CN115753643A - Intelligent recognition method and system for rock mass fissures combining 3D scanning and image spectrum - Google Patents

Abstract

The invention discloses a system and method for intelligent identification of rock mass fissures combining three-dimensional scanning and image spectrum, comprising: obtaining three-dimensional point cloud data of the target rock mass through three-dimensional scanning; processing the three-dimensional point cloud data to obtain a two-dimensional coordinate system The following fracture point cloud data set; the image information and spectral information of the target rock mass are obtained by spectral scanning; the mineral distribution characteristics are extracted based on the spectral information, and the fracture image features are extracted based on the image information; The boundary intelligent recognition model obtains the boundary coordinates and filling mineral information of the fissures; the boundary coordinates of the fissures are checked and merged with the fissure point cloud data set in the two-dimensional coordinate system to realize the comprehensive and accurate identification of rock mass fissures. The invention can effectively make up for the defect that three-dimensional laser scanning cannot realize accurate identification and efficient extraction of parameters for filling cracks, and greatly improves the accuracy of crack identification.

Description

Method and system for intelligent identification of rock mass fissures by combining 3D scanning and image spectrum

technical field

The invention relates to the technical field of rock mass fissure recognition, in particular to an intelligent rock mass fissure recognition method and system that integrates three-dimensional scanning and image spectrum.

Background technique

The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

There are a large number of fracture structures in the rock mass, and these structures have a great influence on the physical and mechanical performance indicators, deformation characteristics, seepage properties and long-term stability of the rock mass. Fissures can be divided into filled and non-filled fissures according to whether they are filled or not. Common fissure fillings include breccia, sand, and limestone formed by tectonic, weathering, and groundwater effects, which are highly differentiated from the composition of surrounding rocks. , the probability of composition consistency is low.

Throughout the field of rock mass fissure identification, the current common identification methods include naked eye observation, image recognition, and three-dimensional laser scanning. The naked eye observation method often relies too much on manual experience, requires a high reserve of professional skills and knowledge of the observer, and cannot achieve accurate quantitative analysis of cracks. It is easily restricted by environmental factors such as weather and light, and has a large workload, low efficiency, and high labor costs. Low economic benefits; image recognition methods are often limited by image quality, have high requirements for algorithm and model integrity, training data often need to be preprocessed, cannot directly use the original model parameters, and the final recognition accuracy needs to be improved; 3D laser scanning technology through 3D laser The scanning equipment directly performs three-dimensional sampling on the surface of the rock mass. Compared with the traditional technology, although it can quickly, in-situ, and high-precision identify the cracks in the rock mass, because the three-dimensional scanning can only sample data on the surface of the rock mass, and The internal space of the fissures filled with fissures is filled with minerals, etc., and the boundaries of such fissures obtained by scanning are prone to distortion or even misjudgment. To sum up, there are certain limitations in the identification accuracy of using only one technology for crack identification.

Contents of the invention

In order to solve the above problems, the present invention proposes a method and system for intelligent identification of rock mass fissures that integrate 3D scanning and image spectrum, using 3D scanning technology to realize accurate identification of non-filled cracks and fuzzy recognition of filled cracks, using image spectrum Technology combined with neural network to realize accurate identification of filling fissures, supplement and check the 3D scanning results, and finally realize accurate, comprehensive, fast and intelligent identification of rock mass fissures.

In some embodiments, the following technical solutions are adopted:

An intelligent identification method for rock mass fissures that combines 3D scanning and image spectrum, including:

Obtain 3D point cloud data of the target rock mass through 3D scanning;

Processing the three-dimensional point cloud data, extracting the coordinates of the discontinuous point set, and performing dimension transformation on the extracted coordinates of the discontinuous point set, to obtain the crack point cloud data set under the two-dimensional coordinate system;

Obtain the image information and spectral information of the target rock mass through spectral scanning;

Mineral distribution features are extracted based on spectral information, and fissure image features are extracted based on image information; based on the mineral distribution features and fissure image features, the boundary coordinates of the filled fissures and the filling mineral information are obtained by using the intelligent identification model of the filled fissure boundary;

The boundary coordinates of the filled fissures are checked and merged with the fissure point cloud data sets in the two-dimensional coordinate system to achieve comprehensive and accurate identification of rock mass fissures.

