WO2024083261A1 - 一种基于图像光谱技术的隧道掌子面地质素描方法及系统 - Google Patents

一种基于图像光谱技术的隧道掌子面地质素描方法及系统 Download PDF

Info

Publication number
WO2024083261A1
WO2024083261A1 PCT/CN2023/128809 CN2023128809W WO2024083261A1 WO 2024083261 A1 WO2024083261 A1 WO 2024083261A1 CN 2023128809 W CN2023128809 W CN 2023128809W WO 2024083261 A1 WO2024083261 A1 WO 2024083261A1
Authority
WO
WIPO (PCT)
Prior art keywords
mineral
image
weathering
tunnel face
geological
Prior art date
Application number
PCT/CN2023/128809
Other languages
English (en)
French (fr)
Inventor
许振浩
林鹏
李珊
谢辉辉
石恒
马瑞阳
贺迎春
邵瑞琦
向航
李术才
Original Assignee
山东大学
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 山东大学 filed Critical 山东大学
Publication of WO2024083261A1 publication Critical patent/WO2024083261A1/zh

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
    • G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—2D [Two Dimensional] image generation
    • G06T11/20—Drawing from basic elements, e.g. lines or circles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—2D [Two Dimensional] image generation
    • G06T11/60—Editing figures and text; Combining figures or text
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/90—Determination of colour characteristics
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/54—Extraction of image or video features relating to texture
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
    • G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/30—Assessment of water resources

