CN117556703B - Method, device and equipment for identifying rock mass structural surface of side slope and readable storage medium - Google Patents
Method, device and equipment for identifying rock mass structural surface of side slope and readable storage medium Download PDFInfo
- Publication number
- CN117556703B CN117556703B CN202311570766.9A CN202311570766A CN117556703B CN 117556703 B CN117556703 B CN 117556703B CN 202311570766 A CN202311570766 A CN 202311570766A CN 117556703 B CN117556703 B CN 117556703B
- Authority
- CN
- China
- Prior art keywords
- rock mass
- mass structural
- model
- point cloud
- cubes
- 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.)
- Active
Links
- 239000011435 rock Substances 0.000 title claims abstract description 149
- 238000000034 method Methods 0.000 title claims abstract description 85
- 238000012549 training Methods 0.000 claims abstract description 45
- 238000012360 testing method Methods 0.000 claims abstract description 43
- 238000013136 deep learning model Methods 0.000 claims abstract description 19
- 230000011218 segmentation Effects 0.000 claims abstract description 11
- 238000013135 deep learning Methods 0.000 claims description 24
- 238000010276 construction Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 7
- 238000004422 calculation algorithm Methods 0.000 claims description 4
- 238000002372 labelling Methods 0.000 claims description 4
- 238000000926 separation method Methods 0.000 claims description 2
- 239000010410 layer Substances 0.000 description 24
- 238000004891 communication Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 238000001514 detection method Methods 0.000 description 6
- 238000011176 pooling Methods 0.000 description 4
- 230000005236 sound signal Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- KLDZYURQCUYZBL-UHFFFAOYSA-N 2-[3-[(2-hydroxyphenyl)methylideneamino]propyliminomethyl]phenol Chemical compound OC1=CC=CC=C1C=NCCCN=CC1=CC=CC=C1O KLDZYURQCUYZBL-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 201000001098 delayed sleep phase syndrome Diseases 0.000 description 1
- 208000033921 delayed sleep phase type circadian rhythm sleep disease Diseases 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000002344 surface layer Substances 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A10/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE at coastal zones; at river basins
- Y02A10/23—Dune restoration or creation; Cliff stabilisation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a method, a device and equipment for identifying a rock mass structural plane of a side slope and a readable storage medium, and relates to the technical field of side slope protection, wherein the method comprises the steps of obtaining point cloud data of the side slope and generating a three-dimensional point cloud model of the side slope; performing cube segmentation on the three-dimensional point cloud model to obtain a plurality of cubes, and eliminating point cloud outliers in the cubes; grouping rock mass structural planes of the cubes to obtain a plurality of rock mass structural planes; constructing a data set by all cubes and rock structural planes, constructing a deep learning model, training and testing the deep learning model, and generating a rock structural plane identification model; the method is used for real-time identification tasks of the point cloud model, and can be used for identifying the structural faces more quickly and classifying the occurrence of the occurrence.
Description
Technical Field
The invention relates to the technical field of slope protection, in particular to a method, a device and equipment for identifying a rock mass structural surface of a slope and a readable storage medium.
Background
Due to the complex topography in western regions, numerous high and steep slopes are distributed, and the surface layer of the rock mass is extremely broken. Therefore, the research on the characterization and mechanical properties of the slope structural surface is more and more urgent. The stability of the side slope is controlled by the structure surface which is generated in the rock side slope, the structural surface failure is an important factor for leading the rock side slope to be unstable, the structural surface occurrence information is an important index for evaluating the development degree of the structural surface of the rock side slope, the structural surface occurrence information is finely investigated, the occurrence information is classified, and the structural surface occurrence information is a basis for rock mass quality classification and side slope stability evaluation, so that the method has important engineering significance on how to efficiently, accurately and intelligently extract the structural surface information.