As a further solution, carry out dimension transformation to the coordinates of the extracted discontinuous point set, the specific process is: for the three-dimensional (x, y, z) coordinates of the extracted discontinuous point set, manually delete the data in the z direction, and convert it into Fracture point cloud dataset in 2D coordinate system.

As a further solution, mineral distribution features are extracted based on spectral information, and the specific process is as follows:

Obtain the spectral curve of each pixel in the image, and determine whether the two pixel points belong to the same type of rock mass based on whether the spectral angle of two adjacent spectral curves is smaller than the set threshold, and match the spectral curve with the spectral database. Determine the specific mineral type corresponding to the spectral curve; thereby determine the mineral type and mineral distribution corresponding to each pixel point, and perform mineral mapping; where the spectral curve and the matching mineral molecular formula are pre-stored in the spectral database.

As a further solution, based on the mineral distribution characteristics and fracture image features, using the intelligent identification model of the filling fracture boundary, the boundary coordinates of the filling fracture and the filling mineral information are obtained. The specific process is: input the fracture image obtained by processing the imaging spectral image cube Features and mineral distribution characteristics, match the crack coordinates at the same position of the target rock mass with the mineral mapping area, extract the overlapping area boundaries of the two areas, regard it as the filling crack area, and finally realize the filling of cracks in combination with mineral mapping Output of boundary coordinates and mineral composition.

As a further solution, the intelligent identification model for filling the fracture boundary adopts a set neural network model, and the identification process of the intelligent identification model for filling the fracture boundary specifically includes:

According to the imaging spectral image cube in the existing database, the image features and spectral features are extracted, and the set ratio is divided into training set and test set, and the training set is input into the neural network model for training, and the trained set is used for training. The neural network model is optimized and verified to obtain the optimal neural network model; the image characteristics and spectral characteristics of the target rock mass that need to be identified for filling cracks are input into the trained neural network model to obtain the identification results of filling cracks and their boundary coordinates , combined with the mineral composition, the boundary coordinates of the filled fractures and the information of the filled minerals are obtained.

As a further solution, check and merge the boundary coordinates of the filled cracks with the crack point cloud data set in the two-dimensional coordinate system, specifically:

Register the coordinates of the crack point cloud data in the two-dimensional coordinate system with the boundary coordinates of the crack filling, unify the two coordinates into the same reference coordinate system, and replace the boundary coordinates of the crack filling with the two-dimensional coordinate system of the corresponding position The coordinates of the fracture point cloud data below.

As a further solution, an identification result of rock mass fissures is obtained, specifically including: the location of the fissures, the information of filling minerals, and the size of the fissures.

In other embodiments, the following technical solutions are adopted:

An intelligent identification system for rock mass fissures that combines 3D scanning and image spectrum, including:

The point cloud data acquisition module is used to obtain the three-dimensional point cloud data of the target rock mass through three-dimensional scanning;

The fracture point cloud data preprocessing module is used to process the three-dimensional point cloud data, extract the coordinates of the discontinuous point set, perform dimension transformation on the extracted discontinuous point set coordinates, and obtain the fissure point cloud data in the two-dimensional coordinate system set;

The image spectrum data acquisition module is used to acquire the image information and spectral information of the target rock mass through spectral scanning;

The fissure filling data acquisition module is used to extract mineral distribution features based on spectral information, and to extract fissure image features based on image information; based on the mineral distribution characteristics and fissure image features, the boundary coordinates of the fissure filling and the filling mineral information;

The data fusion module is used to check and merge the boundary coordinates of the filled cracks with the crack point cloud data sets in the two-dimensional coordinate system to obtain the identification results of rock mass cracks.