Definitions

  • the present invention relates to the technical field of intelligent tunnel geological cataloging, and in particular to a tunnel face geological sketching method and system based on image spectroscopy technology.
  • the geological conditions for tunnel construction are complex, and unfavorable geological bodies such as fault fracture zones, alteration zones, and karst are highly concealed. Carrying out geological cataloging during tunnel construction is an important means to grasp the unfavorable geological conditions along the tunnel. Geological cataloging can not only quickly, accurately and comprehensively understand the engineering geology and hydrogeology of the excavated sections of the tunnel, thereby verifying, correcting and improving the geological data of the preliminary survey and design, but also provide a geological basis for the advanced geological prediction of the tunnel and the optimization of the tunnel construction plan.
  • the use of geological sketch information feedback of the face can achieve the purpose of dynamic feedback design and efficient construction. At the same time, the surrounding rock information of the face is collected, sorted and systematically analyzed to make advanced geological predictions for tunnels under different geological conditions.
  • the traditional tunnel geological sketching method is time-consuming, highly subjective, and heavily dependent on the experience and meticulousness of geologists.
  • the content of the catalog is not comprehensive, prone to misjudgment and omissions, and has certain risks. Workers must bear the safety risks brought by close contact with the tunnel working face to conduct inspections.
  • spectral technology can be used to quantitatively obtain information such as surrounding rock composition, it not only reduces the errors caused by manual analysis, but also greatly improves the accuracy of geological sketches.
  • the current spectral technology has disadvantages such as small testing range, on-site handheld contact operation, and low efficiency.
  • the present invention proposes a tunnel face geological sketching method based on image spectroscopy technology and
  • the system obtains image information and spectral information of the tunnel face based on image spectroscopy technology, fuses the image information and spectral information, analyzes the stratigraphic characteristics, geological structure, and hydrogeology, thereby realizing a digital geological sketch of the tunnel face.
  • the present invention adopts the following technical solution:
  • the present invention provides a method for geological sketching of a tunnel face based on image spectroscopy technology, comprising:
  • the spectral information is subjected to mineral end-member extraction and spectral unmixing by using the mixed pixel decomposition method, the type and content of the minerals are identified, and the spatial distribution of different minerals is determined;
  • the number and relative area of cracks are obtained to obtain the degree of surrounding rock fragmentation.
  • the mineral weathering variation ratio is obtained.
  • the color difference classification results of different weathering types and regions are obtained.
  • the degree of surrounding rock fragmentation, the mineral weathering variation ratio and the color difference classification results is obtained.
  • the stratum lithology, mineral spatial distribution and content, and surrounding rock weathering degree are integrated into the first geological sketch map, and the fracture identification results, water discharge form and water discharge location are integrated into the second geological sketch map to complete the geological sketch of the tunnel face.
  • the process of identifying the type and content of the mineral includes:
  • the identified mineral types are decomposed as end members to obtain the percentage of each mineral.
  • the mineral combination, mineral end members and their content are marked according to each pixel.
  • the weight of each mineral in each pixel end member demixing is weighted with the pixel points occupied by the mineral to obtain the mineral content.
  • the image and spectral information of the measured face are rasterized, and the spectral information in the grid is averaged and then the spectral feature vector is extracted;
  • the image feature vector includes texture features of the feature band
  • the texture feature extraction process includes: obtaining a two-dimensional grayscale image after grayscale processing of the image information, extracting the texture features of the two-dimensional grayscale image using a grayscale co-occurrence matrix method, and calculating feature parameters in four directions of 0°, 45°, 90°, and 135°.
  • the fracture identification result includes the position, occurrence, opening and fracture filling of the fracture
  • the crack identification process includes: using image segmentation method to distinguish cracks from background, eliminating elements other than cracks, obtaining crack skeleton and crack outline; taking crack skeleton as the center and crack outline as the boundary, performing rasterization processing, and identifying crack fillings according to the spectral information in the grid.
  • the process of identifying the weathering degree of surrounding rocks includes:
  • the classification results of surrounding rock fragmentation degree are obtained by using the pre-trained surrounding rock fragmentation degree classification model
  • the spectral color difference is used to characterize the surface color differences between regions with different weathering types and weathering degrees;
  • the water outflow pattern of the tunnel face is obtained based on the image information using a trained water outflow image recognition model, and the water outflow position is determined;
  • the water outflow image recognition model is obtained by using the relationship between the tunnel face water outflow image and the rock water inflow situation of the tunnel face surrounding rock, and using tunnel face images of different water inflow categories to train the constructed network framework;
  • the water discharge forms include no sign of water discharge, seepage, dripping, line-shaped, stream-shaped and gushing water.
  • the geological sketching process of the tunnel face includes:
  • the mineral content is marked with content contour lines
  • the lithology of the formations is marked using different lithology legends and symbols;
  • the water outlet shape is marked with lines.
  • the present invention provides a tunnel face geological sketching system based on image spectroscopy technology, comprising:
  • a spectrum data volume acquisition module is configured to acquire images and spectrum information of the tunnel face to be measured
  • the mineral component identification module is configured to extract mineral end members and spectral information using a hybrid pixel decomposition method. Unmixing, identifying the types and contents of minerals, and determining the spatial distribution of different minerals;
  • a formation lithology identification module is configured to extract a spectral feature vector according to spectral information, extract an image feature vector according to image information, and determine the formation lithology using a trained classifier according to the spectral feature vector and the image feature vector;
  • a crack identification module is configured to extract crack features according to image information, identify crack fillers according to spectral information, and obtain crack identification results according to the crack features and the crack fillers;
  • the weathering degree identification module is configured to obtain the number and relative area of the cracks according to the crack identification results, so as to obtain the degree of surrounding rock fragmentation, obtain the mineral weathering variation ratio according to the comparison between the stratum lithology and the existing mineral types and contents, obtain the color difference classification results of different weathering types and regions according to the color characteristics of different regions, and obtain the degree of surrounding rock weathering according to the degree of surrounding rock fragmentation, the mineral weathering variation ratio and the color difference classification results;
  • a water outflow recognition module is configured to recognize the water outflow form and water outflow position of the tunnel face to be tested according to the image information
  • the geological sketch module is configured to integrate the stratum lithology, mineral spatial distribution and content, and surrounding rock weathering degree into a first geological sketch map, and to integrate the fracture identification results, water outlet form and water outlet position into a second geological sketch map, thereby completing the geological sketch of the tunnel face.
  • the present invention provides an electronic device comprising a memory and a processor, and computer instructions stored in the memory and executed on the processor, wherein when the computer instructions are executed by the processor, the method described in the first aspect is performed.
  • the present invention provides a computer-readable storage medium for storing computer instructions, wherein when the computer instructions are executed by a processor, the method described in the first aspect is performed.
  • the present invention has the following beneficial effects:
  • the tunnel face geological sketching method and system based on image spectroscopy technology proposed in the present invention adopts a non-contact, in-situ measurement method, which is an important technical means for rapid and large-area observation of material composition.
  • a three-dimensional data body can be obtained by shooting once. While obtaining the two-dimensional spatial information of the tunnel face to be measured, dozens to hundreds of continuous bands of one-dimensional spectral information can be collected. It has the advantages of obtaining rich data volume and unified map and spectrum.
  • the present invention proposes a method and system for geological sketching of a tunnel face based on image spectroscopy technology.
  • the method obtains image information of the face to be tested and spectral information of each pixel point based on image spectroscopy technology, fuses image information and spectral information to perform multi-source information fusion identification on stratum lithology, unmixes mixed pixels to identify mineral end members, and performs weighted processing to analyze mineral content.
  • Image information is used to identify crack distribution and crack occurrence, and spectrally detects crack filling minerals.
  • the degree of weathering of the surrounding rock is comprehensively judged by the degree of crushing, mineral weathering variation ratio, and color difference, as well as the hydrogeological analysis of the water outlet position and water outlet form, to achieve quantitative analysis of the geological sketch of the tunnel face.
  • the tunnel face geological sketching method and system based on image spectroscopy technology proposed in this invention replaces the traditional geological
  • the sketching method with the help of image and spectral information from image spectroscopy technology, has greatly avoided the qualitative analysis of subjective judgments, as well as the problem of incomplete cataloging content, which is prone to misjudgment and omission. It has greatly improved the accuracy and efficiency of geological cataloging and realized the intelligent geological sketching of the tunnel face.
  • the tunnel face geological sketching method and system based on image spectroscopy technology proposed in the present invention adopts artificial intelligence and data mining to process engineering geological information and hydrogeological information and establish a prediction and recognition model, thus achieving a leap from qualitative analysis based on subjective judgment to intelligent quantitative analysis.
  • FIG1 is a schematic diagram of a method for geological sketching of a tunnel face based on image spectroscopy technology provided in Example 1 of the present invention
  • FIG2 is a schematic diagram of an imaging method based on image spectroscopy technology provided in Example 1 of the present invention.
  • Figure 3(a)- Figure 3(b) are geological sketches provided in Example 1 of the present invention.
  • This embodiment provides a method for geological sketching of a tunnel face based on image spectroscopy technology.
  • Image spectroscopy technology is a non-destructive in-situ detection method that can simultaneously obtain image information and spectral information, and integrate image and spectral information for comprehensive analysis.
  • the geological record of the tunnel face can be comprehensively and accurately recorded.
  • the mixed pixel decomposition method is used to extract mineral end members and unmix the spectrum, identify the type and content of the mineral, and determine the spatial distribution of different minerals;
  • the spectral information in each grid is averaged to extract the spectral feature vector, and the image feature vector is extracted from the image information.
  • the trained classifier is used to determine the lithology of the formation.
  • the crack features are extracted according to the image information, the crack fillings are identified according to the spectral information, and the crack identification results are obtained according to the crack features and the crack fillings.
  • the crack identification results include the crack location, occurrence, opening and fillings.
  • the number and relative area of cracks are obtained to obtain the degree of surrounding rock fragmentation.
  • the mineral weathering variation ratio is obtained.
  • the color difference classification results of different weathering types and regions are obtained.
  • the degree of surrounding rock fragmentation, the mineral weathering variation ratio and the color difference classification results is obtained.
  • the stratum lithology, mineral spatial distribution and content, and surrounding rock weathering degree are integrated into the first geological sketch map, and the fracture identification results, water outlet form and water outlet position are integrated into the second geological sketch map, so as to complete the geological sketch of the tunnel face; among them, different colors and content contour lines are used for marking according to the mineral type and content, different legend symbols are used for marking the stratum lithology, and the surrounding rock weathering degree is marked in the form of regional marking.
  • the stratum lithology, mineral spatial distribution and content, and surrounding rock weathering degree are summarized to obtain the first geological sketch map; similarly, the fracture identification results, water outlet form and water outlet position are summarized to obtain the second geological sketch map.
  • an imaging spectrometer is used to collect image spectral information of the tunnel face to be measured.
  • FIG2 shows an imaging method based on image spectroscopy technology.
  • a non-contact, non-destructive in-situ scanning method is adopted.
  • a three-dimensional atlas data body can be obtained by shooting once.
  • the three-dimensional atlas data body includes image information and spectral information. While obtaining the two-dimensional spatial information of the tunnel face to be measured, one-dimensional spectral information of dozens to hundreds of continuous bands can be collected. This has the advantages of obtaining a large amount of data and unified atlas.
  • the image information of the tunnel face to be tested and the spectral information of each pixel point are obtained, and the image information and spectral information are integrated to analyze the stratum characteristics such as lithology, mineral composition, weathering degree, etc., the geological structure analysis of the fracture distribution, fracture occurrence and fracture filling, and the hydrogeological analysis of the water outlet position and water outlet form, so as to realize the tunnel Digital geological sketching of the tunnel face solves the problems of traditional tunnel face sketching, which is time-consuming, subjective and incomplete.
  • the image and spectral information are preprocessed; the preprocessing includes denoising, contrast enhancement, target space of interest extraction, etc., and the formation characteristic analysis, geological structure analysis and hydrological characteristic analysis are respectively performed according to the preprocessed image and spectral information to extract the engineering geological information and hydrogeological information of the tunnel face to be measured; wherein, the engineering geological information includes formation characteristics and geological structure, the formation characteristics include formation lithology, mineral composition and weathering degree, the geological structure includes fracture distribution, fracture occurrence, aperture and fracture filling, and the hydrogeological information includes water discharge location and water discharge form.
  • the engineering geological information includes formation characteristics and geological structure
  • the formation characteristics include formation lithology, mineral composition and weathering degree
  • the geological structure includes fracture distribution, fracture occurrence, aperture and fracture filling
  • the hydrogeological information includes water discharge location and water discharge form.
  • the process of mineral spectrum unmixing includes extraction of mineral end members, spectrum matching, identification of mineral types, and inversion of mineral content.
  • the spectrum information of the mineral detection point is extracted by using the mixed pixel decomposition method.
  • the mineral combination, mineral end members and their content are marked according to each pixel of the image.
  • the weight of each mineral unmixed by each pixel end member is weighted with the pixel points occupied by the mineral to obtain the mineral content of the study area.
  • the distribution distances of mineral detection points are selected, and the spectral information of the corresponding detection points is extracted;
  • the spectral information is subjected to water vapor absorption removal and denoising and smoothing processing to remove various accidental errors. This is because the influence of water vapor and atmosphere is inevitable during the spectrum measurement process at the tunnel face.
  • the spectrum shows different degrees of jumps in the atmosphere or water vapor absorption band, which need to be analyzed and removed. De-noising and smoothing can reduce the influence of noise to a certain extent.
  • the commonly used denoising and smoothing methods include moving average method, static average method, Fourier series approximation method, etc.
  • This embodiment takes the mixed pixel decomposition method as an example, and adopts the mixed pixel decomposition method to identify the surrounding rock mineral type and quantitatively analyze the content and perform weighted processing;
  • a mixed pixel decomposition method is used to extract mineral end members and their spectra; the following steps are included: mineral end member extraction, mineral type identification, abundance (mineral content) inversion;
  • PPI pure pixel index method
  • IOA fixed-point component analysis method
