel over-break and under-break calculation and alignment method based on a hierarchical clustering algorithm.

[0039] The technical effects and advantages of the mountain tunnel over-break and under-break calculation registration method based on the hierarchical clustering algorithm of the present invention are as follows:

[0040] The present invention acquires excavation data of point cloud data by means of a three-dimensional laser scanner, clusters the rock wall point cloud data based on a hierarchical clustering algorithm, identifies and removes noise after clustering, and then imports cyclone to extract single-ring point cloud data with a length of about 50 centimeters, plans to build a 3D tunnel single-ring model of a designed size, converts the planned 3D tunnel single-ring model into point cloud data called target point cloud data, aligns and compares the collected point cloud data entity data and the target point cloud data by means of an ICP algorithm, and then calculates the distance between the two models through Boolean operations to obtain specific data of over-excavation and under-excavation of mountain tunnels; at the same time, the present invention provides an expanded reference application for the alignment of two phases of point cloud data. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A flowchart of a mountain tunnel overbreak and underbreak calculation and registration method based on a hierarchical clustering algorithm provided by the present invention;

[0042] Figure 2 A flow chart of the present invention using a hierarchical clustering algorithm to identify clustered rock wall information;

[0043] Figure 3 This is a processing flow chart of the present invention using the ICP algorithm to perform point cloud data registration. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] Example 1

[0046] See also Figure 1As shown, the mountain tunnel overbreak and underbreak calculation registration method based on the hierarchical clustering algorithm described in this embodiment can quickly complete the calculation of the mountain tunnel overbreak and underbreak, including the following steps:

[0047] Step S1: Use a 3D laser scanner to scan the point cloud data excavation data of the rock wall to be excavated in the mountain tunnel, and propose a 3D tunnel single-loop model based on the tunnel size corresponding to the point cloud data excavation data, and obtain point cloud data entity data based on the 3D tunnel single-loop model;

[0048] Step S2: clustering and denoising the excavation data of the point cloud data based on a hierarchical clustering algorithm to obtain target point cloud data;

[0049] Step S3: Use the ICP algorithm to align and compare the target point cloud data and the point cloud data entity data;

[0050] Step S4: obtaining the point cloud data distance between the registered target point cloud data and the point cloud data entity data by means of Boolean operation;

[0051] Step S5: The point cloud data distance is over-break and under-break data, and the over-break and under-break data of the tunnel are output.

[0052] The acquisition logic of the target point cloud data:

[0053] The point cloud data mining data is used as sample data, and the sample data is gradually merged into hierarchical clustered point cloud data mining cluster data through a hierarchical clustering algorithm, wherein the point cloud data mining cluster data is an increasingly larger cluster or data subset;

[0054] Remove the redundant noise points in the point cloud data excavation cluster data, and intercept the single-ring point cloud data with a single-ring length of 50 cm;

[0055] The single-loop point cloud data is imported into the 3D tunnel single-loop model and converted into target point cloud data based on the 3D tunnel single-loop model.

[0056] It should be noted that the hierarchical clustering algorithm is a clustering method based on a tree structure. It gradually merges the point cloud data in the data set into larger and larger clusters or subsets, and finally forms a hierarchical clustering tree. This tree structure can provide clustering results at multiple different levels, allowing the clustering structure of the data to be understood at different levels of abstraction.

[0057] The specific acquisition logic of the point cloud data mining cluster data is as follows:

[0058] Taking the point cloud data excavation data as an independent sample cluster, the point cloud data excavation data includes a plurality of point cloud data points;

[0059] Create a Euclidean distance matrix, use the Euclidean distance to calculate the distance between different point cloud data points, mark them as distance results, and compare the calculated distance results with the preset distance results.

[0060] The point cloud data points whose distance results meet the preset distance results are marked as the same cluster of point cloud data points, and the point cloud data points in the same cluster are combined and iteratively processed through the Euclidean distance matrix until clustering generates a cluster of data;

[0061] Based on the above steps, tree-shaped cluster data is generated by clustering, wherein the tree-shaped cluster data includes a root node and a data node, the root node represents the cloud excavation data corresponding to the entire data set, and the data node represents the point cloud data excavation cluster data corresponding to each cluster data.

[0062] It should be noted that there are usually multiple ways to merge the agglomerative hierarchical clustering, using Euclidean distance and single linkage to decide which clusters should be merged.