Disclosure of Invention
The invention aims to provide a method, a device and equipment for identifying a rock mass structural plane of a side slope and a readable storage medium, so as to solve the problems. In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
in a first aspect, the application provides a method for identifying a rock mass structural plane of a side slope, comprising the following steps:
acquiring point cloud data of a side slope, and generating a three-dimensional point cloud model of the side slope;
Performing cube segmentation on the three-dimensional point cloud model to obtain a plurality of cubes, and eliminating point cloud outliers in the cubes;
Traversing all cubes after eliminating the outliers of the point cloud, and grouping rock structural planes of the cubes to obtain a plurality of rock structural planes;
constructing a data set by all cube and rock mass structural planes, and dividing the data set into a training set and a testing set;
constructing a deep learning model, and training and testing the deep learning model by using the training set and the testing set to generate a rock structural surface recognition model;
And inputting the point cloud data of the side slope to be predicted into the rock mass structural plane identification model to obtain a plurality of rock mass structural planes of the side slope to be predicted, and calculating the occurrence of each rock mass structural plane.
In a second aspect, the present application also provides a device for identifying a rock mass structural plane of a side slope, including:
and a model generation module: the method comprises the steps of obtaining point cloud data of a side slope and generating a three-dimensional point cloud model of the side slope;
and (3) a rejection module: the method comprises the steps of performing cube segmentation on the three-dimensional point cloud model to obtain a plurality of cubes, and eliminating point cloud outliers in the cubes;
And a grouping module: the method comprises the steps of traversing all cubes after point cloud outliers are removed, and grouping rock structural planes of the cubes to obtain a plurality of rock structural planes;
the data set construction module: for constructing a dataset from all cube and rock mass structural faces, dividing the dataset into a training set and a testing set;
model construction module: the method comprises the steps of constructing a deep learning model, and generating a rock structural surface recognition model after training and testing the deep learning model by utilizing the training set and the testing set;
The calculation module: and the point cloud data of the slope to be predicted is input into the rock mass structural plane recognition model to obtain a plurality of rock mass structural planes of the slope to be predicted, and the occurrence of each rock mass structural plane is calculated.
In a third aspect, the present application also provides a rock mass structural plane identification device for a side slope, comprising:
A memory for storing a computer program;
and the processor is used for realizing the step of the rock mass structural plane identification method of the side slope when executing the computer program.
In a fourth aspect, the present application also provides a readable storage medium having stored thereon a computer program which when executed by a processor implements the steps of the slope-based rock mass structural face identification method described above.
The beneficial effects of the invention are as follows:
According to the invention, the rock mass point cloud is divided into cubes with equal size, so that the situation that the quantity difference of each sub-concentrated point cloud is large is avoided, the calculation and identification effects of the point cloud are enhanced, meanwhile, the interference point cloud can be effectively eliminated, the training set and the testing set are constructed by utilizing the eliminated point cloud data, and the degree of human interference is reduced. Meanwhile, the point cloud model is identified through the deep learning network, and the average identification of a single point cloud model only needs 2.8ms, so that the method has shorter training time and faster identification speed, is suitable for the real-time identification task of the point cloud model, and can more quickly identify the structural surface and classify the occurrence.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the embodiments of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a method for identifying a rock mass structural plane of a side slope according to an embodiment of the invention;
FIG. 2 is a schematic diagram of partitioning a point cloud model according to an embodiment of the present invention;
FIG. 3 is a schematic view of RANSAC method plane point fitting in an embodiment of the invention;
FIG. 4 is a schematic structural diagram of a point cloud model according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a deep learning model according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of a yield calculation space coordinate system in an embodiment of the invention;
FIG. 7 is a schematic diagram of a rock mass structural plane identification device of a side slope according to an embodiment of the invention;
fig. 8 is a schematic structural diagram of a rock mass structural plane identification device of a side slope according to an embodiment of the present invention.
The marks in the figure:
800. Rock mass structural plane identification equipment of the side slope; 801. a processor; 802. a memory; 803. a multimedia component; 804. an I/O interface; 805. a communication component.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures. Meanwhile, in the description of the present invention, the terms "first", "second", and the like are used only to distinguish the description, and are not to be construed as indicating or implying relative importance.
Example 1:
the embodiment provides a rock mass structural plane identification method of a side slope.
Referring to fig. 1, the method is shown to include the steps of:
S1, acquiring point cloud data of a side slope, and generating a three-dimensional point cloud model of the side slope;
Specifically, the method comprises the steps of carrying out real-time stepping on a side slope site, possibly comprehensively collecting project data, making project detection schemes and making unmanned aerial vehicle flight plans according to the detection fineness requirements of all parties such as construction parties, construction parties and the like, wherein the project detection schemes comprise unmanned aerial vehicle take-off and landing places, flight time, flight tracks, detection control points, detection key areas and the like.