As a further solution, the data fusion module registers the coordinates of the crack point cloud data in the two-dimensional coordinate system with the boundary coordinates of the crack filling, and unifies the two coordinates into the same reference coordinate system, and the boundary coordinates of the crack filling The coordinates replace the coordinates of the fracture point cloud data in the two-dimensional coordinate system of the corresponding position.

In other embodiments, the following technical solutions are adopted:

An intelligent identification device for rock mass fissures that integrates three-dimensional scanning and image spectra, used to realize the above-mentioned intelligent identification method for rock mass fissures that fuses three-dimensional scanning and image spectra, is characterized in that the system includes:

working platform;

The electromagnetic wave transceiver device arranged on the working platform, the electromagnetic wave transceiver device can emit electromagnetic waves to the side rock wall, and the distance between the electromagnetic wave transceiver device and the side rock wall can be obtained by using the round-trip propagation time of the electromagnetic wave, so as to control the working platform according to the The set route moves automatically;

Relative to the beams arranged on the working platform, the beams are respectively connected to the working platform through liftable devices; the three-dimensional scanning device and the spectral scanning device are respectively mounted on different beams, and can rotate and translate along the beams; the working platform can Drive the three-dimensional scanning device and the spectral scanning device to automatically move to the set target position for data collection;

The control unit is used to receive the data collected by the three-dimensional scanning device and the spectral scanning device, obtain a fracture point cloud data set and a boundary coordinate data set for filling the fracture based on the data, and merge and process the two data sets to obtain the target rock All fissures of the body.

As a further solution, it also includes: a micro-positioning camera arranged on the working platform, the micro-positioning camera accurately locates the identification range of the rock mass to be measured by taking pictures of the feature points of the target rock mass from multiple angles, and transmits the positioning information to the three-dimensional scanning The device realizes the area locking of the target area to be measured.

Compared with prior art, the beneficial effect of the present invention is:

(1) The present invention proposes to use image spectrum technology to accurately identify the filled cracks, based on the existing lithology spectral information database (the database consists of imaging spectral image cubes and spectral images of various minerals and corresponding spectral numbers and molecular formulas, combined with cracks Image information features), combined with the neural network model to train an intelligent identification model for filling cracks, realize the intelligent extraction output of the boundary coordinate set of filling cracks and filling parameters, and finally realize the accurate identification and parameters of filling cracks (parameters include crack shape, Efficient extraction of size, location, mineral composition of fillers, etc.).

(2) The present invention combines three-dimensional laser scanning technology to identify target rock mass fissures, and the two types of crack identification results complement and check each other to realize comprehensive and accurate identification of target rock mass fissures. The method of the invention can effectively make up for the defects that three-dimensional laser scanning cannot realize accurate identification of filling cracks and filling materials cannot be extracted, improves data quality, and greatly improves the accuracy of crack identification.

(3) The present invention designs an intelligent identification system for rock mass fissures that combines 3D scanning and image spectrum. The location can remotely control the crack identification process, realize fully intelligent accurate identification and efficient processing of crack identification, and greatly improve the work efficiency of rock mass crack identification.

Other features and advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.

Description of drawings

Fig. 1 is the flow chart of the intelligent recognition method for rock mass fissures of fusion of three-dimensional scanning and image spectrum in the embodiment of the present invention;

Fig. 2 is the flow chart of the rock mass fissure intelligent recognition system that fuses three-dimensional scanning and image spectrum in the embodiment of the present invention;

Fig. 3 is a schematic structural diagram of an intelligent identification device for rock mass fissures fused with three-dimensional scanning and image spectrum in an embodiment of the present invention;

Fig. 4 is a schematic diagram of the height adjustment of the rock mass fissure intelligent identification device in the embodiment of the present invention;

Among them, 1. 3D laser scanner; 2. Spectral imager; 3. Beam; 4. Mechanical arm; 5. Hydraulic column; 6. Working platform; 7. Intelligent operation display screen; 8. Light source; ; 10. Distribution box; 11. Handle; 12. Fill light; 13 Electromagnetic wave transceiver.