  • the information of mineral end members is extracted by the above method, and then the mineral end members and their corresponding spectra are spectrally matched with the pre-constructed reference spectrum, and the surrounding rock mineral type is determined by calculating the similarity;
  • spectral matching methods include distance similarity measurement, angle similarity measurement (spectral angle), spectral correlation coefficient, spectral binary coding and other methods; taking the spectral angle measurement method as an example, the similarity between the unknown mineral end-member spectrum and the reference spectrum is determined by calculating the angle value between the two, and the unknown mineral end-member spectrum is classified according to the similarity threshold.
  • the inversion of mineral content includes: first, decomposing the identified surrounding rock mineral types as end members to obtain the percentage content of each surrounding rock mineral; second, marking the mineral combination, mineral end members and their content according to each pixel of the image; weighting the weight of each mineral in each pixel end member demixing and the pixel points occupied by the mineral to obtain the mineral content of the study area;
  • the detected surrounding rock mineral composition and content can also be displayed in the form of a brief dot-line schematic diagram to show the changing trend of the mineral content.
  • the mineral content is displayed in the geological sketch summary diagram through content contour lines.
  • the spatial distribution of minerals on the face can also be displayed in the form of mineral mapping.
  • the image and spectral information of the face to be tested are rasterized, and the spectral information in each grid is averaged and then the spectral feature vector is extracted, and one spectral curve is extracted from one grid; then the image feature vector of the image information in the grid is extracted, and the spectral feature vector and the image feature vector are normalized and then the trained classifier is used to identify the lithology of the formation;
  • the processing of spectral information includes spectral preprocessing and extracting characteristic band spectra;
  • the spectrum preprocessing methods include S-G convolution smoothing, baseline correction, standard normal variable transformation, first-order derivative, second-order derivative, detrending, etc.
  • S-G convolution smoothing can effectively improve the smoothness of the spectrum and reduce high-frequency noise interference
  • standard normal variable transformation is mainly used to reduce the impact of uneven size of solid particles, surface scattering of objects and optical path transformation on spectral data
  • detrending is used to deal with the problem of baseline drift of diffuse reflectance spectra, and is generally used in combination with standard normal variable transformation
  • the first-order derivative and second-order derivative methods are used to eliminate background interference and baseline correction to improve resolution and sensitivity.
  • the method for selecting the characteristic band spectrum includes methods based on principal component analysis, minimum noise separation, continuous projection algorithm, boundary decision, etc.
  • the image feature vector extracted from the image information includes the texture features of the characteristic band;
  • the texture feature extraction process includes: obtaining a two-dimensional grayscale image after grayscale processing of the image information, extracting the texture features of the two-dimensional grayscale image using a grayscale co-occurrence matrix method, and calculating characteristic parameters such as energy, entropy, moment of inertia, correlation, etc. in four directions of 0°, 45°, 90°, and 135°.
  • the lithology of the formation is determined using the trained classifier; the classifier includes back propagation neural network, linear discriminant analysis, extreme learning machine, random forest, nonlinear partial least squares support Vector machines, etc.
  • the fracture skeleton and boundary contour are identified based on the image information, and then the fracture characteristics such as the fracture position, fracture occurrence and opening are marked on the sketch; then, based on the fracture skeleton and boundary contour, rasterization is performed so that the grid covers the position of the mineral filling, and the spectral information in each grid is extracted.
  • the fracture filling material in the fracture grid is identified based on the spectral information, and the fracture characteristics are supplemented with the fracture filling mineral, thereby improving the fracture information.
  • the image segmentation methods include threshold segmentation methods, boundary-based segmentation methods, region-based segmentation methods, and segmentation methods combined with specific theoretical tools;
  • binary image skeleton thinning algorithm can be used, such as Zhang Suen thinning algorithm, Thinning-Algorithm algorithm, etc., which includes image skeleton thinning and pruning algorithm to remove side branches;
  • contour extraction method For the extraction of crack contours, contour extraction method, boundary tracking method, etc. can be used;
  • the crack skeleton is used as the center and the crack outline is used as the boundary to perform grid processing, and the filling minerals are identified in each grid.
  • the method for identifying the minerals in each grid is consistent with the method for detecting the mineral composition of the surrounding rock, that is, the minerals are identified by using the mixed pixel decomposition method, which includes the determination of the end members in the spectral information, the spectral characteristics of each end member, spectral matching, identification of mineral types, etc.
  • crack characteristics are supplemented with the identification results of crack fillings to obtain a complete crack identification result, which includes the location, occurrence, opening and filling of the crack.
  • the structural structure of the surface of a weathered rock mass will change under long-term weathering, cracks will develop, and the degree of fragmentation will increase.
  • the mineral composition of its surface will change, with an increase in clay minerals or other secondary minerals and a decrease in primary minerals.
  • the corresponding spectral characteristics will also change, and the color of the rock surface will also change.
  • the long-term action of carbon dioxide in the air on the calcium, magnesium and other hydroxides in the rock mass on its surface will cause the surface color of the rock mass to turn white.
  • the deeper the weathering degree the more obvious the color representation, that is, the difference in surface color of areas with different weathering degrees leads to a large difference in color difference between areas with different weathering degrees and the benchmark.
  • the degree of surrounding rock fragmentation is obtained by the number and relative area of cracks
  • the mineral weathering variation ratio is obtained by analyzing the stratum lithology and the existing mineral types and contents
  • the degree of surrounding rock weathering is determined based on the color difference of different weathering areas.
  • the number and relative area of cracks are calculated based on the crack identification results to characterize the degree of rock fragmentation.
  • the number and relative area adopt the pre-trained surrounding rock fragmentation degree classification model to obtain the rock fragmentation degree classification results of surrounding rock weathering;
  • the image spectrum data of each pixel is used to reflect the color information of different weathering types and weathering degree areas; according to the color characteristics of different areas, the color difference between each pixel and the reference point is calculated, and the weathering degree of each pixel is evaluated.
  • the color difference classification results of surrounding rock weathering are obtained by using the pre-trained color difference classification model of different weathering types and weathering degree areas, and different weathering types and areas on the surface are effectively characterized;
  • the characteristic vectors of the crushing classification results, mineral classification results, and color difference classification results that characterize the weathering of the surrounding rock are extracted, input into the fusion analysis module and weighted respectively, and the final recognition result of the weathering degree of the surrounding rock is obtained.
  • the spectrum-based color measurement performs a linear transformation of the color space on the reflectance spectra of different weathering types and weathering degree areas, and calculates the spectral color difference values between different weathering types and weathering degrees with the reference point as the standard.
  • the relative area is the fracture area/target rock sample area, and the fracture area can be calculated by using a regional growing algorithm on the fracture area.
  • the various types of trained classification models are constructed based on the collected rock images of various weathering levels by establishing mapping relationships between crack characteristics-rock fragmentation degree, mineral composition variation ratio-weathering degree, and color difference-weathering degree. As the data of tunnel scanning recognition and classification accumulates, the corresponding classification models are continuously updated and optimized.
  • the classification of various types of surrounding rock weathering grades is based on the weathering degree classification table, and the weathering degree of the surrounding rock is classified according to the degree of structural damage, whether the mineral composition has changed, color change, etc., including unweathered, slightly weathered, moderately weathered, strongly weathered and fully weathered;
  • the identification characteristics of the unweathered weathering degree grade are that the rock is fresh and there is no trace of weathering;
  • the identification characteristics of slight weathering are that the organizational structure is basically unchanged, only the joint surface has iron and manganese rendering or the minerals are slightly discolored, and there are a small number of weathering cracks;
  • Moderate weathering is characterized by partial destruction of the organizational structure, changes in mineral composition, weathering of minerals near the joint surface into soil, development of weathering fissures, and cutting of the rock mass.
  • the above-mentioned different features are trained to obtain multiple classifiers, which are weighted respectively, and the weathering degree recognition results are fused at a decision level, and finally the final recognition result of the weathering degree of the surrounding rock is obtained through the joint action.
  • the water discharge form and water discharge position of the tunnel face to be tested are identified through image information, and the water discharge forms include no water discharge sign, water seepage, water dripping, linear, water flow and water gushing;
  • the data enhancement processing includes random blurring, local magnification, random horizontal flipping, Gaussian sampling, and channel scaling, etc., to alleviate the imbalance problem between tunnel face images with different water yields, achieve smooth noise reduction and detail removal within the image block, and retain the image edge to the greatest extent;
  • the processed image information is used with the trained water outflow image recognition model to obtain the water outflow pattern of the tunnel face and determine the water outflow location.
  • the water outflow image recognition model utilizes the relationship between the water outflow image of the face and the rock water gushing conditions of the surrounding rock of the tunnel face, and uses tunnel face images of different water gushing categories to train, verify and test the constructed network framework.
  • the engineering geological information and hydrogeological information of the tunnel face to be measured obtained in the above process form a digital tunnel face geological sketch map
  • the output tunnel face geological sketch map contains the digital geological information and the tunnel face geological sketch map
  • digital geological information is used to briefly describe the engineering geological information and hydrogeological information of the tunnel face, including "mileage, surrounding rock mineral type and content distribution, stratum lithology, degree of fracture development and fracture fillings, fracture morphology and distribution, degree of surrounding rock weathering, water richness of rock mass, and mileage, location, and form of water discharge in the tunnel".
  • the geological sketch of the tunnel face includes the distribution of stratum lithology, distribution and content of surrounding rock minerals, fracture information, water output information, etc.
  • the information obtained after observing a specific target with multiple different types of geological and hydrological information sensors is processed locally on the corresponding sensors to form their own geological sketches.
  • the images of the same scene and complementary information are fused, and the fusion judgment is used to form an image with richer information.
  • the mineral type, spatial distribution and content are shown. Different colors are used to represent different mineral types.
  • the content of the mineral is represented by a content contour line to reflect the difference in the content of each mineral at different locations.
  • different minerals are represented by contour lines of different colors.
  • the distribution of mineral I is marked with color 1
  • the distribution of mineral II is marked with color 2...
  • the content value of mineral I at different locations is Z 1
  • the content value of mineral II is Z 2 ...
  • the line color of the content contour line of mineral I is color 1
  • the line color of the content contour line of mineral II is color 2...
  • the points with the same mineral content value are connected into a curve.
  • Content contour lines can reflect the changes in the content of specific minerals.
  • the mineral identification results are superimposed on the lithology classification results, which will eventually represent the engineering geological information of the face, that is, the symbols and annotations of the stratum lithology, mineral spatial distribution and content, and the degree of weathering of the surrounding rock will be combined on the first geological sketch map.
  • the position, occurrence and opening of the fractures are marked with lines according to the identified fracture skeleton, outline and filling material, and the types of fracture filling minerals are marked with different colors; for the hydrogeological information of the face, the form of water is indicated by lines, etc.; thus, the annotations of the information representing the fracture characteristics and hydrological characteristics of the face are merged on the second geological sketch map.
  • This embodiment provides a tunnel face geological sketching system based on image spectroscopy technology, including:
  • a spectrum data volume acquisition module is configured to acquire images and spectrum information of the tunnel face to be measured
  • the mineral component identification module is configured to extract mineral end members and perform spectral unmixing using a mixed pixel decomposition method on spectral information, identify the type and content of minerals, and determine the spatial distribution of different minerals;
  • a formation lithology identification module is configured to extract a spectral feature vector according to spectral information, extract an image feature vector according to image information, and determine the formation lithology using a trained classifier according to the spectral feature vector and the image feature vector;
  • a crack identification module is configured to extract crack features according to image information, identify crack fillers according to spectral information, and obtain crack identification results according to the crack features and the crack fillers;
  • the weathering degree identification module is configured to obtain the number and relative area of the cracks according to the crack identification results, so as to obtain the degree of surrounding rock fragmentation, obtain the mineral weathering variation ratio according to the comparison between the stratum lithology and the existing mineral types and contents, obtain the color difference classification results of different weathering types and regions according to the color characteristics of different regions, and obtain the degree of surrounding rock weathering according to the degree of surrounding rock fragmentation, the mineral weathering variation ratio and the color difference classification results;
  • a water outflow recognition module is configured to recognize the water outflow form and water outflow position of the tunnel face to be tested according to the image information
  • the geological sketch module is configured to integrate the stratum lithology, mineral spatial distribution and content, and surrounding rock weathering degree into a first geological sketch map, and to integrate the fracture identification results, water outlet form and water outlet position into a second geological sketch map, thereby completing the geological sketch of the tunnel face.
  • Example 1 corresponds to the steps described in Example 1, and the examples and application scenarios implemented by the above modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above Example 1.
  • the blocks may be executed as part of a system in a computer system such as a set of computer executable instructions.
  • An electronic device includes a memory and a processor, and computer instructions stored in the memory and executed on the processor, wherein when the computer instructions are executed by the processor, the method described in Embodiment 1 is performed. For the sake of brevity, it will not be described in detail here.
  • the processor may be a central processing unit CPU, and the processor may also be other general-purpose processors, digital signal processors DSP, application-specific integrated circuits ASIC, off-the-shelf programmable gate arrays FPGA or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.
  • the memory may include a read-only memory and a random access memory, and provide instructions and data to the processor.
  • a portion of the memory may also include a non-volatile random access memory.
  • the memory may also store information about the device type.
  • a computer-readable storage medium is used to store computer instructions, and when the computer instructions are executed by a processor, the method described in Example 1 is completed.
  • the method in Example 1 can be directly embodied as a hardware processor, or a combination of hardware and software modules in the processor.
  • the software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc.
  • the storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware. To avoid repetition, it is not described in detail here.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Multimedia (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Databases & Information Systems (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Geometry (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Geophysics And Detection Of Objects (AREA)