Preferably, the method adopts a five-eyepiece camera carried by the unmanned aerial vehicle Buddhist L1 in the Xinjiang to carry out close-up photogrammetry on the side slope. And collecting slope point cloud data, storing the data, and importing intelligent map software into the intelligent map software to generate a real three-dimensional point cloud model of the slope.
Referring to fig. 2, based on the above embodiment, the method further includes:
s2, performing cube segmentation on the three-dimensional point cloud model to obtain a plurality of cubes, and eliminating point cloud outliers in the cubes;
Specifically, the step S2 includes:
S21, performing cube segmentation on point cloud data in the three-dimensional point cloud model to obtain cubes with a plurality of first preset volumes;
specifically, a minimum number (nmin) and a maximum number (nmax) of point clouds in a minimum cube are set, and then the point cloud data is divided into 8 cubes of equal volume.
S22, continuously dividing the cube with the first preset volume into cubes with the second preset volume;
the cube generated in step S21 is divided into 8 cubes of equal volume again.
S23, continuously dividing the cube with the second preset volume until the number of point clouds in the divided cube is within a preset range, namely the number of point clouds in the cube is larger than the minimum number (nmin) and smaller than the maximum number (nmax).
Based on the above embodiment, the method further includes:
s3, traversing all cubes after eliminating point cloud outliers, and grouping rock structural faces of the cubes to obtain a plurality of rock structural faces;
Specifically, the step S3 includes:
s31, acquiring point clouds forming a cube;
Preferably, an open3d construction point cloud outlier rejection program is adopted to reject discrete point clouds and noise data in the cube, and effective data point clouds are reserved.
S32, calculating a normal vector included angle between any two point clouds;
S33, when the normal vector included angle is smaller than a preset threshold value, the two point clouds belong to the same group of rock mass structural planes;
preferably, the preset threshold e=20°, and when the included angle of two point clouds is smaller than the threshold, they are considered to belong to the same group of structural planes.
S34, calculating a normal vector of a rock mass structural plane by adopting a RANSAC method, and endowing the normal vector of the rock mass structural plane to all point clouds on the rock mass structural plane, wherein the point clouds are shown in FIG. 3;
preferably, the point clouds in the cube are subjected to coplanar detection and normal vector calculation. Then a general expression of the plane equation is obtained:
Ax+By+Cz+D＝0 (1)
Wherein A, B, C, D are parameters, n (A, B, C) is the normal vector of the plane, and the normal vector of the fitted plane is given to all point clouds on the plane.
S35, labeling the structural faces of each rock body, such as structural faces J1 and J2 …, see FIG. 4.
Based on the above embodiment, the method further includes:
S4, constructing a data set by all cube and rock mass structural planes, and dividing the data set into a training set and a testing set;
Specifically, the step S4 includes:
S41, selecting 25% -40% of point clouds corresponding to the cubes in all the cubes as a training set, using the rest point clouds as a test set, and respectively storing training set data and test set data in a classified manner;
S42, taking point clouds in the training set and the testing set as input labels of the model and rock mass structural planes as output labels of the model.
Based on the above embodiment, the method further includes:
S5, referring to FIG. 5, constructing a deep learning model, and training and testing the deep learning model by using the training set and the testing set to generate a rock structural surface recognition model;
specifically, the step S5 includes:
S51, constructing a deep learning network model by an input layer, a convolution layer, a pooling layer, a full connection layer and an output layer, wherein the deep learning network model is preferably LightPointNet;
specifically, n point cloud data in the training set are input into the input layer, each point is composed of three-dimensional coordinates (x, y, z), and the actual input is an n×3-dimensional tensor.
The convolution layers are provided with 3 layers, wherein 3 parameters in the 1 st convolution layer (64,3,1) sequentially represent the channel number 64, the size of a convolution kernel 1 multiplied by 3 and the step length of 1 multiplied by 1; the 3 parameters in the 2 nd convolution layer (128,1,1) represent the channel number 128, the size of the convolution kernel 1×1, and the step size 1×1; 3 rd convolution layer (256,1,1).
And connecting a maximum pooling layer after the layer 3 convolution, and carrying out maximum pooling on each channel in the layer 3 convolution to obtain a tensor of 1 multiplied by 128 dimensions.
