CN115713645A - Lithology identification method and system based on spectral imaging technology - Google Patents
Lithology identification method and system based on spectral imaging technology Download PDFInfo
- Publication number
- CN115713645A CN115713645A CN202211282405.XA CN202211282405A CN115713645A CN 115713645 A CN115713645 A CN 115713645A CN 202211282405 A CN202211282405 A CN 202211282405A CN 115713645 A CN115713645 A CN 115713645A
- Authority
- CN
- China
- Prior art keywords
- lithology
- grid
- spectral
- image
- grids
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 56
- 238000000701 chemical imaging Methods 0.000 title claims abstract description 28
- 238000005516 engineering process Methods 0.000 title claims abstract description 19
- 230000003595 spectral effect Effects 0.000 claims abstract description 57
- 238000001228 spectrum Methods 0.000 claims abstract description 48
- 238000012545 processing Methods 0.000 claims abstract description 23
- 238000001514 detection method Methods 0.000 claims abstract description 14
- 238000012937 correction Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 9
- 238000010606 normalization Methods 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 8
- 238000003384 imaging method Methods 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 3
- 238000007689 inspection Methods 0.000 claims 1
- 230000000452 restraining effect Effects 0.000 claims 1
- 239000011435 rock Substances 0.000 description 17
- 230000008901 benefit Effects 0.000 description 8
- 230000002159 abnormal effect Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 238000009499 grossing Methods 0.000 description 6
- 238000010998 test method Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 239000000523 sample Substances 0.000 description 5
- 238000007405 data analysis Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 238000010238 partial least squares regression Methods 0.000 description 3
- 238000012706 support-vector machine Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 229910052500 inorganic mineral Inorganic materials 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 239000011707 mineral Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 229910052736 halogen Inorganic materials 0.000 description 1
- 150000002367 halogens Chemical class 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000004611 spectroscopical analysis Methods 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention belongs to the technical field of lithology identification of engineering geology, and provides a lithology identification method and a lithology identification system based on a spectral imaging technology. Acquiring image spectrum data of a tunnel face of a tunnel, and rasterizing an interested area; carrying out regional trace detection by using image data of a tunnel face, and judging whether a trace exists in a grid; performing second rasterization processing on the grids with the traces; extracting all traceless grids and spectral features and image features of the grids obtained by rasterization again; the spectral features and the image features of the same grid are fused, and then the lithology of each grid is identified through a pre-trained lithology prediction model; and checking and correcting the lithology recognition result of the grid obtained by the second rasterization, and finally obtaining the lithology of all the grids.
Description
Technical Field
The invention belongs to the technical field of engineering geology lithology identification, and particularly relates to a lithology identification method and system based on a spectral imaging technology.
Background
The statements in this section merely provide background information related to the present disclosure and may not constitute prior art.
Lithology identification has been an important and fundamental problem in the fields of geology, resource exploration, tunnel and underground engineering unfavorable geology identification, disaster prevention and reduction and the like. The lithology recognition of the unfavorable geological region is the premise and the basis for geological forecast of the tunnel engineering, and has important guiding significance for optimization of the tunnel engineering design scheme, safety assessment and risk evaluation. The traditional lithology identification method adopts a visual observation method and a slice identification method, which depend too much on manual experience, not only consumes long time and has strong specialization, but also is easily influenced by subjective factors, so that the accuracy is not ideal. Geologists have conducted research on lithology recognition based on images, but as a result, the recognition accuracy is inaccurate due to problems such as high similarity of images caused by similar rock components or textures, fine lithology features are easily lost in the feature extraction process, visible rock features are damaged by weathering or human activities, poor imaging quality caused by shooting conditions or technical differences, and the like. The method for acquiring rock mineral type and content information is various, wherein the spectrum technology is a common means at present, wherein XRF and XRD spectrum tests require contact measurement, sample grinding and the like, a large amount of time is consumed, and by means of single characteristics of rocks, misclassification and misclassification phenomena are easily generated, so that the problems of large lithology identification error, low lithology interpretation precision and the like are caused.
The inventor finds that rocks can be simply classified by means of image information or spectral information of rocks and mines, but the phenomena of 'same objects, different spectrums' and 'same spectrum foreign bodies' exist, the classification precision of the methods is not high, and lithology identification by using spectral information at present mostly depends on contact type or sample grinding processing, so that the method has certain limitation when being used for tunnel sites.