Detailed ways

It should be pointed out that the following detailed description is exemplary and intended to provide further explanation to the present application. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

It should be noted that the terminology used here is only for describing specific implementations, and is not intended to limit the exemplary implementations according to the present application. As used herein, unless the context clearly dictates otherwise, the singular is intended to include the plural, and it should also be understood that when the terms "comprising" and/or "comprising" are used in this specification, they mean There are features, steps, operations, means, components and/or combinations thereof.

Embodiment one

In one or more embodiments, a method for intelligent identification of rock mass fissures that combines 3D scanning and image spectrum is disclosed, and the fissures are divided into two types: filled fissures and non-filled fissures, and three-dimensional scanning technology and data processing model are used to identify Accurate extraction of non-filled fissures, fuzzy extraction of filled fissures, image spectrum technology combined with image features and mineral distribution characteristics, combined with neural network to accurately identify filled fissures, both perform their duties, and the results complement each other. Check, and finally realize the complete, intelligent and efficient identification of the full coverage of cracks.

In conjunction with Figure 1, the method of this embodiment specifically includes the following processes:

Step (1): Obtain the 3D point cloud data of the target rock mass through 3D scanning;

Step (2): Perform noise filtering on the 3D point cloud data, and then extract the coordinates of the discontinuous point set, perform dimension transformation on the coordinates of the extracted discontinuous point set, and perform three-dimensional (x, y, z) coordinates, manually delete the data in the depth (z) direction, and obtain the fracture point cloud data set in the two-dimensional coordinate system, which includes the non-filled fracture point cloud data set and the filled fracture point cloud data set, wherein, There may be deviations in the identification results of the point cloud data set for filling fractures. Therefore, by merging and merging with the coordinates of the two-dimensional filling fracture data set obtained by the intelligent identification model of the filling fracture boundary, accurate and comprehensive non-filling and filling fracture point cloud data can be obtained. set.

It should be noted that in the process of 3D point cloud data acquisition, the surface of the object is measured by a 3D scanning instrument, and the data obtained generally have many geometric discontinuities, which are usually in the form of discrete points, called discontinuities. continuous point.

In this embodiment, the processing of the three-dimensional point cloud data specifically includes: point cloud intensity correction, point cloud data registration under different viewing angles, and non-target point cloud filtering processes;

Among them, the point cloud intensity correction is specifically to correct the point cloud intensity through the preset point cloud intensity check model, select the most suitable laser incident angle, the optimal laser scanner position, etc.;

The point cloud data registration under different viewing angles is specifically to use the point cloud data automatic splicing method based on reflection value images to register the point cloud data acquired by multiple scans.

Step (3): Obtain image information and spectral information of the target rock mass through spectral scanning;

In this embodiment, the obtained spectral information is preprocessed by variance normalization, which is used to eliminate the spectral error between samples caused by different solid particle sizes, scattering or measurement optical path. The variance normalization method is as follows:

Among them, I _snv is the spectral data obtained by spectral scanning, the spectral data obtained by the first scanning is set as a sample, and its variance E _α and spectral variance σ are calculated.

Step (4): Mineral distribution features are extracted based on spectral information, and fissure image features are extracted based on image information; based on the mineral distribution features and fissure image features, the boundary coordinates of the filled fissures and the filling mineral information are obtained by using the intelligent recognition model of the filled fissure boundary ;

In this embodiment, mineral distribution features are extracted based on spectral information, and the specific process is as follows:

Obtain the spectral curve of each pixel in the image, and determine whether the two pixel points belong to the same type of rock mass based on whether the spectral angle of two adjacent spectral curves is smaller than the set threshold, and match the spectral curve with the spectral database. Determine the specific mineral type corresponding to the spectral curve; thus determine the mineral type and mineral distribution corresponding to each pixel point, and perform mineral mapping on this basis; wherein, the spectral curve and its matching Mineral formula.

Perform crack detection, filtering, point distribution, boundary line fitting, connection and correction on the image information containing cracked rock mass to obtain the crack image features.

Then, using the extracted image features and spectral features, and using the trained intelligent recognition model of the filled fracture boundary, the extraction of the boundary coordinates of the filled fractures and the information output of the filled minerals are realized.