Abstract

本发明公开一种基于图像光谱技术的隧道掌子面地质素描方法及系统，包括：采集掌子面的图像光谱数据，根据光谱信息提取矿物端元及其光谱，以确定矿物类型及含量；对图像信息提取纹理特征，对光谱信息提取特征波段光谱，以确定地层岩性；根据图像信息提取裂隙特征，根据光谱信息识别裂隙填充物，以得到裂隙识别结果；根据裂隙数量、相对面积得到围岩破碎程度，根据地层岩性、围岩矿物分析结果分析矿物风化变异比率，根据光谱色差识别结果获取不同风化程度的颜色差异，以得到围岩风化程度；根据图像信息识别待测掌子面的出水形态及出水位置；根据上述结果采用不同的标注符号进行标注汇总得到隧道掌子面地质素描图，提高地质编录的精度和效率。

Description

一种基于图像光谱技术的隧道掌子面地质素描方法及系统
本发明要求于2022年10月19日提交中国专利局、申请号为202211281307.4、发明名称为“一种基于图像光谱技术的隧道掌子面地质素描方法及系统”的中国专利申请的优先权，其全部内容通过引用结合在本发明中。
技术领域
本发明涉及隧道智能化地质编录技术领域，特别是涉及一种基于图像光谱技术的隧道掌子面地质素描方法及系统。
背景技术
本部分的陈述仅仅是提供了与本发明相关的背景技术信息，不必然构成在先技术。
隧道施工的地质条件复杂，断层破碎带、蚀变带、岩溶等不良地质体具有较强的隐蔽性，在隧道施工过程中开展地质编录是掌握隧道沿线不良地质情况的一种重要手段。地质编录不仅可以快速、准确和全面了解隧道已开挖段落的工程地质和水文地质情况，从而对前期勘察设计的地质资料进行核实、修正和完善，还可以为隧道超前地质预报和隧道施工方案的优化提供地质依据。在隧道施工中利用掌子面的地质素描信息反馈可以达到动态反馈设计和高效施工的目的，同时对掌子面的围岩信息进行采集整理和系统分析，为隧道在不同地质条件下进行超前地质预报。
传统的隧道地质素描法耗时、主观性强、严重依赖于地质工作者的经验和细致程度，编录内容不够全面，易出现误判漏判现象，且具有一定的风险性，工作人员必须承受与隧道工作面密切接触带来的安全风险来进行检查。
而且，发明人发现，现有的隧道掌子面智能化地质编录大都依赖于采集拍摄的掌子面的图片，但基于图像的深度学习技术在获取地质编录信息时，无法准确识别岩石的矿物含量信息以及岩相特征不明显的地质条件等。
利用光谱技术虽然可以定量化获取围岩成分等信息，不仅降低了人工分析的带来的误差，同时极大提升了地质素描的精度，但是目前的光谱技术存在测试范围小、需现场手持接触式操作、效率低等缺点。
发明内容
为了解决上述问题，本发明提出了一种基于图像光谱技术的隧道掌子面地质素描方法及 系统，基于图像光谱技术获取掌子面的图像信息和光谱信息，融合图像信息和光谱信息，进行地层特征、地质构造、水文地质的分析，从而实现对隧道掌子面的数字化地质素描。
为了实现上述目的，本发明采用如下技术方案：
第一方面，本发明提供一种基于图像光谱技术的隧道掌子面地质素描方法，包括：
获取待测掌子面的图像与光谱信息；
对光谱信息利用混合像元分解法进行矿物端元提取和光谱解混，识别矿物的类型及含量，并确定不同矿物的空间分布；
根据光谱信息提取光谱特征向量，根据图像信息提取图像特征向量，根据光谱特征向量和图像特征向量采用训练后的分类器确定地层岩性；
根据图像信息提取裂隙特征，根据光谱信息识别裂隙填充物，根据裂隙特征与裂隙填充物得到裂隙识别结果；
根据裂隙识别结果获取裂隙的数量和相对面积，以此得到围岩破碎程度，根据地层岩性与现有矿物类型及含量的比较，得到矿物风化变异比率，根据不同区域的颜色特征，得到不同风化类型及区域的色差分类结果，根据围岩破碎程度、矿物风化变异比率和色差分类结果得到围岩风化程度；
根据图像信息识别待测掌子面的出水形态及出水位置；
将地层岩性、矿物空间分布和含量以及围岩风化程度整合为第一地质素描图，将裂隙识别结果、出水形态及出水位置整合为第二地质素描图，以此完成对隧道掌子面的地质素描。
作为可选择的一种实施方式，识别矿物的类型及含量的过程包括：
提取矿物端元，将矿物端元及其对应的光谱与预构建的参考波谱进行光谱匹配，以确定矿物类型；
将识别出的矿物类型作为端元进行分解，得到每种矿物的百分含量，根据每个像元标注出矿物组合、矿物端元及其所占含量，将每个像元端元解混的每种矿物的权重与矿物所占像素点做加权处理，得到矿物含量。
作为可选择的一种实施方式，对待测掌子面的图像与光谱信息进行栅格化处理，对网格内的光谱信息均值化处理后提取光谱特征向量；
所述图像特征向量包括特征波段的纹理特征，纹理特征的提取过程包括：对图像信息进行灰度处理后得到二维灰度图像，采用灰度共生矩阵方法提取二维灰度图像的纹理特征，在0°、45°、90°、135°四个方向计算特征参数。
作为可选择的一种实施方式，所述裂隙识别结果包括裂隙的位置、产状、开度和裂隙填充物；
裂隙识别过程包括：采用图像分割法区分裂隙与背景，剔除除裂隙外的其他元素，获取裂隙骨架和裂隙轮廓；以裂隙骨架为中心、裂隙轮廓为边界，进行栅格化处理，根据网格内的光谱信息进行裂隙填充物的识别。
作为可选择的一种实施方式，围岩风化程度的识别过程包括：
根据裂隙的数量和相对面积，采用预训练的围岩破碎程度分类模型，得到围岩破碎程度的分类结果；
根据地层岩性与现有矿物类型及含量的比较，分析是否有矿物成分的变化，采用预训练的矿物风化变异比率模型，得到矿物风化变异比率；
根据不同区域的颜色特征，以用光谱色差来表征不同风化类型及风化程度区域间的表面颜色差异；
对围岩破碎程度、矿物风化变异比率和色差分类结果分别赋权后，得到围岩风化程度。
作为可选择的一种实施方式，根据图像信息采用训练后的出水图像识别模型得到隧道掌子面的出水形态，并确定出水位置；
所述出水图像识别模型是利用掌子面出水图像和掌子面围岩的岩石涌水情况间的关系，使用不同涌水类别的掌子面图像对构建的网络框架进行训练而得；
所述出水形态包括无出水迹象、渗水、滴水、线状、股水和涌水。
作为可选择的一种实施方式，隧道掌子面的地质素描过程包括：
对矿物类型采用不同的颜色进行标注；
对矿物含量采用含量等值线进行标注；
地层岩性采用不同的岩性图例和符号进行标注；
对存在围岩风化的区域标注围岩风化程度；
对裂隙的位置、产状和开度采用线条标注；
对裂隙充填物的种类采用不同的颜色标注；
对出水形态采用线条等标注。
第二方面，本发明提供一种基于图像光谱技术的隧道掌子面地质素描系统，包括：
图谱数据体获取模块，被配置为获取待测掌子面的图像与光谱信息；
矿物成分识别模块，被配置为对光谱信息利用混合像元分解法进行矿物端元提取和光谱 解混，识别矿物的类型及含量，并确定不同矿物的空间分布；
地层岩性识别模块，被配置为根据光谱信息提取光谱特征向量，根据图像信息提取图像特征向量，根据光谱特征向量和图像特征向量采用训练后的分类器确定地层岩性；
裂隙识别模块，被配置为根据图像信息提取裂隙特征，根据光谱信息识别裂隙填充物，根据裂隙特征与裂隙填充物得到裂隙识别结果；
风化程度识别模块，被配置为根据裂隙识别结果获取裂隙的数量和相对面积，以此得到围岩破碎程度，根据地层岩性与现有矿物类型及含量的比较，得到矿物风化变异比率，根据不同区域的颜色特征，得到不同风化类型及区域的色差分类结果，根据围岩破碎程度、矿物风化变异比率和色差分类结果得到围岩风化程度；
出水识别模块，被配置为根据图像信息识别待测掌子面的出水形态及出水位置；
地质素描模块，被配置为将地层岩性、矿物空间分布和含量以及围岩风化程度整合为第一地质素描图，将裂隙识别结果、出水形态及出水位置整合为第二地质素描图，以此完成对隧道掌子面的地质素描。
第三方面，本发明提供一种电子设备，包括存储器和处理器以及存储在存储器上并在处理器上运行的计算机指令，所述计算机指令被处理器运行时，完成第一方面所述的方法。
第四方面，本发明提供一种计算机可读存储介质，用于存储计算机指令，所述计算机指令被处理器执行时，完成第一方面所述的方法。
与现有技术相比，本发明的有益效果为：
1.本发明提出的基于图像光谱技术的隧道掌子面地质素描方法及系统，采用非接触式、原位测量的方式，是一种快速、大面积观测物质组成的重要技术手段，拍摄一次可获取一个三维数据体，在获取待测掌子面的二维空间信息的同时能采集几十到上百个连续波段的一维光谱信息，具有获取数据量丰富，图谱合一的优势。
2.本发明提出一种基于图像光谱技术的隧道掌子面地质素描方法及系统，基于图像光谱技术获取待测掌子面的图像信息和每个像素点的光谱信息，融合图像信息和光谱信息对地层岩性进行多源信息融合识别，混合像元解混识别矿物端元、加权处理分析矿物含量，图像信息识别裂隙分布、裂隙产状，光谱检测裂隙填充矿物，借助破碎程度、矿物风化变异比率、色差综合判识围岩风化程度，以及对出水位置、出水形态的水文地质的分析，实现对隧道掌子面地质素描的定量化分析。
3.本发明提出的基于图像光谱技术的隧道掌子面地质素描方法及系统，取代了传统地质 素描方法，借助图像光谱技术的图像和光谱信息，极大的规避了主观层面判断的定性分析，以及编录内容不够全面，易出现误判漏判现象的问题，极大的提高了地质编录的精度和效率，实现了隧道掌子面的智能化地质素描。
4.本发明提出的基于图像光谱技术的隧道掌子面地质素描方法及系统，采用人工智能和数据挖掘的手段进行工程地质信息和水文地质信息的数据处理和预测识别模型的建立，实现从主观层面判断的定性分析到智能化定量分析的跨越。