And then taking the tensor of 1×128 dimensions as input, obtaining the tensor of 1×258 dimensions through the full-connection layer, adding a layer of Dropout behind the full-connection layer for preventing overfitting, and setting the Dropout rate to be 0.5.
The Dropout is followed by an output layer and a loss layer of k dimensions.
S52, inputting the training set into a deep learning network model, and obtaining a trained deep learning network model after the deep learning network model learns the relation between the point cloud in the training set and the rock mass structural plane;
S53, inputting the test set into a trained deep learning network model, and generating a prediction structural surface by the deep learning network model according to point clouds in the test set;
s54, when the error between the predicted structural surface and the rock mass structural surface in the test set reaches a preset threshold value, training the deep learning model to obtain a rock mass structural surface identification model.
Based on the above embodiment, the method further includes:
s6, inputting point cloud data of the side slope to be predicted into the rock mass structural plane identification model to obtain a plurality of rock mass structural planes of the side slope to be predicted, and calculating the occurrence of each rock mass structural plane;
specifically, the step S6 includes:
s61, separating point clouds forming a plurality of rock mass structural planes, and dividing the separated point clouds into the rock mass structural planes by adopting a DBSCAN algorithm based on density;
S62, referring to FIG. 6, a RANSAC method is adopted to calculate a normal vector n of a rock mass structural plane, and the normal vector is calculated based on the following:
an included angle alpha between the structural surface and the horizontal plane, namely an included angle between a component C of a normal vector n on the Z axis and the Z axis: α=arccos (C);
the angle beta between the projection of the normal vector n in the horizontal direction and the north direction:
where A represents the component of normal vector n on the X axis and B represents the component of normal vector n on the Y axis.
Example 2:
As shown in fig. 7, the present embodiment provides a rock mass structural plane identification device for a side slope, the device including:
and a model generation module: the method comprises the steps of obtaining point cloud data of a side slope and generating a three-dimensional point cloud model of the side slope;
and (3) a rejection module: the method comprises the steps of performing cube segmentation on the three-dimensional point cloud model to obtain a plurality of cubes, and eliminating point cloud outliers in the cubes;
And a grouping module: the method comprises the steps of traversing all cubes after point cloud outliers are removed, and grouping rock structural planes of the cubes to obtain a plurality of rock structural planes;
the data set construction module: for constructing a dataset from all cube and rock mass structural faces, dividing the dataset into a training set and a testing set;
model construction module: the method comprises the steps of constructing a deep learning model, and generating a rock structural surface recognition model after training and testing the deep learning model by utilizing the training set and the testing set;
The calculation module: and the point cloud data of the slope to be predicted is input into the rock mass structural plane recognition model to obtain a plurality of rock mass structural planes of the slope to be predicted, and the occurrence of each rock mass structural plane is calculated.
Based on the above embodiment, the culling module includes:
A first dividing unit: the method comprises the steps of performing cube segmentation on point cloud data in a three-dimensional point cloud model to obtain cubes with a plurality of first preset volumes;
A second dividing unit: the method comprises the steps of continuously dividing a cube with a first preset volume into cubes with a second preset volume;
a third dividing unit: and the method is used for continuously dividing the cube with the second preset volume until the number of the point clouds in the divided cube is within a preset range.
Based on the above embodiments, the grouping module includes:
An acquisition unit: for obtaining a point cloud constituting a cube;
a first calculation unit: the method is used for calculating the normal vector included angle between any two point clouds;
a judging unit: when the normal vector included angle is smaller than a preset threshold value, the two point clouds belong to the same group of rock mass structural planes;
A second calculation unit: the method comprises the steps of calculating a normal vector of a rock mass structural plane by adopting a RANSAC method, and giving the normal vector of the rock mass structural plane to all point clouds on the rock mass structural plane;
Marking unit: the method is used for labeling each rock mass structural surface.
Based on the above embodiments, the data set construction module includes:
The selecting unit: the method comprises the steps of selecting 25% -40% of point clouds corresponding to cubes in all cubes as a training set, and using the rest point clouds as a testing set;
the setting unit: the method is used for taking point clouds in a training set and a testing set as input labels of the model and taking a rock mass structural plane as output labels of the model.