Disclosure of Invention
In order to solve the technical problems in the background art, the invention provides a lithology identification method and system based on a spectral imaging technology, which can identify lithology and spatial distribution conditions on a tunnel face in real time, judge the geological conditions of rock masses in front of a tunnel and provide an important reference basis for mastering the geological conditions of the rock masses in front of the tunnel. The invention utilizes the image spectrum technology and adopts the image spectrum information acquired by a long-distance photographing mode, thereby not only acquiring the image data, but also acquiring the spectrum data of each pixel point on the image, namely a three-dimensional cubic data body.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention provides a lithology identification method based on a spectral imaging technology, which comprises the following steps:
acquiring image spectrum data of a tunnel face, and rasterizing an interested area;
carrying out regional trace detection by using image data of a tunnel face, and judging whether a trace exists in a grid;
performing second rasterization processing on the grids with the traces;
extracting all traceless grids and spectral features and image features of the grids obtained by rasterization again;
the spectral features and the image features of the same grid are fused, and then the lithology of each grid is identified through a pre-trained lithology prediction model;
and checking and correcting the lithology recognition result of the grid obtained by the second rasterization, and finally obtaining the lithology of all the grids.
As an implementation mode, the lithology of all the finally obtained grids is displayed on the image of the tunnel face in a map filling mode.
The technical scheme has the advantages that lithology and spatial distribution conditions on the tunnel face are intuitively recognized in real time, the geological condition of the rock body in front of the tunnel is judged, and an important reference basis is provided for mastering the geological condition of the rock body in front of the tunnel.
And as an implementation mode, constraining the lithology of the grid obtained by the second rasterization by adopting a Sudoku test method and an area trace, and correcting the lithology recognition result of the corresponding grid.
The nine-grid check method is to check the grids of the abnormal recognition results of the small grids divided for the second time by using a nine-grid graph; firstly, the grids with different recognition results in the nine grids are calibrated, and the nine-square grids are established by taking the abnormal grids as the center and are classified as lithology with the largest proportion in the eight adjacent grids.
The technical scheme has the advantages that the error of the lithology identification result of the small grid through which the regional trace passes is corrected, and the accuracy of the lithology identification result of the small grid divided for the second time is improved.
As an embodiment, before fusing the spectral features and the image features of the same grid, the method further includes:
and carrying out normalization processing on the spectral characteristics and the image characteristics of the same grid.
The technical scheme has the advantages that the dimensionality of the feature vector of the image dimensionality is kept consistent with the dimensionality of the feature vector of the spectrum dimensionality, and the accuracy of the lithology identification result of the grid is finally improved.
As an embodiment, after acquiring the image spectrum data of the tunnel face, the method further includes:
and preprocessing the spectral data of the tunnel face.
The spectrum signal acquired by the spectrometer contains useful information required by an experiment, and random noise is brought by the precision of the instrument, so that a plurality of preprocessing methods are available, such as convolution smoothing, area normalization, baseline correction, first-order derivative, standard normal variable transformation, multivariate scattering correction and the like. The most common method for noise removal (SG) convolution smoothing is to spectrally smooth the acquired hyperspectral curve, which both removes noise and preserves spectral contours.
In one embodiment, the lithology prediction model is a classifier.
It should be noted here that the classifier may employ a model such as an extreme learning machine, partial least squares regression, support vector machine, etc.
A second aspect of the invention provides a lithology identification system based on spectral imaging technology, comprising:
the primary rasterization module is used for acquiring image spectrum data of a tunnel face and rasterizing an interested area;
the regional trace detection module is used for carrying out regional trace detection by utilizing the image data of the tunnel face and judging whether a trace exists in the grid;
the secondary rasterization module is used for carrying out secondary rasterization processing on the grids with traces;
the characteristic extraction module is used for extracting all traceless grids and spectral characteristics and image characteristics of the grids obtained by rasterization again;
the lithology recognition module is used for fusing the spectral characteristics and the image characteristics of the same grid and recognizing the lithology of each grid through a pre-trained lithology prediction model;
and the lithology correction module is used for detecting and correcting the lithology recognition result of the grid obtained by the second rasterization, and finally obtaining the lithology of all the grids.
As an embodiment, the lithology identification system based on the spectral imaging technology further includes:
and the recognition result display module is used for displaying the lithology of all the finally obtained grids on the image of the tunnel face in a map filling mode.
As an implementation manner, in the lithology correction module, the lithology of the grid obtained by the second rasterization is constrained by using a squared figure checking method and an area trace, and the lithology recognition result of the corresponding grid is corrected.
As an embodiment, before fusing the spectral features and the image features of the same grid, the lithology identification module further includes:
and carrying out normalization processing on the spectral characteristics and the image characteristics of the same grid.