本发明附加方面的优点将在下面的描述中部分给出，部分将从下面的描述中变得明显，或通过本发明的实践了解到。
附图说明
构成本发明的一部分的说明书附图用来提供对本发明的进一步理解，本发明的示意性实施例及其说明用于解释本发明，并不构成对本发明的不当限定。
图1为本发明实施例1提供的基于图像光谱技术的隧道掌子面地质素描方法流程示意图；
图2为本发明实施例1提供的基于图像光谱技术的成像方式示意图；
图3(a)-图3(b)为本发明实施例1提供的地质素描图。
具体实施方式
下面结合附图与实施例对本发明做进一步说明。
应该指出，以下详细说明都是示例性的，旨在对本发明提供进一步的说明。除非另有指明，本文使用的所有技术和科学术语具有与本发明所属技术领域的普通技术人员通常理解的相同含义。
需要注意的是，这里所使用的术语仅是为了描述具体实施方式，而非意图限制根据本发明的示例性实施方式。如在这里所使用的，除非上下文另外明确指出，否则单数形式也意图包括复数形式，此外，还应当理解的是，术语“包括”和“具有”以及他们的任何变形，意图在于覆盖不排他的包含，例如，包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元，而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
在不冲突的情况下，本发明中的实施例及实施例中的特征可以相互组合。
实施例1
本实施例提供一种基于图像光谱技术的隧道掌子面地质素描方法，图像光谱技术是一种可以同时获取图像信息和光谱信息的无损原位检测方法，融合图像和光谱信息进行综合分析 可以全面精确地对隧道掌子面进行地质编录。
如图1所示，具体包括：
获取待测掌子面的三维图谱数据体，包括图像信息和光谱信息；
根据每个像素点的光谱信息利用混合像元分解法进行矿物端元提取和光谱解混，识别矿物的类型及含量，并确定不同矿物的空间分布；
对待测掌子面的图像与光谱信息进行栅格化处理；
对每个网格中的光谱信息经均值化处理后提取光谱特征向量，对图像信息提取图像特征向量，并将光谱特征向量和图像特征向量归一化后，采用训练后的分类器确定地层岩性；
对每个网格，根据图像信息提取裂隙特征，根据光谱信息识别裂隙填充物，根据裂隙特征与裂隙填充物得到裂隙识别结果，裂隙识别结果包括裂隙位置、产状、开度和填充物；
根据裂隙识别结果获取裂隙的数量和相对面积，以此得到围岩破碎程度，根据地层岩性与围岩现有矿物类型及含量的比较，得到矿物风化变异比率，根据不同区域的颜色特征，得到不同风化类型及区域的色差分类结果，根据围岩破碎程度、矿物风化变异比率和色差分类结果得到围岩风化程度；
根据图像信息识别待测掌子面的出水形态及出水位置；
将地层岩性、矿物空间分布和含量以及围岩风化程度整合为第一地质素描图，将裂隙识别结果、出水形态及出水位置整合为第二地质素描图，以此完成对隧道掌子面的地质素描；其中，根据矿物类型及含量采用不同的颜色和含量等值线进行标注，地层岩性采用不同的图例符号进行标注，围岩风化程度采用区域标注形式，最终将地层岩性、矿物空间分布和含量以及围岩风化程度汇总得到第一地质素描图；同理，汇总裂隙识别结果、出水形态及出水位置得到第二地质素描图。
在本实施例中，利用成像光谱仪对待测掌子面进行图像光谱信息的采集，如图2所示为基于图像光谱技术的成像方式，采用非接触式的无损原位扫描方式，拍摄一次可获取一个三维图谱数据体，三维图谱数据体包括图像信息和光谱信息，在获取待测掌子面的二维空间信息的同时能采集几十到上百个连续波段的一维光谱信息，具有获取数据量大，图谱合一的优势。
基于图像光谱技术获取待测掌子面的图像信息和每个像素点的光谱信息，融合图像信息和光谱信息，对地层岩性、矿物成分、风化程度等地层特征的分析，对裂隙分布、裂隙产状和裂隙填充的地质构造分析，以及对出水位置、出水形态的水文地质分析，从而实现对隧道 掌子面的数字化地质素描，解决传统隧道掌子面素描耗时、主观性强、不全面等问题。
在本实施例中，在获取到待测掌子面的图像信息和光谱信息后，对图像和光谱信息进行预处理；所述预处理包括去噪、增强对比度、目标感兴趣空间提取等，根据预处理后的图像和光谱信息分别进行地层特征分析、地质构造分析和水文特征分析，以提取待测掌子面的工程地质信息和水文地质信息；其中，工程地质信息包括地层特征和地质构造，地层特征包括地层岩性、矿物成分和风化程度，地质构造包括裂隙分布、裂隙产状、开度和裂隙填充物，水文地质信息包括出水位置、出水形态。
在本实施例中，矿物光谱解混的过程包括矿物端元的提取、光谱的匹配、矿物类型的识别、矿物含量的反演；首先利用混合像元分解的方法提取矿物检测点的光谱信息，其次根据图像的每个像元标注出矿物组合、矿物端元及其所占含量，将每个像元端元解混的每种矿物的权重与矿物所占像素点做加权处理，得到研究区域的矿物含量；
具体地：
根据不同隧洞地质编录的精度要求，选取矿物检测点分布的距离，并提取相应检测点的光谱信息；
对光谱信息进行水汽吸收去除和去噪平滑处理，以去除各种偶然性误差；这是因为在隧道掌子面进行波谱实测过程中，水汽和大气的影响是不可避免的，波谱在大气或水汽吸收带中呈现不同程度的跳动，需加以分析并去除；去噪平滑可以在一定程度上减少噪声的影响，通常采用的去噪平滑方法包括移动平均法、静态平均法、傅里叶级数求近似法等；
本实施例以混合像元分解方法为例，采用混合像元分解的方法进行围岩矿物类型识别及含量定量分析、加权处理；
根据处理后的光谱信息采用混合像元分解方法进行矿物端元及其光谱的提取；包括以下步骤：矿物端元提取、矿物类型识别、丰度(矿物含量)反演；
对于矿物端元的提取，可以基于图像的端元进行提取，比如纯净像元指数法(PPI)和定点成分分析法(IEA)等；
通过上述的方法提取出矿物端元的信息，然后将矿物端元及其对应的光谱与预构建的参考波谱进行光谱匹配，通过计算相似度确定围岩矿物类型；
其中，光谱匹配方法包括距离相似性度量、角度相似性度量(光谱角)、光谱相关系数、光谱二值编码等方法；以光谱角度量方法为例，通过计算未知矿物端元光谱与参考波谱间的夹角值，来确定两者的相似性，根据相似性阈值对未知矿物端元光谱进行分类。
矿物含量的反演包括：首先将识别出的围岩矿物类型作为端元进行分解，得到每种围岩矿物的百分含量，其次根据图像的每个像元标注出矿物组合、矿物端元及其所占含量，将每个像元端元解混的每种矿物的权重与矿物所占像素点做加权处理，得到研究区域的矿物含量；
不同矿物的数字化情况根据区域含量加权：
在本实施例中，还可以将检测出的围岩矿物成分和含量以简要的点线示意图的形式展示出矿物含量的变化趋势，在地质素描汇总图中矿物含量通过含量等值线进行展示，还可以通过矿物填图的形式展示掌子面的矿物空间分布情况。
在本实施例中，对待测掌子面的图像与光谱信息进行栅格化处理，对每个网格中的光谱信息经均值化处理后提取光谱特征向量，一个网格提取出一条光谱曲线；然后提取网格内图像信息的图像特征向量，并将光谱特征向量和图像特征向量归一化处理后采用训练后的分类器识别地层岩性；
具体地：
对隧道掌子面进行网格划分，提取每个网格的图像信息和光谱信息；
对光谱信息进行的处理包括光谱预处理和提取特征波段光谱；
所述光谱预处理的方法包括有S-G卷积平滑、基线校正、标准正态变量变换、一阶导数、二阶导数、去趋势等；
其中，S-G卷积平滑可以有效提高光谱的平滑性，减少高频噪音干扰；标准正态变量变换主要是减少固体颗粒物大小不均和物体表面散射以及光程变换对光谱数据的影响；去趋势用于处理漫反射光谱基线漂移的问题，一般和标准正态变量变换组合使用；一阶导数和二阶导数的方法用于消除背景干扰和基线校正，来提高分辨率和灵敏度。
所述特征波段光谱的选择方法包括基于主成分分析、最小噪声分离、连续投影算法、边界决策等。
对图像信息提取的图像特征向量包括特征波段的纹理特征；所述纹理特征的提取过程包括：对图像信息进行灰度处理后得到二维灰度图像，采用灰度共生矩阵方法提取二维灰度图像的纹理特征，在0°、45°、90°、135°四个方向计算能量、熵、惯性矩、相关性等特征参数。
将特征波段光谱和纹理特征经归一化处理后，采用训练后的分类器确定地层岩性；分类器包括反向传播神经网络、线性判别分析、极限学习机、随机森林、非线性偏最小二乘支持 向量机等。
在本实施例中，根据图像信息识别裂隙骨架和边界轮廓，进而在素描图上标记出裂隙位置、裂隙产状和开度的裂隙特征；然后以裂隙骨架和边界轮廓为基准，进行栅格化处理，使网格覆盖矿物填充的位置，提取每个网格内的光谱信息，根据光谱信息识别裂隙网格内的裂隙填充物，将裂隙特征与裂隙填矿物进行补充，进而完善裂隙信息。
具体地：
采用图像分割法区分裂隙与背景；所述图像分割法包括阈值分割法、基于边界的分割法、基于区域的分割法以及结合特定理论工具的分割方法等；
剔除图像中除裂隙外的其他元素，对裂隙的骨架和轮廓进行提取；
对于裂隙骨架的提取，可以采用二值图骨架细化算法，如Zhang Suen细化算法、Thining-Algorithm的算法等，其包括图像骨架细化和剪枝算法清除旁支；
对于裂隙轮廓的提取，可以采用轮廓提取法、边界跟踪法等；
待圈定出裂隙之后，以裂隙骨架为中心、裂隙轮廓为边界，进行栅格化处理，在每个网格内进行填充矿物的识别；
对每个网格内的矿物识别的方法使用上述围岩矿物成分检测的方法一致，即利用混合像元分解的方法对矿物进行识别，其包括光谱信息中端元的确定、每个端元的光谱特征、光谱匹配、识别矿物类型等；