[0063] Assuming that each point cloud data point is a cluster, calculate the similarity between each cluster to obtain a similarity matrix; use the Euclidean distance matrix to obtain the distance results between different point cloud data points, and merge the two clusters with the closest distance results into one, that is, merge two point cloud data points into one point cloud data point, merge the two point cloud data points with the highest similarity, and then update the similarity matrix. When the number of point cloud data clusters is 1, the loop terminates.

[0064] Use Euclidean distance to decide which clusters should be merged; identify rock wall information and remove noise outside the rock wall information.

[0065] Another logic for obtaining the distance results between different point cloud data points is:

[0066] Calculate the point cloud data centroid based on each of the point cloud data points, mark the distance between the centroids corresponding to the two point cloud data points as a distance result,

[0067] Merge the distance results of two point cloud data points whose centroid distance is less than the preset merging distance to generate point cloud data excavation cluster data;

[0068] Merge two point cloud data points that are less than the preset merge distance, and then update the centroid distance. When the number of point cloud data clusters in the point cloud data excavation cluster data is 1, the loop terminates.

[0069] The specific logic of point cloud data registration based on ICP algorithm is:

[0070] First, two point cloud data sets are given, representing the target point cloud data and the point cloud data entity data respectively; specifically:

[0071] Target point cloud data:

[0072] Point cloud data entity data:

[0073] Among them: a i and p i Represents the point cloud data coordinates of the target point cloud data and the point cloud data entity data, N A and N P Respectively represent the number of point cloud data of target point cloud data and point cloud data entity data;

[0074] The point cloud data coordinates corresponding to the target point cloud data and the point cloud data entity data are stored in xyz format and marked as rock wall point cloud data coordinate information and model point cloud data coordinate information;

[0075] Calculate the rotation matrix and translation matrix from the rock wall point cloud data coordinate information to the model point cloud data coordinate information;

[0076] The calculated rotation matrix and translation vector are used for the next generation of iterative calculations until the error value is less than the set threshold, that is, the two groups of point cloud data are infinitely close to coincidence, or the distance is less than a fixed limit; output the rotation matrix and translation vector

[0077] The translation vector t and the rotation matrix R are used to calculate the point cloud data distance from the rock wall point cloud data coordinate information to the model point cloud data coordinate information through the formula;

[0078] The specific formula is:

[0079]

[0080] Among them, E(R,t) represents the point cloud data distance from the rock wall point cloud data coordinate information to the model point cloud data coordinate information.

[0081] The present invention discloses a method for calculating and registering over-excavation and under-excavation of mountain tunnels based on a hierarchical clustering algorithm. For the excavation data of point cloud data collected or obtained by means of a three-dimensional laser scanner, the rock wall point cloud data is first clustered by means of a hierarchical clustering algorithm, and identification and denoising are performed after clustering. Then, cycle one is imported to extract single-ring point cloud data with a length of about 50 centimeters. A 3D tunnel single-ring model of a designed size is proposed to be built, and the proposed 3D tunnel single-ring model is converted into point cloud data called target point cloud data. The collected point cloud data entity data and the target point cloud data are registered and compared by means of the ICP algorithm, and then the distance between the two models is calculated by Boolean operation to obtain the specific data of over-excavation and under-excavation of the mountain tunnel. At the same time, the present invention provides an expanded reference application for the registration of two-phase point cloud data.

[0082] Example 2

[0083] In an exemplary embodiment, a computer program product stored on a computer-readable medium includes a computer-readable program, which, when executed on an electronic device, provides a user input interface to implement a mountain tunnel over-break and under-break calculation and registration method based on a hierarchical clustering algorithm.

[0084] For example, the computer readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a compact disc (CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0085] In an exemplary embodiment, a computer program product or a computer program is also provided, the computer program product or the computer program includes one or more program codes, and the one or more program codes are stored in a computer-readable storage medium. One or more processors of an electronic device can read the one or more program codes from the computer-readable storage medium, and the one or more processors execute the one or more program codes, so that the electronic device can perform the above-mentioned mountain tunnel over-break and under-break calculation and registration method based on the hierarchical clustering algorithm.

[0086] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0087] It should be understood that determining B based on A does not mean determining B only based on A. B can also be determined based on A and / or other information.

[0088] A person of ordinary skill in the art will understand that all or part of the steps to implement the above embodiments can be completed by hardware, or can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. The above-mentioned storage medium can be a read-only memory, a disk or an optical disk, etc.