As an embodiment, in the primary rasterization module, after acquiring the image spectrum data of the tunnel face, the method further includes:
and (4) preprocessing the spectral data of the tunnel face.
As an embodiment, in the lithology identification module, the lithology prediction model is a classifier.
A third aspect of the invention provides a computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method for lithology identification based on spectral imaging techniques as described above.
A fourth aspect of the invention provides a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the method for lithology identification based on spectral imaging as described above when executing the program.
Compared with the prior art, the invention has the beneficial effects that:
(1) Because the data volume of the three-dimensional data acquired by the image spectrum technology is large, the full-waveband spectrum data volume of the sample is large, and the information is mixed, the acquired image spectrum data of the tunnel face is utilized to perform rasterization processing on the region of interest, the full-waveband average value of the pixel point of the spectrum image in each grid is obtained, the image characteristics in the grid and the average spectrum are fused to identify the lithology, the whole acquisition of the face information can be ensured, the model calculation amount can be reduced, and the model robustness and the working efficiency can be improved.
(2) The method fuses image data and spectral data of a tunnel face, extracts all traceless grids through rasterization processing, obtains spectral characteristics and image characteristics of the grids through rasterization again, and identifies the lithology of each grid through characteristic fusion; and finally, the lithology of all the grids is obtained by detecting and correcting the lithology recognition result of the grid obtained by the second rasterization, so that the lithology of the tunnel face is recognized on the tunnel site, the lithology and the spatial distribution condition on the tunnel face can be recognized in real time only by collecting information of the tunnel face through a hyperspectral imaging technology and carrying out data processing and analysis, the geological condition of the rock mass in front of the tunnel is judged, and an important reference basis is provided for mastering the geological condition of the rock mass in front of the tunnel.
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.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are included to illustrate an exemplary embodiment of the invention and not to limit the invention.
FIG. 1 is a flow chart of a lithology identification method based on spectral imaging technology according to an embodiment of the invention;
FIG. 2 is a diagram illustrating the meshing of a tunnel face according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a lithology identification system based on a spectral imaging technology according to an embodiment of the present invention.
Wherein, 1, standard white board; 2. a spectral imager; 3. an imaging lens; 4. a battery; 5. a retractable table; 6. a telescopic rod; 7. a data analysis platform; 8. a light source; 9. a holder.
Detailed Description
The invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. 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 invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an", and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
Example one
Referring to fig. 1, the present embodiment provides a lithology identification method based on a spectral imaging technology, which specifically includes the following steps:
step 1: acquiring image spectrum data of the tunnel face of the tunnel, and rasterizing the region of interest.
In the implementation process, the image spectrum data of the tunnel face can be realized by adopting a data acquisition and storage system, and the structure of the data acquisition and storage system is shown in fig. 3. The data acquisition and storage system comprises a hyperspectral imaging camera, a holder 9, a standard white board 1, a battery 4 and a light source 8. The hyperspectral imaging camera comprises a spectral imager 2 and an imaging lens 3. The cloud platform 9 is fixed on the robot for installing and fixing the hyperspectral imaging camera, and the cloud platform 9 can rotate at will, so that the operation of an actual field is facilitated. The bottom of the cloud deck 9 is also provided with a telescopic table 5. The light source 8 is fixed on the robot and located around the hyperspectral imaging camera, the type of the light source is a halogen lamp, and the height and the angle of the light source 8 are adjustable. The standard white board 1 is mainly used for optical calibration measurement for spectral analysis.
The embodiment utilizes the high spectrum appearance of formation of image to have the advantage of "map unification", carries out face rock scanning not only can acquire the rock image in step, can also acquire the mineral spectral information on the whole rock face, through the analysis and the integration of high spectrum three-dimensional data, has compensatied the not enough of single data, can high efficiency acquire the spatial distribution information of face lithology.
In order to reduce the influence of uneven illumination and camera dark current, black and white correction needs to be carried out on the acquired original hyperspectral data, under the same environment, a standard white board is used for acquiring white references, a lens cover is covered after a light source is closed, dark references are acquired, and the corrected hyperspectral image is obtained.
According to the distance between the robot and the tunnel face, the angle of the cradle head and the height of the imaging spectrometer, the details of the main target can be clearly seen and distinguished under the spatial resolution scale without being interfered by other factors.
The standard white board 1 of the imaging spectrometer is adjusted and calibrated through the telescopic rod 5 connected with the standard white board 1, the position of the standard white board 1 is moved to the front of the camera, and the camera probe is vertically aligned with the standard white board for calibration.