最后将裂隙特征与裂隙填充物的识别结果进行补充，得到完整的裂隙识别结果，裂隙识别结果包括裂隙的位置、产状、开度、填充物。
风化岩体的岩体表面在长期风化作用下结构构造会发生改变，裂隙发育，破碎化程度增加，其表面矿物成分会发生改变，黏土矿物或其他次生矿物增多，原生矿物减少，相应的光谱特征也会发生改变，岩体表面颜色也会发生改变，例如其表面岩体中的钙、镁等氢氧化物在空气中二氧化碳的长期作用下，使得岩体表面颜色变为白色，风化程度越深，颜色表征越为明显，即不同风化程度区域表面颜色的不同，导致不同风化程度区域与基准间的色差存在较大差异。
所以，基于以上分析，在本实施例中，通过裂隙的数量、相对面积得到围岩破碎程度，通过分析地层岩性与现有矿物类型及含量得到矿物风化变异比率，以及根据不同风化区域的色差，从而判识围岩风化程度。
具体地，根据裂隙识别结果计算裂隙数量和相对面积，来表征岩体的破碎程度，对裂隙 数量和相对面积采用预训练的围岩破碎程度分类模型，得到围岩风化的岩石破碎程度分类结果；
根据地层岩性以及围岩现有矿物类型、含量，分析是否有矿物成分的变化，计算矿物风化的变异比率，来表征围岩风化矿物的变化，采用预训练的矿物风化变异比率模型，得到围岩风化的矿物分类结果；
利用每个像素点的图像光谱数据反映不同风化类型及风化程度区域的颜色信息；根据不同区域的颜色特征，计算每个像素点与基准点颜色的差异，对每个像素点的风化程度进行评估，采用预训练的不同风化类型及风化程度区域的色差分类模型，得到围岩风化的色差分类结果，对表面不同风化类型及区域进行有效表征；
提取表征围岩风化的破碎分类结果、矿物分类结果、色差分类结果特征向量，输入融合分析模块分别赋权后，得到最终的围岩风化程度的识别结果。
作为可选择的实施方式，基于光谱的颜色测量通过对不同的风化类型及风化程度区域的反射光谱进行颜色空间的线性变换，以基准点为标准，计算不同风化类型及风化程度间的光谱色差值。
作为可选择的实施方式，相对面积为裂隙面积/目标岩样面积，裂隙面积可由对裂隙区域采用区域生长算法计算得到。
作为可选择的实施方式，训练后的各类分类模型是根据已采集的各类不同风化等级的岩石图像，通过建立裂隙特征-岩体破碎程度、矿物成分变异比率-风化程度、色差-风化程度的映射关系而构建，且随着隧道扫描识别分类的数据积累，不断更新、优化对应的分类模型。
作为可选择的实施方式，各类不同围岩风化等级的划分依据风化程度分级表，根据组织结构破坏程度、矿物成分是否发生变化、颜色变化等情况对围岩风化程度进行分类，包括未风化、微风化、中等风化、强风化以及全风化；
其中，对于未风化的风化程度等级其判识特征为岩质新鲜，未见风化痕迹；
微风化的识别特征为组织结构基本未变，仅节理面有铁锰质渲染或矿物略有变色，有少量风化裂隙；
中等风化的特征为组织结构部分破坏，矿物成分发生变化，节理面附件的矿物风化成土状，风化裂隙发育，岩体被切割；
强风化的特征为组织结构大部分被破坏，矿物成分已显著变化，含大量粘土质粘土矿物，风化裂隙很发育，岩体被切割为碎块；
若组织结构已全部被破坏，矿物成分已全部改变并已风化成土状，则分为全风化。
在本实施例中，将上述的不同特征训练得到多个分类器分别赋权，对风化程度识别结果做一个决策级的融合，最终在共同作用下得到围岩风化程度的最终识别结果。
在本实施例中，通过图像信息识别待测掌子面的出水形态及出水位置，出水形态包括无出水迹象、渗水、滴水、线状、股水和涌水；
具体地：
对图像信息进行数据增强处理；所述数据增强处理包括随机模糊、局部放大、随机水平翻转、高斯采样和通道缩放等，缓解不同出水量掌子面图像之间的不平衡问题，实现图像块内平滑降噪、去细节，并最大程度保留图像边缘；
将处理后的图像信息采用训练后的出水图像识别模型得到隧道掌子面的出水形态，并确定出水位置；其中，出水图像识别模型是利用掌子面出水图像和隧道掌子面围岩的岩石涌水情况间的关系，使用不同涌水类别的隧道掌子面图像对构建的网络框架进行训练、验证和测试。
在本实施例中，根据上述过程得到的待测掌子面的工程地质信息和水文地质信息形成数字化的隧道掌子面地质素描图，输出的隧道掌子面地质素描图中含有数字化地质信息和掌子面地质素描图；
其中，数字化地质信息用于简述隧道掌子面的工程地质信息和水文地质信息，具体包括“里程、围岩矿物类型以及含量分布情况、地层岩性、裂隙发育程度以及裂隙充填物、裂隙的形态以及分布、围岩风化程度，岩体富水程度以及洞内出水里程、部位、出水形态”等。
隧道掌子面地质素描图中包括地层岩性分布情况、围岩矿物分布及含量情况、裂隙信息、出水信息等，经过多个不同类型的地质水文信息传感器观测某个特定目标后得到的信息在相应的传感器上完成本地处理，形成各自的地质素描图，将同一场景、信息互补的图像进行融合，融合判决形成信息量更丰富的图像。
如图3(a)-图3(b)所示，在隧道掌子面地质素描图中，对于矿物类型、空间分布及含量；矿物类型选用不同的颜色表示不同的矿物类型，对于矿物的含量采用含量等值线反应每种矿物在不同位置处含量的差异，同理不同的矿物用不同颜色的等值线表示；比如，矿物Ⅰ的分布采用颜色1进行标注，矿物II的分布采用颜色2进行标注……；分析不同位置处矿物Ⅰ的含量值为Z1、矿物II的含量值为Z2……，同时矿物Ⅰ的含量等值线的线条颜色为颜色1，矿物II的含量等值线的线条颜色为颜色2……；将同种矿物含量数值的个点连接成曲线， 含量等值线可以反映出特定矿物的含量变化情况。
对于地层岩性，不同的岩性用地质学的岩性图例和符号进行标注；例如花岗岩用“+”填充、页岩用“—”填充、辉绿岩用“X”填充；
对于掌子面不同的围岩风化程度，若存在风化区域，则在风化的部位标注出风化的程度；
将矿物识别结果叠加到岩性分类结果上，最终将表示掌子面的工程地质信息，即将地层岩性、矿物空间分布及含量以及围岩风化程度的符号和标注汇合在第一地质素描图上。
对于裂隙识别结果的标注，根据识别出的裂隙骨架、轮廓以及填充物，用线条标注裂隙的位置、产状、开度，用不同的颜色标注裂隙充填矿物的种类；对于掌子面的水文地质信息，用线条等表示出水的形态；从而将表示掌子面裂隙特征和水文特征的信息的标注汇合在第二地质素描图上。
实施例2
本实施例提供一种基于图像光谱技术的隧道掌子面地质素描系统，包括：
图谱数据体获取模块，被配置为获取待测掌子面的图像与光谱信息；
矿物成分识别模块，被配置为对光谱信息利用混合像元分解法进行矿物端元提取和光谱解混，识别矿物的类型及含量，并确定不同矿物的空间分布；
地层岩性识别模块，被配置为根据光谱信息提取光谱特征向量，根据图像信息提取图像特征向量，根据光谱特征向量和图像特征向量采用训练后的分类器确定地层岩性；
裂隙识别模块，被配置为根据图像信息提取裂隙特征，根据光谱信息识别裂隙填充物，根据裂隙特征与裂隙填充物得到裂隙识别结果；
风化程度识别模块，被配置为根据裂隙识别结果获取裂隙的数量和相对面积，以此得到围岩破碎程度，根据地层岩性与现有矿物类型及含量的比较，得到矿物风化变异比率，根据不同区域的颜色特征，得到不同风化类型及区域的色差分类结果，根据围岩破碎程度、矿物风化变异比率和色差分类结果得到围岩风化程度；
出水识别模块，被配置为根据图像信息识别待测掌子面的出水形态及出水位置；
地质素描模块，被配置为将地层岩性、矿物空间分布和含量以及围岩风化程度整合为第一地质素描图，将裂隙识别结果、出水形态及出水位置整合为第二地质素描图，以此完成对隧道掌子面的地质素描。
此处需要说明的是，上述模块对应于实施例1中所述的步骤，上述模块与对应的步骤所实现的示例和应用场景相同，但不限于上述实施例1所公开的内容。需要说明的是，上述模 块作为系统的一部分可以在诸如一组计算机可执行指令的计算机系统中执行。
在更多实施例中，还提供：
一种电子设备，包括存储器和处理器以及存储在存储器上并在处理器上运行的计算机指令，所述计算机指令被处理器运行时，完成实施例1中所述的方法。为了简洁，在此不再赘述。
应理解，本实施例中，处理器可以是中央处理单元CPU，处理器还可以是其他通用处理器、数字信号处理器DSP、专用集成电路ASIC，现成可编程门阵列FPGA或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
存储器可以包括只读存储器和随机存取存储器，并向处理器提供指令和数据、存储器的一部分还可以包括非易失性随机存储器。例如，存储器还可以存储设备类型的信息。
一种计算机可读存储介质，用于存储计算机指令，所述计算机指令被处理器执行时，完成实施例1中所述的方法。
实施例1中的方法可以直接体现为硬件处理器执行完成，或者用处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器、闪存、只读存储器、可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器，处理器读取存储器中的信息，结合其硬件完成上述方法的步骤。为避免重复，这里不再详细描述。
本领域普通技术人员可以意识到，结合本实施例描述的各示例的单元即算法步骤，能够以电子硬件或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
上述虽然结合附图对本发明的具体实施方式进行了描述，但并非对本发明保护范围的限制，所属领域技术人员应该明白，在本发明的技术方案的基础上，本领域技术人员不需要付出创造性劳动即可做出的各种修改或变形仍在本发明的保护范围以内。