[0089] The above description is only an optional embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0090] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0091] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0092] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0093] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A mountain tunnel over-break and under-break calculation registration method based on a hierarchical clustering algorithm, characterized in that: The following steps are involved: Step S1: Use a 3D laser scanner to scan the point cloud data excavation data of the rock wall to be excavated in the mountain tunnel, and propose a 3D tunnel single-loop model based on the tunnel size corresponding to the point cloud data excavation data, and obtain point cloud data entity data based on the 3D tunnel single-loop model; Step S2: clustering and denoising the excavation data of the point cloud data based on a hierarchical clustering algorithm to obtain target point cloud data; The acquisition logic of the target point cloud data: The point cloud data mining data is used as sample data, and the sample data is gradually merged into hierarchical clustered point cloud data mining cluster data through a hierarchical clustering algorithm, wherein the point cloud data mining cluster data is cluster data or a data subset of the sample data; Remove the redundant noise points in the point cloud data excavation cluster data, and intercept the single-ring point cloud data with a single-ring length of 50 cm; Import the single-loop point cloud data into the 3D tunnel single-loop model, and convert it into target point cloud data based on the 3D tunnel single-loop model; Step S3: Use the ICP algorithm to align and compare the target point cloud data and the point cloud data entity data; Step S4: obtaining the point cloud data distance between the registered target point cloud data and the point cloud data entity data by means of Boolean operation; The logic for obtaining the distance of the point cloud data is: The translation vector t and the rotation matrix R are used to calculate the point cloud data distance from the rock wall point cloud data coordinate information to the model point cloud data coordinate information through the Boolean operation formula; The Boolean operation formula is: Among them, E(R,t) represents the point cloud data distance from the rock wall point cloud data coordinate information to the model point cloud data coordinate information; Step S5: The point cloud data distance is over-break and under-break data, and the over-break and under-break data of the tunnel are output.

2. The method for calculating and registering over-break and under-break of mountain tunnels based on a hierarchical clustering algorithm according to claim 1, characterized in that: The specific acquisition logic of the point cloud data excavation cluster data is: Taking the point cloud data excavation data as an independent sample cluster, the point cloud data excavation data includes a plurality of point cloud data points; Create a Euclidean distance matrix, use the Euclidean distance to calculate the distance between different point cloud data points, mark them as distance results, and compare the calculated distance results with the preset distance results. The point cloud data points whose distance results meet the preset distance results are marked as the same cluster of point cloud data points, and the point cloud data points in the same cluster are combined and iteratively processed through the Euclidean distance matrix until clustering generates a cluster of data; Based on the above steps, tree-shaped cluster data is generated by clustering, wherein the tree-shaped cluster data includes a root node and a data node, the root node represents the cloud excavation data corresponding to the entire data set, and the data node represents the point cloud data excavation cluster data corresponding to each cluster data.

3. The method for calculating and registering over-break and under-break of mountain tunnels based on a hierarchical clustering algorithm according to claim 1, characterized in that: Another logic for obtaining the distance results between different point cloud data points is: Calculate the point cloud data centroid based on each of the point cloud data points, mark the distance between the centroids corresponding to the two point cloud data points as a distance result, Merge the distance results of two point cloud data points whose centroid distance is less than the preset merging distance to generate point cloud data excavation cluster data; Merge two point cloud data points that are less than the preset merge distance, and then update the centroid distance. When the number of point cloud data clusters in the point cloud data excavation cluster data is 1, the loop terminates.

4. The method for calculating and registering over-break and under-break of mountain tunnels based on a hierarchical clustering algorithm according to claim 1, characterized in that: The specific logic of point cloud data registration based on the ICP algorithm is: First, two point cloud data sets are given, representing the target point cloud data and the point cloud data entity data respectively; specifically: Target point cloud data: Point cloud data entity data: Among them: a i and p i Represents the point cloud data coordinates of the target point cloud data and the point cloud data entity data, N A and N P Respectively represent the number of point cloud data of target point cloud data and point cloud data entity data; The point cloud data coordinates corresponding to the target point cloud data and the point cloud data entity data are stored in xyz format and marked as rock wall point cloud data coordinate information and model point cloud data coordinate information; Calculate the rotation matrix and translation matrix from the rock wall point cloud data coordinate information to the model point cloud data coordinate information; The calculated rotation matrix and translation vector are used for the next generation of iterative calculations until the error value is less than the set threshold, and the rotation matrix and translation vector are output.

5. A computer program product stored on a computer readable medium, characterized in that: It comprises a computer readable program, which, when executed on an electronic device, provides a user input interface to implement a mountain tunnel over-break and under-break calculation and registration method based on a hierarchical clustering algorithm as described in any one of claims 1 to 4.

Images