The data acquisition and storage system is connected with the data analysis platform 7, and the image spectrum acquired by the data acquisition and storage system is transmitted to the data analysis platform 7 for corresponding rasterization and other processing.
Because the image data collected by the image spectrum system has numerous pixels, each pixel can extract a complete high-resolution spectrum curve, the data volume is large, and the increase of wave bands can lead to the increase of information redundancy and data processing complexity. If the whole spectrum data of the whole palm surface is subjected to equalization processing, the whole identification precision is reduced, the palm surface is subjected to grid division in the embodiment, namely, the spectrum values of N pixel points in a small grid under the same wave band are averaged, and finally, a complete spectrum curve can be extracted from each grid.
The division of the grids is divided according to the actual tunnel engineering environment (depending on the shooting range of the tunnel face and the area size of the tunnel face), and under the condition of correct other settings, the more dense the grids are, the higher the solving precision is; however, the denser the grid, the more the number, the longer the solution time; therefore, a trade-off between the solution efficiency and the solution accuracy is required. For example: when the mesh division is debugged for the first time, the area of a single mesh may be set to about 0.1 square meter, and if the area of one palm surface is 40 square meters, the palm surface is equally divided into 400 meshes.
As a specific implementation manner, after acquiring the image spectrum data of the tunnel face, the method further includes:
and preprocessing the spectral data of the tunnel face.
The spectrum signal acquired by the spectrometer contains useful information required by an experiment, and random noise is brought by the precision of the instrument, so that a plurality of preprocessing methods are available, such as convolution smoothing, area normalization, baseline correction, first-order derivative, standard normal variable transformation, multivariate scattering correction and the like. The most common method for noise removal (SG) convolution smoothing is to spectrally smooth the acquired hyperspectral curve, which both removes noise and preserves spectral contours.
The effect of preprocessing the spectral data of the tunnel face comprises functions of black and white correction, spectrum smoothing and the like. In order to overcome the influence of nonuniformity of light source intensity under each wave band and the influence of dark current in the hyperspectral image acquisition process, the acquired hyperspectral curve is subjected to spectrum smoothing, so that the noise is eliminated, and the spectrum profile is reserved.
Step 2: and (4) carrying out regional trace detection by using image data of the tunnel face, and judging whether traces exist in the grids.
In the specific implementation process, the image data of the tunnel face is used for performing area trace detection, and the detection result includes two cases, namely no trace in the grid and trace in the grid, as shown in fig. 2.
When the detection result is that no trace is in the grid, the grid does not need to be rasterized again.
And 3, step 3: the second rasterization process is performed on the traced mesh.
And 4, step 4: and extracting the spectral characteristics and the image characteristics of all the traceless grids and grids obtained by rasterizing again.
Specifically, the spectral characteristics include spectral value equalization, characteristic band extraction, and spectral band number. And (4) spectrum equalization, averaging the spectrum values of all pixel points in the small grid in the same wave band, and finally extracting a spectrum curve from one grid. And extracting the characteristic wave band, and screening and extracting the characteristic wavelength.
Specifically, the spectrum is equalized, the mean value of the spectrum values of the same waveband of all the pixels in the small grid is obtained, that is, the hyperspectral image of each grid is a three-dimensional tensor of nxnxnxp, wherein nxn is a spatial dimension, and P is a spectral dimension, the three-dimensional tensor is expanded along a third dimension to obtain (nxn) xp, the (nxn) xp is obtained by representing N pixels corresponding to each waveband, then the pixels are averaged to obtain a 1 xp vector, and m can also be understood as being the mean value of the spectrum values of the same waveband of all the pixels in the small grid
The number of pixel points, iij is the spectral value of the ith pixel in the jth waveband, I j Is the j wave in a grid
Average spectral values in sections I j (j =1, 2 \8230;) constitutes a complete spectral curve, and finally a grid extracts a spectral curve.
In the extraction of the characteristic wave band, because the spectral data has a plurality of variables and may have redundant information, if each spectral value is substituted into the model analysis, the accuracy of the identification prediction is affected, the computation amount of the system processing analysis is increased, and the computation speed of the model is reduced, so that the characteristic wavelength needs to be screened and extracted. In the actual application analysis of a hyperspectral image, spectroscopy requires that an obvious peak or a trough exists in a spectral curve of a selected characteristic wave band, namely, a target to be detected can absorb or reflect light of the characteristic wave band, principal component analysis is carried out on original data, redundant information among wave bands is removed, and multiband image information is compressed to a few more effective conversion wave bands than the original wave bands. The characteristic with the maximum correlation with the principal component in the original characteristics can be obtained according to the factor load between the principal component and the original characteristics, and therefore characteristic wavelength extraction is achieved. Finally, characteristic wave bands b1, b2 and b3 are selected for forming a characteristic space of 8230, and an average spectrum curve of the sample under the characteristic space is obtained.