Claims (10)

  1. 一种基于图像光谱技术的隧道掌子面地质素描方法，其特征在于，包括：
    获取待测掌子面的图像与光谱信息；
    对光谱信息利用混合像元分解法进行矿物端元提取和光谱解混，识别矿物的类型及含量，并确定不同矿物的空间分布；
    根据光谱信息提取光谱特征向量，根据图像信息提取图像特征向量，根据光谱特征向量和图像特征向量采用训练后的分类器确定地层岩性；
    根据图像信息提取裂隙特征，根据光谱信息识别裂隙填充物，根据裂隙特征与裂隙填充物得到裂隙识别结果；
    根据裂隙识别结果获取裂隙的数量和相对面积，以此得到围岩破碎程度，根据地层岩性与现有矿物类型及含量的比较，得到矿物风化变异比率，根据不同区域的颜色特征，得到不同风化类型及区域的色差分类结果，根据围岩破碎程度、矿物风化变异比率和色差分类结果得到围岩风化程度；
    根据图像信息识别待测掌子面的出水形态及出水位置；
    将地层岩性、矿物空间分布和含量以及围岩风化程度整合为第一地质素描图，将裂隙识别结果、出水形态及出水位置整合为第二地质素描图，以此完成对隧道掌子面的地质素描。
  2. 如权利要求1所述的一种基于图像光谱技术的隧道掌子面地质素描方法，其特征在于，识别矿物的类型及含量的过程包括：
    提取矿物端元，将矿物端元及其对应的光谱与预构建的参考波谱进行光谱匹配，以确定矿物类型；
    将识别出的矿物类型作为端元进行分解，得到每种矿物的百分含量，根据每个像元标注出矿物组合、矿物端元及其所占含量，将每个像元端元解混的每种矿物的权重与矿物所占像素点做加权处理，得到矿物含量。
  3. 如权利要求1所述的一种基于图像光谱技术的隧道掌子面地质素描方法，其特征在于，对待测掌子面的图像与光谱信息进行栅格化处理，对网格内的光谱信息均值化处理后提取光谱特征向量；
    所述图像特征向量包括特征波段的纹理特征，纹理特征的提取过程包括：对图像信息进行灰度处理后得到二维灰度图像，采用灰度共生矩阵方法提取二维灰度图像的纹理特征，在0°、45°、90°、135°四个方向计算特征参数。
  4. 如权利要求1所述的一种基于图像光谱技术的隧道掌子面地质素描方法，其特征在于，所述裂隙识别结果包括裂隙的位置、产状、开度和裂隙填充物；
    裂隙识别过程包括：采用图像分割法区分裂隙与背景，剔除除裂隙外的其他元素，获取裂隙骨架和裂隙轮廓；以裂隙骨架为中心、裂隙轮廓为边界，进行栅格化处理，根据网格内的光谱信息进行裂隙填充物的识别。
  5. 如权利要求1所述的一种基于图像光谱技术的隧道掌子面地质素描方法，其特征在于， 围岩风化程度的识别过程包括：
    根据裂隙的数量和相对面积，采用预训练的围岩破碎程度分类模型，得到围岩破碎程度的分类结果；
    根据地层岩性与现有矿物类型及含量的比较，分析是否有矿物成分的变化，采用预训练的矿物风化变异比率模型，得到矿物风化变异比率；
    根据不同区域的颜色特征，以用光谱色差来表征不同风化类型及风化程度区域间的表面颜色差异；
    对围岩破碎程度、矿物风化变异比率和色差分类结果分别赋权后，得到围岩风化程度。
  6. 如权利要求1所述的一种基于图像光谱技术的隧道掌子面地质素描方法，其特征在于，根据图像信息采用训练后的出水图像识别模型得到隧道掌子面的出水形态，并确定出水位置；
    所述出水图像识别模型是利用掌子面出水图像和掌子面围岩的岩石涌水情况间的关系，使用不同涌水类别的掌子面图像对构建的网络框架进行训练而得；
    所述出水形态包括无出水迹象、渗水、滴水、线状、股水和涌水。
  7. 如权利要求1所述的一种基于图像光谱技术的隧道掌子面地质素描方法，其特征在于，隧道掌子面的地质素描过程包括：
    对矿物类型采用不同的颜色进行标注；
    对矿物含量采用含量等值线进行标注；
    地层岩性采用不同的岩性图例和符号进行标注；
    对存在围岩风化的区域标注围岩风化程度；
    对裂隙的位置、产状和开度采用线条标注；
    对裂隙充填物的种类采用不同的颜色标注；
    对出水形态采用线条等标注。
  8. 一种基于图像光谱技术的隧道掌子面地质素描系统，其特征在于，包括：
    图谱数据体获取模块，被配置为获取待测掌子面的图像与光谱信息；
    矿物成分识别模块，被配置为对光谱信息利用混合像元分解法进行矿物端元提取和光谱解混，识别矿物的类型及含量，并确定不同矿物的空间分布；
    地层岩性识别模块，被配置为根据光谱信息提取光谱特征向量，根据图像信息提取图像特征向量，根据光谱特征向量和图像特征向量采用训练后的分类器确定地层岩性；
    裂隙识别模块，被配置为根据图像信息提取裂隙特征，根据光谱信息识别裂隙填充物，根据裂隙特征与裂隙填充物得到裂隙识别结果；
    风化程度识别模块，被配置为根据裂隙识别结果获取裂隙的数量和相对面积，以此得到围岩破碎程度，根据地层岩性与现有矿物类型及含量的比较，得到矿物风化变异比率，根据不同区域的颜色特征，得到不同风化类型及区域的色差分类结果，根据围岩破碎程度、矿物风化变异比率和色差分类结果得到围岩风化程度；
    出水识别模块，被配置为根据图像信息识别待测掌子面的出水形态及出水位置；
    地质素描模块，被配置为将地层岩性、矿物空间分布和含量以及围岩风化程度整合为第一地质素描图，将裂隙识别结果、出水形态及出水位置整合为第二地质素描图，以此完成对隧道掌子面的地质素描。
  9. 一种电子设备，其特征在于，包括存储器和处理器以及存储在存储器上并在处理器上运行的计算机指令，所述计算机指令被处理器运行时，完成权利要求1-7任一项所述的方法。
  10. 一种计算机可读存储介质，其特征在于，用于存储计算机指令，所述计算机指令被处理器执行时，完成权利要求1-7任一项所述的方法。
PCT/CN2023/128809 2022-10-19 2023-10-31 一种基于图像光谱技术的隧道掌子面地质素描方法及系统 WO2024083261A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211281307.4 2022-10-19
CN202211281307.4A CN115761038B (zh) 2022-10-19 2022-10-19 一种基于图像光谱技术的隧道掌子面地质素描方法及系统