In this embodiment, the intelligent identification model for filling the crack boundary can be realized by selecting an existing neural network model, such as: a deep neural network model, and the specific processing process of the input data for the intelligent identification model for filling the crack boundary is as follows:

Input the fissure image features and mineral distribution characteristics obtained from the processed imaging spectral image cube, match the fissure coordinates at the same position of the target rock mass with the mineral mapping area, extract the overlapping area boundaries of the two areas, and regard them as filling fissures area, combined with mineral mapping to finally realize the output of the boundary coordinates and mineral composition of filled fractures.

In this embodiment, the training process for the neural network model is as follows:

According to the imaging spectral image cube in the existing database, the above-mentioned image features and spectral features are extracted, and it is divided into a training set and a test set according to a ratio of 3:1, and the training set is input into the neural network model for training. Set-pair the trained neural network model optimization verification to obtain the optimal neural network model; input the image characteristics and spectral characteristics of the target rock mass that needs to be identified for filling fissures into the trained neural network model to obtain the results of fissure filling identification And its boundary coordinate set, combined with mineral composition, completes the output of filling cracks.

Step (5): Merge the crack point cloud data set in the two-dimensional coordinate system and the boundary coordinates of the filled cracks to obtain all the crack parameter data of the target rock mass; specifically include: crack shape, size, position and filler minerals ingredient data.

Among them, the process of merging is specifically as follows: register the two-dimensional discontinuous point set obtained through dimension transformation with the boundary coordinates of the crack filling, so that the two coordinate systems are placed in the same reference coordinate system, and the result obtained by combining the two techniques Based on the image information, it is determined that there are two fracture coordinate datasets in the same position, and the fracture boundary data will be filled to replace the original fracture coordinate set, and finally all the fracture data of the target rock mass will be determined.

The merged fracture boundary data is processed twice, and the boundary line fitting, connection and correction are performed sequentially.

Combined with the distribution of fillers for filling fissures, the image spectrum data and fissure data are input into the sketch system to realize the drawing of rock mass fissure sketches.

Embodiment two

In one or more embodiments, an intelligent identification system for rock mass fissures that integrates three-dimensional scanning and image spectrum is disclosed; in combination with Figure 2, the system of this embodiment specifically includes:

The area delineation module is used to delineate the target area to be measured, reduce redundant data acquisition, and reduce the difficulty of later data processing.

The point cloud data acquisition module is used to obtain the three-dimensional point cloud data of the target rock mass through three-dimensional scanning;

The fissure point cloud data preprocessing module is used to perform noise filtering processing on the three-dimensional point cloud data, extract the coordinates of the discontinuous point set, perform dimension transformation on the extracted discontinuous point set coordinates, and obtain the fissure point under the two-dimensional coordinate system cloud dataset;

The image spectrum data acquisition module is used to acquire the image information and spectral information of the target rock mass through spectral scanning;

The filling crack intelligent extraction module is used for extracting mineral distribution features based on spectral information, and extracting crack image features based on image information; based on the mineral distribution features and crack image features, using the filling crack boundary intelligent recognition model to obtain the boundary coordinates of the filling crack and filling mineral information;

The data fusion module is used for checking and merging the crack point cloud data set and the boundary coordinates of filling cracks to obtain the recognition result of rock mass cracks.

It should be noted that the specific implementation methods of the above modules have been described in Embodiment 1, and will not be described in detail here.

Embodiment three

In one or more embodiments, an intelligent identification system for rock mass fissures that integrates 3D scanning and image spectra is disclosed. method;

In conjunction with Fig. 3, the system of this embodiment specifically includes:

The two sides of the platform are respectively provided with beams 3, and the two beams 3 are respectively connected to the working platform 6 through liftable devices; the three-dimensional scanning device and the spectrum scanning device are respectively mounted on different beams, and the beams are provided with transmission devices and rotating shafts. The transmission device and the rotating shaft on the beam can realize the horizontal displacement and rotation of the three-dimensional scanning device or the spectral scanning device.