Based on the above embodiments, the model building module includes:
the construction unit: the method comprises the steps of constructing a deep learning network model by an input layer, a convolution layer, a pooling layer, a full connection layer and an output layer;
Training unit: the method comprises the steps of inputting a training set into a deep learning network model, and obtaining a trained deep learning network model after the deep learning network model learns the relation between point cloud and rock mass structural planes in the training set;
test unit: the method comprises the steps that a test set is input into a deep learning network model, and the deep learning network model generates a prediction structural surface according to point clouds in the test set;
a generation unit: and the method is used for obtaining a rock mass structural plane recognition model after the deep learning model training is completed when the error between the predicted structural plane and the rock mass structural plane in the test set reaches a preset threshold value.
Based on the above embodiments, the calculation module includes:
A separation unit: the method comprises the steps of separating point clouds forming a plurality of rock mass structural planes, and dividing the separated point clouds into the rock mass structural planes by adopting a DBSCAN algorithm based on density;
A third calculation unit: the method is used for calculating the normal vector of the rock mass structural plane by adopting the RANSAC method, and calculating the included angle between the structural member and the horizontal plane and the included angle between the projection of the normal vector in the horizontal direction and the north direction based on the normal vector.
It should be noted that, regarding the apparatus in the above embodiments, the specific manner in which the respective modules perform the operations has been described in detail in the embodiments regarding the method, and will not be described in detail herein.
Example 3:
Corresponding to the above method embodiment, a device for identifying a rock structural surface of a side slope is further provided in this embodiment, and a device for identifying a rock structural surface of a side slope described below and a method for identifying a rock structural surface of a side slope described above may be referred to correspondingly with each other.
Fig. 8 is a block diagram of a rock mass structural face identification device 800 of a side slope, according to an example embodiment. As shown in fig. 8, the rock mass structural face identification apparatus 800 of the side slope may include: a processor 801, a memory 802. The rock mass structural face identification device 800 of the side slope may also include one or more of a multimedia component 803, an i/O interface 804, and a communication component 805.
The processor 801 is configured to control the overall operation of the slope rock mass structural plane identification device 800, so as to complete all or part of the steps in the above-mentioned slope rock mass structural plane identification method. The memory 802 is used to store various types of data to support the operation of the rock mass structural face identification device 800 on the slope, which may include, for example, instructions for any application or method operating on the rock mass structural face identification device 800 on the slope, as well as application-related data, such as contact data, messages, pictures, audio, video, and so forth. The Memory 802 may be implemented by any type or combination of volatile or non-volatile Memory devices, such as static random access Memory (Static Random Access Memory, SRAM for short), electrically erasable programmable Read-Only Memory (ELECTRICALLY ERASABLE PROGRAMMABLE READ-Only Memory, EEPROM for short), erasable programmable Read-Only Memory (Erasable Programmable Read-Only Memory, EPROM for short), programmable Read-Only Memory (Programmable Read-Only Memory, PROM for short), read-Only Memory (ROM for short), magnetic Memory, flash Memory, magnetic disk, or optical disk. The multimedia component 803 may include a screen and an audio component. Wherein the screen may be, for example, a touch screen, the audio component being for outputting and/or inputting audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in the memory 802 or transmitted through the communication component 805. The audio assembly further comprises at least one speaker for outputting audio signals. The I/O interface 804 provides an interface between the processor 801 and other interface modules, which may be a keyboard, mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 805 is configured to provide wired or wireless communication between the rock mass face identification device 800 of the side slope and other devices. Wireless communication, such as Wi-Fi, bluetooth, near field communication (Near FieldCommunication, NFC for short), 2G, 3G, or 4G, or a combination of one or more thereof, the corresponding communication component 805 may therefore include: wi-Fi module, bluetooth module, NFC module.
In an exemplary embodiment, the rock mass structural face identification device 800 of the side slope may be implemented by one or more Application Specific Integrated Circuits (ASIC), digital signal processor (DIGITALSIGNAL PROCESSOR DSP), digital signal processing device (DIGITAL SIGNAL Processing Device DSPD), programmable logic device (Programmable Logic Device PLD), field programmable gate array (Field Programmable GATE ARRAY FPGA), controller, microcontroller, microprocessor or other electronic component for performing the above-described rock mass structural face identification method of the side slope.