Publications (1)

Publication Number Publication Date
WO2024083261A1 true WO2024083261A1 (zh) 2024-04-25

Family

ID=85353896

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/128809 WO2024083261A1 (zh) 2022-10-19 2023-10-31 一种基于图像光谱技术的隧道掌子面地质素描方法及系统

Country Status (2)

Country Link
CN (1) CN115761038B (zh)
WO (1) WO2024083261A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115761038B (zh) * 2022-10-19 2023-06-30 山东大学 一种基于图像光谱技术的隧道掌子面地质素描方法及系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103698817A (zh) * 2013-12-27 2014-04-02 上海川海信息科技有限公司 基于图像的地下洞室掌子面快速安全预测方法
WO2021121224A1 (zh) * 2019-12-17 2021-06-24 山东大学 基于岩石矿物分析的tbm隧道断层破碎带预报系统及方法
CN114067094A (zh) * 2021-11-11 2022-02-18 山东大学 用于凿岩台车的多光谱摄像系统及围岩智能判识方法
CN114926101A (zh) * 2022-07-18 2022-08-19 石家庄铁道大学 一种面向施工速度与安全的tbm适应性围岩分级方法
US20220317021A1 (en) * 2020-01-21 2022-10-06 Shandong University Determining system and method for weathering resistant capability of clastic rocks in tunnel based on feldspar features
CN115761038A (zh) * 2022-10-19 2023-03-07 山东大学 一种基于图像光谱技术的隧道掌子面地质素描方法及系统

Family Cites Families (24)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103389527A (zh) * 2013-08-01 2013-11-13 中国建筑第四工程局有限公司 隧道地质的预报方法和系统
CN104502990A (zh) * 2015-01-06 2015-04-08 中国地质大学（武汉） 一种基于数码图像的隧道掌子面地质调查方法
CN106651893A (zh) * 2016-12-23 2017-05-10 贵州电网有限责任公司电力科学研究院 一种基于边缘检测的墙体裂缝识别方法
CN106596896A (zh) * 2016-12-30 2017-04-26 山东大学 一种现场隧道实时围岩分级系统及方法
CN108089237A (zh) * 2017-12-13 2018-05-29 中国铁建大桥工程局集团有限公司 一种隧道底部基础多层岩溶溶洞的探测方法
CN109187548B (zh) * 2018-07-24 2020-04-07 西南交通大学 一种岩体裂隙识别方法
CN109191521B (zh) * 2018-12-05 2019-05-24 中国铁建重工集团有限公司 一种隧道点云数据分析方法及系统
CN110221341A (zh) * 2019-07-03 2019-09-10 武汉中科岩土工程有限责任公司 一种隧道施工不良地质超前预报方法
WO2021042668A1 (zh) * 2019-09-06 2021-03-11 山东大学 Tbm搭载式隧道围岩结构虚拟再现系统与方法
CN110514126A (zh) * 2019-09-30 2019-11-29 西南石油大学 一种基于机器视觉的隧道掌子面位移监测的方法
CN110850435A (zh) * 2019-11-11 2020-02-28 山东大学 车载式隧道地质素描装置及使用方法
CN110988992A (zh) * 2019-12-23 2020-04-10 中铁十二局集团第二工程有限公司 一种矿山法施工超前地质预报方法
CN111267984B (zh) * 2020-01-21 2021-04-13 山东大学 基于高光谱技术分析的隧道内不良地质体识别系统与方法
CN111638200B (zh) * 2020-04-22 2021-11-23 山东大学 基于拉曼光谱分析的地质预报系统及方法
CN111505733A (zh) * 2020-04-30 2020-08-07 山东科技大学 一种地下工程裂隙岩体赋存特殊地质体多源异构信息融合分析与动态预测方法
CN111626358B (zh) * 2020-05-27 2023-07-18 大连海事大学 一种基于bim图片识别的隧道围岩分级方法
CN111677530B (zh) * 2020-06-04 2022-04-26 中铁六局集团太原铁路建设有限公司 隧道贯通段三台阶临时仰拱安全施工工法
CN113074695A (zh) * 2021-04-06 2021-07-06 西南交通大学 一种基于掌子面三维变形监测的隧道稳定性判释方法
CN113177302B (zh) * 2021-04-14 2023-07-14 北京科技大学 一种裂隙岩石颗粒流模型构建及新生裂纹产状分析方法
CN113256082A (zh) * 2021-05-07 2021-08-13 中交第二航务工程局有限公司 一种隧道掌子面智能地质素描及地质评价方法
CN113484910B (zh) * 2021-06-28 2024-03-01 西安建筑科技大学 基于地震干涉法的隧道超前地质预报方法及系统
CN114076552A (zh) * 2021-07-16 2022-02-22 中交一公局集团有限公司 一种隧道智能爆破方法及系统
CN113936110A (zh) * 2021-10-15 2022-01-14 青海省地质调查院(青海省地质矿产研究院、青海省地质遥感中心) 一种基于数字地图的造山带填图方法及系统
CN114419088A (zh) * 2022-01-24 2022-04-29 重庆中环建设有限公司 一种基于数字图像的隧道围岩信息提取方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103698817A (zh) * 2013-12-27 2014-04-02 上海川海信息科技有限公司 基于图像的地下洞室掌子面快速安全预测方法
WO2021121224A1 (zh) * 2019-12-17 2021-06-24 山东大学 基于岩石矿物分析的tbm隧道断层破碎带预报系统及方法
US20220317021A1 (en) * 2020-01-21 2022-10-06 Shandong University Determining system and method for weathering resistant capability of clastic rocks in tunnel based on feldspar features
CN114067094A (zh) * 2021-11-11 2022-02-18 山东大学 用于凿岩台车的多光谱摄像系统及围岩智能判识方法
CN114926101A (zh) * 2022-07-18 2022-08-19 石家庄铁道大学 一种面向施工速度与安全的tbm适应性围岩分级方法
CN115761038A (zh) * 2022-10-19 2023-03-07 山东大学 一种基于图像光谱技术的隧道掌子面地质素描方法及系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
LIU YANGFEI: "Analysis of Applicability of Tunnel Advanced Geological Forecast Method", CHINESE JOURNAL OF ENGINEERING GEOPHYSICS, vol. 15, no. 6, 1 November 2018 (2018-11-01), XP093160779 *

Also Published As

Publication number Publication date
CN115761038A (zh) 2023-03-07
CN115761038B (zh) 2023-06-30

Similar Documents

Publication Publication Date Title
Chen et al. Automated extraction and evaluation of fracture trace maps from rock tunnel face images via deep learning
Lemy et al. Discontinuity trace map construction using photographs of rock exposures
CN103400151B (zh) 一体化的光学遥感影像与gis自动配准与水体提取方法
CN104050252B (zh) 一种高光谱遥感蚀变信息提取方法
CN102879099B (zh) 一种基于高光谱成像的壁画信息提取方法
CN109711288A (zh) 基于特征金字塔和距离约束fcn的遥感船舶检测方法
WO2024083261A1 (zh) 一种基于图像光谱技术的隧道掌子面地质素描方法及系统
Mostafa et al. Shadow identification in high resolution satellite images in the presence of water regions
CN103020649A (zh) 一种基于纹理信息的森林类型识别方法
Van Ruitenbeek et al. Measuring rock microstructure in hyperspectral mineral maps
CN110188716B (zh) 一种矿化信息提取方法及系统
CN108021890A (zh) 一种基于plsa和bow的高分遥感影像港口检测方法
Salati et al. Lithological mapping and fuzzy set theory: Automated extraction of lithological boundary from ASTER imagery by template matching and spatial accuracy assessment
Zhou et al. Unmixing urban hyperspectral imagery using probability distributions to represent endmember variability
Chen et al. A mathematical morphology-based multi-level filter of LiDAR data for generating DTMs
Liu et al. Extraction of earthquake-induced collapsed buildings from bi-temporal VHR images using object-level homogeneity index and histogram
Omran Rapid prediction of soil mineralogy using imaging spectroscopy
CN113850769A (zh) 一种基于Siamese空谱联合卷积网络的高光谱变化检测方法
Zhang et al. Building footprint and height information extraction from airborne LiDAR and aerial imagery
Lemy et al. A digital face mapping case study in an underground hard rock mine
Daghigh et al. Discontinuity plane extraction from a rock mass point cloud using unsupervised machine learning
Cal High-resolution object-based building extraction using PCA of LiDAR nDSM and aerial photos
Zhao et al. Detection and Identification of Surface Cover in Coalbed Methane Enrichment Area Based on Spectral Unmixing
Shamsiev et al. Structural and technological complex of methods for processing satellite images
CN107590824A (zh) 基于三维图像处理技术的岩石颗粒识别及位移跟踪方法