In this embodiment, the liftable device adopts a hydraulic column 5, which can be adjusted freely. As shown in Figure 4, it is ensured that the devices on the two beams do not block each other. The hydraulic column is also equipped with a mechanical arm 4, which can realize the adjustment of equipment remote control.

The three-dimensional scanning device is mainly composed of a three-dimensional laser scanner 1. The spectrum scanning device is mainly composed of a light source and a spectral imager 2. The light source 8 is a halogen lamp, and the brightness and angle of the light source are adjustable.

The working platform is also equipped with a micro-positioning camera 9, a supplementary light 12, an intelligent operation display screen 7 and other test equipment and objects on the stage, and a micro-positioning camera 9, which can identify the rock mass to be tested by taking pictures from multiple angles and locating the feature points of the target rock mass. Accurate positioning within the range, and the positioning information is synchronized to the 3D laser scanner through the sensor, so as to realize the area locking of the target area to be measured and reduce the difficulty of later data processing. The supplementary light 12 adopts a high-brightness LED light source to supplement light in environments with insufficient light such as tunnels, so as to ensure the smooth progress of image data collection and micro-camera positioning.

The working platform is equipped with cushioning and wear-resistant high-performance tires to ensure the smooth movement of the device; the working platform can drive the three-dimensional scanning device and the spectral scanning device to move to the set target position for data collection.

A plurality of electromagnetic wave transceivers are also arranged on the working platform. The electromagnetic wave transceiver 13 emits electromagnetic waves through a built-in pulse generator, and utilizes the number of pulses at the time interval of the pulse to and fro on the survey line to obtain the distance from the surrounding rock walls, thereby enabling Control the working platform to drive automatically on the established route. The device also includes a power distribution box 10, the purpose of which is to provide energy for the entire device and reasonably configure the electric energy.

The device also includes a handle 11 for short-distance device transportation and manual position adjustment.

The system of this embodiment also includes: a control unit, which is used to receive the data collected by the three-dimensional scanning device and the spectral scanning device, and obtain the non-filled fracture point cloud data set and the boundary coordinate data set of the filled fracture based on these data respectively. For the two data The sets are merged to obtain all the fractures of the target rock mass.

The specific working process of the control unit is consistent with the method disclosed in Embodiment 1, and the control unit mainly includes the following components:

(1) Data collection and processing system: It consists of data collection module and data processing module. The data collection module is responsible for receiving images, spectra, point clouds and other data obtained by the fracture scanning integrated system, and classifying and transferring these data; the data processing module is composed of a spectral data processing unit and a point cloud data processing unit composition. The spectral data processing unit first preprocesses the spectral information and extracts image and mineral features, and then inputs the extracted features into the intelligent identification model for filling cracks. The intelligent identification model for filling cracks, through the extracted image features and spectral features, uses the neural network to realize the extraction of the boundary coordinate set of filling cracks and the output of filling parameters. The point cloud data processing unit inputs the point cloud data and image data into the point cloud data intelligent processing model to process the point cloud data. The point cloud data intelligent processing model can sequentially perform point cloud data including point cloud intensity correction, point cloud data registration under different viewing angles, non-target point cloud filtering, and coordinate extraction of discontinuous point sets. On this basis, the obtained coordinates The set performs depth dimension elimination to realize the dimension transformation of coordinates.

(2) Data merging and storage system: It is composed of a data integration module and a storage module; the data integration module merges the received data processing filled crack boundary coordinate data set and the crack point cloud data after dimension conversion. Fissure numbering, and realize the pairing of filling fissures fissures and fillings. The merging process realizes coordinate registration, coordinate checking, coordinate elimination and coordinate merging of two-dimensional coordinate sets from different sources, and the merged crack boundary coordinate data is input to the rock mass crack output system after secondary processing; the data storage The module is used to store data information and program information, so as to facilitate data migration backup analysis and program upgrade detection and repair.