In another exemplary embodiment, a computer readable storage medium is also provided comprising program instructions which, when executed by a processor, implement the steps of the rock mass structural face identification method of a side slope described above. For example, the computer readable storage medium may be the memory 802 described above including program instructions executable by the processor 801 of the rock mass structural face identification device 800 of a side slope to perform the rock mass structural face identification method of a side slope described above.
Example 4:
Corresponding to the above method embodiment, a readable storage medium is also provided in this embodiment, and a readable storage medium described below and a method for identifying a rock mass structural plane of a side slope described above may be referred to correspondingly.
A readable storage medium, on which a computer program is stored, which when being executed by a processor implements the steps of the method for identifying a rock mass structural plane of a side slope according to the above method embodiment.
The readable storage medium may be a usb disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, which may store various program codes.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.
Claims (10)
1. The method for identifying the rock mass structural surface of the side slope is characterized by comprising the following steps of:
acquiring point cloud data of a side slope, and generating a three-dimensional point cloud model of the side slope;
the three-dimensional point cloud model is subjected to cube segmentation, the minimum quantity and the maximum quantity of point clouds in the minimum cube are set, point cloud data are segmented into cubes with the same volume, and point cloud outliers in the cubes are removed, and the method comprises the following steps:
Adopting open3d to construct a point cloud outlier removing program, removing discrete point clouds and noise data in the cube, and retaining effective data point clouds;
Traversing all cubes after eliminating point cloud outliers, grouping rock mass structural planes on the cubes to obtain a plurality of rock mass structural planes, and comprising the following steps:
calculating a normal vector included angle between any two point clouds;
when the normal vector included angle is smaller than a preset threshold value, the two point clouds belong to the same group of rock mass structural planes;
calculating a normal vector of a rock mass structural plane by adopting a RANSAC method, and giving the normal vector of the rock mass structural plane to all point clouds on the rock mass structural plane;
Labeling each rock mass structural surface;
constructing a data set by all cube and rock mass structural planes, and dividing the data set into a training set and a testing set;
Constructing a deep learning model, training and testing the deep learning model by using the training set and the testing set, and generating a rock mass structural plane recognition model, wherein the method comprises the following steps of:
Inputting the training set into a deep learning network model, and obtaining a trained deep learning network model after the deep learning network model learns the relation between the point cloud in the training set and the rock mass structural plane;
Inputting the test set into a trained deep learning network model, and generating a prediction structural plane by the deep learning network model according to point clouds in the test set;
When the error between the predicted structural surface and the rock mass structural surface in the test set reaches a preset threshold value, training the deep learning model to obtain a rock mass structural surface identification model;
And inputting the point cloud data of the side slope to be predicted into the rock mass structural plane identification model to obtain a plurality of rock mass structural planes of the side slope to be predicted, and calculating the occurrence of each rock mass structural plane.
2. The method for identifying a rock mass structural plane of a side slope according to claim 1, wherein the three-dimensional point cloud model is subjected to cube segmentation to obtain a plurality of cubes, and point cloud outliers in the cubes are removed, and the method comprises the steps of:
performing cube segmentation on point cloud data in the three-dimensional point cloud model to obtain cubes with a plurality of first preset volumes;
continuously dividing the cube with the first preset volume into cubes with the second preset volume;
and continuously dividing the cube with the second preset volume until the number of point clouds in the divided cube is within a preset range.
3. A method of identifying rock mass face of a side slope according to claim 1, wherein a dataset is constructed from all cubes and rock mass faces, the dataset being divided into a training set and a test set, comprising:
selecting 25% -40% of point clouds corresponding to the cubes from all cubes as a training set, and using the rest point clouds as a test set;
and taking the point clouds in the training set and the testing set as input labels of the model and the rock mass structural plane as output labels of the model.
4. The method for identifying rock mass structural planes of a side slope according to claim 1, wherein the step of inputting point cloud data of the side slope to be predicted into the rock mass structural plane identification model to obtain a plurality of rock mass structural planes of the side slope to be predicted and calculating the occurrence of each rock mass structural plane comprises the steps of:
Separating point clouds forming a plurality of rock mass structural planes, and dividing the separated point clouds into the rock mass structural planes by adopting a DBSCAN algorithm based on density;
And calculating a normal vector of the rock mass structural plane by adopting a RANSAC method, and calculating an included angle between a structural member and a horizontal plane and an included angle between the projection of the normal vector in the horizontal direction and the north direction based on the normal vector.