The image features are very rich, the features are closely related, and information fusion among various features is an important factor for identification. Texture features can be calculated from the image by adopting methods such as gray level co-occurrence matrix and the like at four angles of 0 degrees, 45 degrees, 90 degrees and 135 degrees respectively, and characteristic values such as energy, entropy, moment of inertia, correlation and the like of the image spectrum image are extracted as the texture features.
And 5: and fusing the spectral characteristics and the image characteristics of the same grid, and identifying the lithology of each grid through a pre-trained lithology prediction model.
Specifically, before fusing the spectral features and the image features of the same grid, the method further includes:
and carrying out normalization processing on the spectral characteristics and the image characteristics of the same grid.
The technical scheme has the advantages that the dimensionality of the feature vector of the image dimensionality is kept consistent with the dimensionality of the feature vector of the spectrum dimensionality, and the accuracy of the lithology identification result of the grid is finally improved.
And fusing the spectral features and the image features subjected to normalization processing, and sending the fused features into a lithology prediction model trained in advance for lithology classification.
Wherein the lithology prediction model is a classifier.
It should be noted here that the classifier may employ a model such as an extreme learning machine, partial least squares regression, support vector machine, etc.
And 6: and checking and correcting the lithology recognition result of the grid obtained by the second rasterization, and finally obtaining the lithology of all the grids.
Through the model established in the early stage, the model can carry out feature description on target data through a certain rule, and then carries out qualitative classification or quantitative prediction on the data according to the features, so as to identify the lithology of each small grid. Meanwhile, a Sudoku test method is adopted, and the regional trace detection results processed by image information are jointly determined, so that the influence of local region identification on the overall identification result can be reduced.
In some embodiments, the lithology of the grid obtained by the second rasterization is constrained by a Sudoku test method and an area trace, and the lithology recognition result of the corresponding grid is corrected.
The nine-grid check method is to check the grids of the abnormal recognition results of the small grids divided for the second time by using a nine-grid graph; firstly, the grids with different recognition results in the nine grids are calibrated, and the nine-square grids are established by taking the abnormal grids as the center and are classified as lithology with the largest proportion in the eight adjacent grids.
Generally, the lithology identification result of the small grid penetrated by the area trace has errors because two lithologies may exist in the grid, and the grid penetrated by the area trace is continuously divided again, and the area of the division is about 0.01m 2 (namely, the spectral curve is divided into 9 grids), and the extracted spectral curve is fused with the image characteristics in the small grids by adopting the same method to identify the lithology. And correcting the recognition result of the small grids divided for the second time by using a Sudoku test method.
The nine-grid test method is to test the grids of the abnormal recognition results of the small grids divided for the second time by using a nine-grid graph; firstly, the grids with different recognition results in the nine grids are calibrated, and the nine-square grids are established by taking the abnormal grids as the center and are classified as lithology with the largest proportion in the eight adjacent grids.
The technical scheme has the advantages that the error of the lithology identification result of the small grid through which the regional trace passes is corrected, and the accuracy of the lithology identification result of the small grid divided for the second time is improved.
In one or more embodiments, the lithology of all the finally obtained grids can be displayed on the image of the tunnel face in a map filling mode.
The technical scheme has the advantages that lithology and spatial distribution conditions on the tunnel face are intuitively recognized in real time, the geological condition of the rock mass in front of the tunnel is judged, and an important reference basis is provided for mastering the geological condition of the rock mass in front of the tunnel.
Example two
The embodiment provides a lithology identification system based on a spectral imaging technology, which comprises a primary rasterization module, an area trace detection module, a secondary rasterization module, a feature extraction module, a lithology identification module and a lithology correction module.
(1) And the primary rasterizing module is used for acquiring image spectrum data of the tunnel face of the tunnel and rasterizing the region of interest.
Specifically, in the primary rasterizing module, after acquiring image spectrum data of a tunnel face, the method further includes:
and preprocessing the spectral data of the tunnel face.
(2) And the area trace detection module is used for detecting the area traces by using the image data of the tunnel face and judging whether traces exist in the grids.
(3) And the secondary rasterizing module is used for performing secondary rasterizing processing on the grids with the traces.
(4) And the characteristic extraction module is used for extracting all the traceless grids and the spectral characteristics and the image characteristics of the grids obtained by rasterizing again.
(5) And the lithology recognition module is used for fusing the spectral characteristics and the image characteristics of the same grid and recognizing the lithology of each grid through a pre-trained lithology prediction model.