(3) Rock mass fissure output system: it is composed of a rock mass fissure sketch module and a fissure parameter output module; the rock mass fissure sketch module refers to inputting fissure data into the fissure sketch software to realize the graphical depiction and analysis of rock mass fissures Output, a detailed plot of each fracture. The crack output module is composed of a crack output unit and a liquid crystal display. The crack output unit realizes the matching and output of the crack image and parameter data, which is finally displayed by the liquid crystal display. The control of the system and the query and extraction of crack data can be realized through the display. and other functions. The output parameter information mainly includes: the orientation, shape, size of the fracture, and the mineral composition of the filler filling the fracture;

(4) Intelligent control system: The intelligent control system can remotely control the system, realize the start-stop and attitude adjustment of the spectrometer and 3D laser scanner through the wireless signal transceiver and sensor in the control unit, and realize the control of the robot arm and beam. The free adjustment of the transmission device, the rotating shaft and the light source realizes the control of the electromagnetic wave transceiver device and ensures the precise movement of the device. Realize the optimal scheduling and reasonable allocation of various systems and supporting facilities. It is also possible to realize the remote operation of the mobile terminal of the above process through the combination of the intelligent control system and the remote control platform, as well as the viewing and extraction of crack and parameter information.

Although the specific implementation of the present invention has been described above in conjunction with the accompanying drawings, it does not limit the protection scope of the present invention. Those skilled in the art should understand that on the basis of the technical solution of the present invention, those skilled in the art do not need to pay creative work Various modifications or variations that can be made are still within the protection scope of the present invention.

Claims (10)

1. A method for intelligent identification of rock mass fissures that combines three-dimensional scanning and image spectrum, characterized in that it includes: Obtain 3D point cloud data of the target rock mass through 3D scanning; Processing the three-dimensional point cloud data, extracting the coordinates of the discontinuous point set, and performing dimension transformation on the extracted coordinates of the discontinuous point set, to obtain the crack point cloud data set under the two-dimensional coordinate system; Obtain the image information and spectral information of the target rock mass through spectral scanning; Mineral distribution features are extracted based on spectral information, and fissure image features are extracted based on image information; based on the mineral distribution features and fissure image features, the boundary coordinates of the filled fissures and the filling mineral information are obtained by using the intelligent identification model of the filled fissure boundary; The boundary coordinates of the filled fissures are checked and merged with the fissure point cloud data sets in the two-dimensional coordinate system to obtain the identification results of rock mass fissures. 2. A kind of rock mass fissure intelligent identification method of fusion three-dimensional scanning and image spectrum as claimed in claim 1, it is characterized in that, described three-dimensional point cloud data is processed, after extracting the coordinates of discontinuous point set, extracting The coordinates of the discontinuous point set are dimensionally transformed. The specific process is: for the extracted three-dimensional (x, y, z) coordinates of the discontinuous point set, manually delete the data in the z direction, and convert it into a crack in the two-dimensional coordinate system. point cloud dataset. 3. A kind of rock mass fissure intelligent identification method of fusion three-dimensional scanning and image spectrum as claimed in claim 1, it is characterized in that, extract mineral distribution feature based on spectral information, specific process is: Obtain the spectral curve of each pixel in the image, and determine whether the two pixel points belong to the same type of rock mass based on whether the spectral angle of two adjacent spectral curves is smaller than the set threshold, and match the spectral curve with the spectral database. Determine the specific mineral type corresponding to the spectral curve; thereby determine the mineral type and mineral distribution corresponding to each pixel point, and perform mineral mapping; where the spectral curve and the matching mineral molecular formula are pre-stored in the spectral database. 4. A kind of rock mass fissure intelligent identification method of fusion three-dimensional scanning and image spectrum as claimed in claim 1, it is characterized in that, described filling fissure boundary intelligent recognition model adopts the neural network model of setting, for filling fissure boundary intelligence Identify the identification process of the model, specifically including: According to the imaging spectral image cube in the existing database, the image features and spectral features are extracted, and the set ratio is divided into training set and test set, and the training set is input into the neural network model for training, and the trained set is used for training. The neural network model is optimized and verified to obtain the optimal neural network model; the image characteristics and spectral characteristics of the target rock mass that need to be identified for filling cracks are input into the trained neural network model to obtain the identification results of filling cracks and their boundary coordinates , combined with the mineral composition, the boundary coordinates of the filled fractures and the information of the filled minerals are obtained. 5. A method for intelligent identification of rock mass fissures fused with three-dimensional scanning and image spectrum as claimed in claim 1, characterized in that the boundary coordinates of the filled fissures are checked against the fissure point cloud datasets under the two-dimensional coordinate system Merge, specifically: Register the coordinates of the crack point cloud data in the two-dimensional coordinate system with the boundary coordinates of the crack filling, unify the two coordinates into the same reference coordinate system, and replace the boundary coordinates of the crack filling with the two-dimensional coordinate system of the corresponding position The coordinates of the fracture point cloud data below. 6. A method for intelligent identification of rock mass fissures fused with three-dimensional scanning and image spectrum as claimed in claim 1, characterized in that the identification results of rock mass fissures are obtained, which specifically include: the position of the fissure, the filling mineral information and the information of the fissure. size. 7. An intelligent identification system for rock mass fissures that integrates three-dimensional scanning and image spectrum, characterized in that it includes: The point cloud data acquisition module is used to obtain the three-dimensional point cloud data of the target rock mass through three-dimensional scanning; The fracture point cloud data preprocessing module is used to process the three-dimensional point cloud data, extract the coordinates of the discontinuous point set, perform dimension transformation on the extracted discontinuous point set coordinates, and obtain the fissure point cloud data in the two-dimensional coordinate system set; The image spectrum data acquisition module is used to acquire the image information and spectral information of the target rock mass through spectral scanning; The fissure filling data acquisition module is used to extract mineral distribution features based on spectral information, and to extract fissure image features based on image information; based on the mineral distribution characteristics and fissure image features, the boundary coordinates of the fissure filling and the filling mineral information; The data fusion module is used to check and merge the boundary coordinates of the filled cracks with the crack point cloud data sets in the two-dimensional coordinate system to obtain the identification results of rock mass cracks. 8. A kind of rock mass fissure intelligent recognition system of fusion three-dimensional scanning and image spectrum as claimed in claim 7, it is characterized in that, described data fusion module combines the fissure point cloud data coordinates under the two-dimensional coordinate system and the fissure filling fissure The boundary coordinates are registered, and the two coordinates are unified into the same reference coordinate system, and the boundary coordinates of the filled cracks are replaced with the crack point cloud data coordinates in the two-dimensional coordinate system of the corresponding position. 9. An intelligent identification system for rock mass fissures that fuses three-dimensional scanning and image spectra, for realizing the intelligent identification method for rock mass fissures that fuses three-dimensional scanning and image spectra according to any one of claims 1-7, characterized in that, The system includes: working platform; The electromagnetic wave transceiver device arranged on the working platform, the electromagnetic wave transceiver device can emit electromagnetic waves to the side rock wall, and the distance between the electromagnetic wave transceiver device and the side rock wall can be obtained by using the round-trip propagation time of the electromagnetic wave, so as to control the working platform according to the The set route moves automatically; Relative to the beams arranged on the working platform, the beams are respectively connected to the working platform through liftable devices; the three-dimensional scanning device and the spectral scanning device are respectively mounted on different beams, and can rotate and translate along the beams; the working platform can Drive the three-dimensional scanning device and the spectral scanning device to automatically move to the set target position for data collection; The control unit is used to receive the data collected by the three-dimensional scanning device and the spectral scanning device, obtain a fracture point cloud data set and a boundary coordinate data set for filling the fracture based on the data, and merge and process the two data sets to obtain the target rock All fissures of the body. 10. A rock mass fissure intelligent recognition system that integrates three-dimensional scanning and image spectrum as claimed in claim 9, further comprising: a miniature positioning camera arranged on the working platform, and the miniature positioning camera passes through multi-angle Shoot and locate the feature points of the target rock mass to accurately locate the identification range of the rock mass to be measured, and transmit the positioning information to the 3D scanning device to realize the area locking of the target area to be measured.

Images