5. A rock mass structural face identification device of a side slope, comprising:
and a model generation module: the method comprises the steps of obtaining point cloud data of a side slope and generating a three-dimensional point cloud model of the side slope;
And (3) a rejection module: the method for segmenting the three-dimensional point cloud model into cubes, setting the minimum quantity and the maximum quantity of point clouds in the minimum cubes, segmenting the point cloud data into a plurality of cubes with equal volumes, and eliminating point cloud outliers in the cubes comprises the following steps:
Adopting open3d to construct a point cloud outlier removing program, removing discrete point clouds and noise data in the cube, and retaining effective data point clouds;
And a grouping module: all cubes used for traversing and rejecting point cloud outliers are grouped into a plurality of rock mass structural planes after the rock mass structural planes of the cubes are obtained, and the method comprises the following steps:
calculating a normal vector included angle between any two point clouds;
when the normal vector included angle is smaller than a preset threshold value, the two point clouds belong to the same group of rock mass structural planes;
calculating a normal vector of a rock mass structural plane by adopting a RANSAC method, and giving the normal vector of the rock mass structural plane to all point clouds on the rock mass structural plane;
Labeling each rock mass structural surface;
the data set construction module: for constructing a dataset from all cube and rock mass structural faces, dividing the dataset into a training set and a testing set;
Model construction module: the method is used for constructing a deep learning model, training and testing the deep learning model by utilizing the training set and the testing set to generate a rock mass structural plane recognition model, and comprises the following steps:
Inputting the training set into a deep learning network model, and obtaining a trained deep learning network model after the deep learning network model learns the relation between the point cloud in the training set and the rock mass structural plane;
Inputting the test set into a trained deep learning network model, and generating a prediction structural plane by the deep learning network model according to point clouds in the test set;
When the error between the predicted structural surface and the rock mass structural surface in the test set reaches a preset threshold value, training the deep learning model to obtain a rock mass structural surface identification model;
The calculation module: and the point cloud data of the slope to be predicted is input into the rock mass structural plane recognition model to obtain a plurality of rock mass structural planes of the slope to be predicted, and the occurrence of each rock mass structural plane is calculated.
6. The slope rock mass face identification device of claim 5, wherein the culling module includes:
A first dividing unit: the method comprises the steps of performing cube segmentation on point cloud data in a three-dimensional point cloud model to obtain cubes with a plurality of first preset volumes;
A second dividing unit: the method comprises the steps of continuously dividing a cube with a first preset volume into cubes with a second preset volume;
a third dividing unit: and the method is used for continuously dividing the cube with the second preset volume until the number of the point clouds in the divided cube is within a preset range.
7. The slope rock mass face identification device of claim 5, wherein the data set construction module comprises:
The selecting unit: the method comprises the steps of selecting 25% -40% of point clouds corresponding to cubes in all cubes as a training set, and using the rest point clouds as a testing set;
the setting unit: the method is used for taking point clouds in a training set and a testing set as input labels of the model and taking a rock mass structural plane as output labels of the model.
8. The slope rock mass face identification device of claim 5, wherein the computing module comprises:
A separation unit: the method comprises the steps of separating point clouds forming a plurality of rock mass structural planes, and dividing the separated point clouds into the rock mass structural planes by adopting a DBSCAN algorithm based on density;
A third calculation unit: the method is used for calculating the normal vector of the rock mass structural plane by adopting the RANSAC method, and calculating the included angle between the structural member and the horizontal plane and the included angle between the projection of the normal vector in the horizontal direction and the north direction based on the normal vector.
9. A rock mass structural face identification device for a side slope, comprising:
A memory for storing a computer program;
a processor for carrying out the steps of the rock mass face identification method of a side slope as claimed in any one of claims 1 to 4 when executing said computer program.