Specifically, in the lithology identification module, before fusing the spectral features and the image features of the same grid, the method further includes:
and carrying out normalization processing on the spectral characteristics and the image characteristics of the same grid.
Specifically, in the lithology identification module, the lithology prediction model is a classifier.
For example: the classifier can adopt models such as an extreme learning machine, partial least squares regression, support vector machine and the like.
(6) And the lithology correction module is used for checking and correcting the lithology recognition result of the grid obtained by the second rasterization, and finally obtaining the lithology of all the grids.
Specifically, in the lithology correction module, the lithology of the grid obtained by the second rasterization is constrained by adopting a Sudoku test method and an area trace, and the lithology recognition result of the corresponding grid is corrected.
In one or more embodiments, the lithology identification system based on spectral imaging technology further comprises:
and the recognition result display module is used for displaying the lithology of all the finally obtained grids on the image of the tunnel face in a map filling mode.
It should be noted that, each module in the present embodiment corresponds to each step in the first embodiment one to one, and the specific implementation process is the same, which is not described herein again.
EXAMPLE III
The present embodiment provides a computer-readable storage medium having stored thereon a computer program which, when being executed by a processor, carries out the steps of the method for lithology identification based on spectral imaging technique as described above.
Example four
The present embodiment provides a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps in the method for lithology identification based on spectral imaging technique as described above when executing the program.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A lithology identification method based on a spectral imaging technology is characterized by comprising the following steps:
acquiring image spectrum data of a tunnel face, and rasterizing an interested area;
carrying out regional trace detection by using image data of a tunnel face, and judging whether a trace exists in a grid;
performing second rasterization processing on the grids with the traces;
extracting all traceless grids and obtaining the spectral characteristics and the image characteristics of the grids through re-rasterization;
the spectral features and the image features of the same grid are fused, and then the lithology of each grid is identified through a pre-trained lithology prediction model;
and checking and correcting the lithology recognition result of the grid obtained by the second rasterization, and finally obtaining the lithology of all the grids.
2. The lithology-based imaging technique of claim 1, wherein the lithology of all the resulting meshes is shown on the image of the tunnel face in a map-filling manner.
3. The lithology-based imaging technique of claim 1, wherein the lithology of the second rasterized mesh is constrained using a Sudoku test and regional traces to correct the lithology recognition results of the corresponding mesh.
4. The lithology-based identification method of claim 1, wherein before fusing the spectral features and the image features of the same grid, further comprising:
and carrying out normalization processing on the spectral characteristics and the image characteristics of the same grid.
5. The lithology-based imaging technique of claim 1, wherein after acquiring the image spectral data of the tunnel face, further comprising:
and (4) preprocessing the spectral data of the tunnel face.
6. The method of spectral imaging technique-based lithology identification of claim 1, wherein the lithology prediction model is a classifier.
7. A lithology identification system based on spectral imaging technology, comprising:
the primary rasterization module is used for acquiring image spectrum data of the tunnel face of the tunnel and rasterizing the region of interest;
the regional trace detection module is used for carrying out regional trace detection by utilizing the image data of the tunnel face and judging whether a trace exists in the grid;
the secondary rasterization module is used for carrying out secondary rasterization processing on the grids with traces;
the characteristic extraction module is used for extracting all traceless grids and spectral characteristics and image characteristics of the grids obtained by rasterization again;
the lithology recognition module is used for fusing the spectral characteristics and the image characteristics of the same grid and recognizing the lithology of each grid through a pre-trained lithology prediction model;
and the lithology correction module is used for detecting and correcting the lithology recognition result of the grid obtained by the second rasterization, and finally obtaining the lithology of all the grids.
8. The spectral imaging technique based lithology identification system of claim 7, further comprising:
the recognition result display module is used for displaying the lithology of all the finally obtained grids on the image of the tunnel face in a map filling mode;
or
In the lithology correction module, restraining the lithology of the grid obtained by the second rasterization by adopting a Sudoku inspection method and an area trace, and correcting the lithology recognition result of the corresponding grid;
or
In the lithology identification module, before fusing the spectral features and the image features of the same grid, the method further includes:
carrying out normalization processing on the spectral characteristics and the image characteristics of the same grid;
or
In the primary rasterizing module, after acquiring image spectrum data of a tunnel face, the method further includes:
preprocessing the spectral data of the tunnel face;
or
In the lithology identification module, the lithology prediction model is a classifier.
9. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method for lithology identification based on spectral imaging techniques according to any one of claims 1 to 6.
10. A computer arrangement comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor, when executing the program, carries out the steps of the method for lithology identification based on spectral imaging technique according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211282405.XA CN115713645B (en) | 2022-10-19 | 2022-10-19 | Lithology recognition method and system based on spectral imaging technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211282405.XA CN115713645B (en) | 2022-10-19 | 2022-10-19 | Lithology recognition method and system based on spectral imaging technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115713645A true CN115713645A (en) | 2023-02-24 |
| CN115713645B CN115713645B (en) | 2024-06-28 |
Family
ID=85229933
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211282405.XA Active CN115713645B (en) | 2022-10-19 | 2022-10-19 | Lithology recognition method and system based on spectral imaging technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115713645B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116883852A (en) * | 2023-08-29 | 2023-10-13 | 北京建工环境修复股份有限公司 | Core data acquisition method and system based on hyperspectrum |
| CN118503905A (en) * | 2024-04-18 | 2024-08-16 | 山东大学 | Rock mass quality evaluation method, system and device based on hyperspectral imaging technology |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6374185B1 (en) * | 2000-02-18 | 2002-04-16 | Rdsp I, L.P. | Method for generating an estimate of lithological characteristics of a region of the earth's subsurface |
| CN103279616A (en) * | 2013-06-03 | 2013-09-04 | 西安近代化学研究所 | Virtual test method for influence of solid propellant smoke upon visible-light guidance signals |
| CN104808255A (en) * | 2015-04-30 | 2015-07-29 | 武汉光谷北斗控股集团有限公司 | Fractal theory-based mineralization anomaly information mining method |
| CN105117734A (en) * | 2015-07-28 | 2015-12-02 | 江南大学 | Corn seed hyper-spectral image classification identification method based on model on-line updating |
| CN111914847A (en) * | 2020-07-23 | 2020-11-10 | 厦门商集网络科技有限责任公司 | OCR recognition method and system based on template matching |
| WO2021108838A1 (en) * | 2019-12-02 | 2021-06-10 | Plotlogic Pty Ltd | Real time mine monitoring system and method |
| CN113486869A (en) * | 2021-09-07 | 2021-10-08 | 中国自然资源航空物探遥感中心 | Method, device and medium for lithology identification based on unsupervised feature extraction |
| CN114067094A (en) * | 2021-11-11 | 2022-02-18 | 山东大学 | Multispectral camera system for rock drilling jumbo and intelligent surrounding rock identification method |
| CN114386497A (en) * | 2021-12-31 | 2022-04-22 | 核工业北京地质研究院 | Aviation hyperspectral and gamma spectrum data fusion method oriented to uranium mineralization structure |
| CN114662626A (en) * | 2022-05-26 | 2022-06-24 | 湖北省国土测绘院 | Lithology automatic classification method and system for multi-source remote sensing data multi-feature fusion |
| CN114895711A (en) * | 2022-06-14 | 2022-08-12 | 北京华能新锐控制技术有限公司 | Automatic unmanned aerial vehicle flight path line planning method for fan blade inspection |
-
2022
- 2022-10-19 CN CN202211282405.XA patent/CN115713645B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6374185B1 (en) * | 2000-02-18 | 2002-04-16 | Rdsp I, L.P. | Method for generating an estimate of lithological characteristics of a region of the earth's subsurface |
| CN103279616A (en) * | 2013-06-03 | 2013-09-04 | 西安近代化学研究所 | Virtual test method for influence of solid propellant smoke upon visible-light guidance signals |
| CN104808255A (en) * | 2015-04-30 | 2015-07-29 | 武汉光谷北斗控股集团有限公司 | Fractal theory-based mineralization anomaly information mining method |
| CN105117734A (en) * | 2015-07-28 | 2015-12-02 | 江南大学 | Corn seed hyper-spectral image classification identification method based on model on-line updating |
| WO2021108838A1 (en) * | 2019-12-02 | 2021-06-10 | Plotlogic Pty Ltd | Real time mine monitoring system and method |
| CN111914847A (en) * | 2020-07-23 | 2020-11-10 | 厦门商集网络科技有限责任公司 | OCR recognition method and system based on template matching |
| CN113486869A (en) * | 2021-09-07 | 2021-10-08 | 中国自然资源航空物探遥感中心 | Method, device and medium for lithology identification based on unsupervised feature extraction |
| CN114067094A (en) * | 2021-11-11 | 2022-02-18 | 山东大学 | Multispectral camera system for rock drilling jumbo and intelligent surrounding rock identification method |
| CN114386497A (en) * | 2021-12-31 | 2022-04-22 | 核工业北京地质研究院 | Aviation hyperspectral and gamma spectrum data fusion method oriented to uranium mineralization structure |
| CN114662626A (en) * | 2022-05-26 | 2022-06-24 | 湖北省国土测绘院 | Lithology automatic classification method and system for multi-source remote sensing data multi-feature fusion |
| CN114895711A (en) * | 2022-06-14 | 2022-08-12 | 北京华能新锐控制技术有限公司 | Automatic unmanned aerial vehicle flight path line planning method for fan blade inspection |
Non-Patent Citations (4)
| Title |
|---|
| SIMA PEIGHAMBARI，YUN ZHANG: "Hyperspectral remote sensing in lithological mapping, mineral exploration, and environmental geology: An updated review", 《JOURNAL OF APPLIED REMOTE SENSING》, vol. 15, no. 3, 31 July 2021 (2021-07-31), pages 1 - 19, XP060145360, DOI: 10.1117/1.JRS.15.031501 * |
| 潘建平等: "顾及植被的复杂艰险地区多光谱遥感岩性", 《测绘科学》, vol. 46, no. 8, 31 August 2021 (2021-08-31), pages 120 - 126 * |
| 田淑芳等: "《热红外高光谱数据预处理与岩性填图方法研究》", 31 December 2021, 地质出版社, pages: 82 - 83 * |
| 许振浩等: "岩性识别:方法、现状及智能化发展趋势", 《地质论评》, 31 August 2022 (2022-08-31), pages 1 - 16 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116883852A (en) * | 2023-08-29 | 2023-10-13 | 北京建工环境修复股份有限公司 | Core data acquisition method and system based on hyperspectrum |
| CN116883852B (en) * | 2023-08-29 | 2024-03-08 | 北京建工环境修复股份有限公司 | Core data acquisition method and system based on hyperspectrum |
| CN118503905A (en) * | 2024-04-18 | 2024-08-16 | 山东大学 | Rock mass quality evaluation method, system and device based on hyperspectral imaging technology |
| CN118503905B (en) * | 2024-04-18 | 2024-11-01 | 山东大学 | Rock mass quality evaluation method, system and device based on hyperspectral imaging technology |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115713645B (en) | 2024-06-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115713645B (en) | Lithology recognition method and system based on spectral imaging technology | |
| CN110689531A (en) | Automatic power transmission line machine inspection image defect identification method based on yolo | |
| CN109190473A (en) | The application of a kind of " machine vision understanding " in remote monitoriong of electric power | |
| CN101101338A (en) | Exploration method and system for oil and gas | |
| CN113252700A (en) | Structural crack detection method, equipment and system | |
| CN115761038B (en) | Tunnel face geological sketch method and system based on image spectrum technology | |
| Kumar et al. | Deep remote sensing methods for methane detection in overhead hyperspectral imagery | |
| CN102590247A (en) | Steel wire rope defect detection method based on X-ray image processing technology | |
| CN115656053A (en) | Rock mineral content testing method and system | |
| CN111879709B (en) | Lake water body spectral reflectivity inspection method and device | |
| CN117423013B (en) | Ore potential prediction method based on geological big data | |
| CN115309956A (en) | Sampling position determining method, device, equipment and storage medium | |
| CN115761463A (en) | Shallow sea water depth inversion method, system, equipment and medium | |
| CN117422699A (en) | Highway detection method, highway detection device, computer equipment and storage medium | |
| CN117804368B (en) | Tunnel surrounding rock deformation monitoring method and system based on hyperspectral imaging technology | |
| CN114324202A (en) | Small watershed water quality monitoring method based on spectral analysis | |
| CN101615254B (en) | Method for extracting coal fire information by hyperspectral remote sensing based on generalized addition model | |
| CN107944474A (en) | Multiple dimensioned cooperation table based on local auto-adaptive dictionary reaches hyperspectral classification method | |
| CN116385593A (en) | Hyperspectral remote sensing mineral mapping method based on quantitative semi-supervised learning | |
| Aragua et al. | A cost-effective approach for chicken egg weight estimation through computer vision | |
| Quental et al. | Imaging spectroscopy answers to acid mine drainage detection at S. Domingos, Iberian Pyrite Belt, Portugal | |
| CN114359136B (en) | Stealth effect evaluation method and system based on ground imaging data | |
| Torsvik et al. | Detection of macroplastic on beaches using drones and object-based image analysis | |
| CN118279672B (en) | Image detection method and system for fuel rod assembly in nuclear power station pool | |
| RU2823446C1 (en) | Method and system for automated determination of core characteristics |
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 |