10. A readable storage medium, characterized by: the readable storage medium has stored thereon a computer program which, when executed by a processor, implements the steps of the rock mass structural face identification method of a side slope as claimed in any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311570766.9A CN117556703B (en) | 2023-11-23 | 2023-11-23 | Method, device and equipment for identifying rock mass structural surface of side slope and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311570766.9A CN117556703B (en) | 2023-11-23 | 2023-11-23 | Method, device and equipment for identifying rock mass structural surface of side slope and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117556703A CN117556703A (en) | 2024-02-13 |
| CN117556703B true CN117556703B (en) | 2024-06-14 |
Family
ID=89814459
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311570766.9A Active CN117556703B (en) | 2023-11-23 | 2023-11-23 | Method, device and equipment for identifying rock mass structural surface of side slope and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117556703B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114608476A (en) * | 2022-03-09 | 2022-06-10 | 东北大学 | Intelligent analysis and extraction method for three-dimensional point cloud structural surface of complex rock mass |
| CN115147551A (en) * | 2022-07-18 | 2022-10-04 | 南昌工程学院 | Method for extracting homonymous feature points on surface of monitored object by remote ground laser scanning technology |
| CN116843946A (en) * | 2023-05-31 | 2023-10-03 | 中铁二院工程集团有限责任公司 | Tunnel rock mass main structural surface identification method and device based on image identification |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111178214B (en) * | 2019-12-23 | 2023-04-18 | 天津大学 | High and steep slope dangerous rock mass rapid identification method based on unmanned aerial vehicle photography technology |
| CN116403124A (en) * | 2022-11-30 | 2023-07-07 | 中国铁建昆仑投资集团有限公司 | Structural plane intelligent recognition method of three-dimensional point cloud surrounding rock based on DResNet-PointNet |
| CN116258972B (en) * | 2023-05-16 | 2023-08-01 | 四川安信科创科技有限公司 | Deep learning-based rock high-steep slope structural surface extraction method |
-
2023
- 2023-11-23 CN CN202311570766.9A patent/CN117556703B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114608476A (en) * | 2022-03-09 | 2022-06-10 | 东北大学 | Intelligent analysis and extraction method for three-dimensional point cloud structural surface of complex rock mass |
| CN115147551A (en) * | 2022-07-18 | 2022-10-04 | 南昌工程学院 | Method for extracting homonymous feature points on surface of monitored object by remote ground laser scanning technology |
| CN116843946A (en) * | 2023-05-31 | 2023-10-03 | 中铁二院工程集团有限责任公司 | Tunnel rock mass main structural surface identification method and device based on image identification |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117556703A (en) | 2024-02-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ding et al. | Object detection in aerial images: A large-scale benchmark and challenges | |
| CN112801050B (en) | Intelligent luggage tracking and monitoring method and system | |
| CN109948497A (en) | A kind of object detecting method, device and electronic equipment | |
| CN108604303A (en) | General image feature from bottom to top and the from top to bottom system and method for entity classification are merged for precise image/video scene classification | |
| CN111429698A (en) | Geological disaster early warning system | |
| CN113724388B (en) | High-precision map generation method, device, equipment and storage medium | |
| CN115861400B (en) | Target object detection method, training device and electronic equipment | |
| CN114387506A (en) | Transmission tower monitoring method and device, computer equipment and storage medium | |
| CN115331048B (en) | Image classification method, device, equipment and storage medium | |
| CN116482711A (en) | Local static environment sensing method and device for autonomous selection of landing zone | |
| CN111753911A (en) | Method and apparatus for fusing models | |
| Stefanidis et al. | Summarizing video datasets in the spatiotemporal domain | |
| JP6705886B2 (en) | Surveillance | |
| CN116266359A (en) | Target tracking method, device, computer equipment and storage medium | |
| CN114373162A (en) | Dangerous area personnel intrusion detection method and system for transformer substation video monitoring | |
| CN114169425A (en) | Training target tracking model and target tracking method and device | |
| CN117556703B (en) | Method, device and equipment for identifying rock mass structural surface of side slope and readable storage medium | |
| Zhou | 3D urban modeling from city-scale aerial LiDAR data | |
| CN112926681B (en) | Target detection method and device based on deep convolutional neural network | |
| CN113705651B (en) | Air condition target identification method and device based on radar information | |
| CN115116080A (en) | Table analysis method and device, electronic equipment and storage medium | |
| Shekade et al. | Vehicle Classification in Traffic Surveillance System using YOLOv3 Model | |
| Kakodra et al. | Query-by-object based video synopsis | |
| Akilandasowmya et al. | Human action analysis using K-NN classifier | |
| CN113505834A (en) | Method for training detection model, determining image updating information and updating high-precision map |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |