A method for quantitatively characterizing natural structural plane inhomogeneity

Abstract

The application discloses a natural structure surface non-uniformity quantitative characterization method and relates to the technical field of geotechnical engineering. First, the structure surface image is standardized collected, and a single structure surface image is taken as an analysis object. Based on preset mineral color characteristics, pixel in the image is matched and determined by class, so that various minerals are identified and spatial distribution information of the minerals is extracted. Then, a multi-index comprehensive evaluation system is constructed, and quantitative characterization of mineral composition of the structure surface and spatial non-uniform distribution characteristics is realized. Compared with the limitation of traditional mineralogical analysis method which only focuses on mineral composition, the method can characterize the spatial distribution characteristics of the minerals on the structure surface, realize overall description of mineral distribution form and aggregation characteristics, realize mineral region extraction and parameter calculation based on image recognition, relatively simplify the test process, and has higher analysis efficiency. The method is suitable for rapid quantitative analysis of the structure surface sample in engineering and has good engineering application value.

Description

Technical Field

[0001] This application belongs to the field of geotechnical engineering technology, specifically relating to a quantitative characterization method for the non-uniformity of natural structural surfaces, applicable to the quantitative analysis of the mineral composition and spatial distribution characteristics of natural hard structural surfaces. Background Technology

[0002] Structural planes are key structural units controlling the mechanical properties of rock masses and the stability of underground engineering projects. In deep geological environments, natural structural planes typically undergo complex tectonic activities and alteration, often developing various alteration minerals within them and forming spatially unevenly distributed patchy structures, such as chlorite, sericite, and calcite. Different minerals exhibit significant differences in strength, deformation, and frictional properties, displaying a distinct non-uniform distribution characteristic on the structural plane.

[0003] However, in existing structural surface mechanical analysis and engineering evaluation, structural surfaces are usually simplified as homogeneous weakened interfaces or uniform filling layers, and described using a single overall mechanical parameter. This homogenization approach ignores the differences in mineral composition and spatial distribution within the structural surface, making it difficult to reflect the true mechanical response characteristics of the structural surface, thus affecting the accuracy of engineering analysis results.

[0004] On the other hand, existing methods for mineral analysis of structural surfaces mainly include thin section identification, scanning electron microscopy analysis, and X-ray diffraction analysis. These methods are mostly for local observation of mineral composition at the microscale. Although they can obtain information on mineral types, they are difficult to reflect the spatial distribution characteristics of minerals within the overall range of the structural surface. Moreover, the testing process is complex and inefficient, which is not conducive to rapid quantitative analysis of structural surface samples in engineering.

[0005] In summary, existing methods are insufficient for the overall and quantitative characterization of mineral composition and its spatial distribution heterogeneity, and a unified evaluation method capable of simultaneously depicting differences in mineral composition and spatial distribution characteristics is lacking. Therefore, it is necessary to provide a quantitative characterization method for the heterogeneity of natural structural planes to achieve the overall identification and comprehensive evaluation of the heterogeneity characteristics of structural planes. Summary of the Invention

[0006] In view of at least one of the above-mentioned technical problems, this application provides a method for quantitative characterization of the heterogeneity of natural structural planes. This method uses a single structural plane image as the analysis object, identifies the mineral composition and spatial distribution characteristics of the structural plane, constructs a multi-index evaluation system, and achieves quantitative characterization of the heterogeneity of the mineral composition and spatial distribution of the structural plane.

[0007] The technical solution of this application is:

[0008] A quantitative characterization method for the non-uniformity of natural structural planes, comprising the following steps:

[0009] Step 1: Acquire structural surface images and establish a structural surface image dataset;

[0010] Step 2: Based on the structural surface images in the structural surface image dataset, extract typical color feature parameters of various mineral regions, and perform parametric representation and vectorized modeling of the typical color features to construct a mineral-color feature database;

[0011] Step 3: Obtain the structural surface image to be analyzed from the structural surface image dataset. Based on the mineral-color feature database, perform mineral-by-mineral color matching and recognition on the structural surface image to be analyzed, identify and extract the segmentation regions of various minerals, reconstruct the mineral distribution map of the structural surface, and iteratively optimize the segmentation parameters by comparing and verifying the reconstructed mineral distribution map of the structural surface with the original image until the comparison and verification results of the reconstructed mineral distribution map of the structural surface and the original image are consistent.

[0012] Step 4: Based on the segmented regions of each mineral, the pixel-spatial distribution information of each mineral in the structural surface image to be analyzed is statistically analyzed and extracted to obtain the mineral pixel-spatial distribution parameters;

[0013] Step 5: Based on the mineral pixel-spatial distribution parameters, calculate the mineral spatial distribution characteristic parameters used to characterize the mineral composition ratio, spatial concentration degree, and patch-scale distribution characteristics;

[0014] Step 6: Based on the spatial distribution characteristic parameters of minerals, construct the mineral composition non-uniformity index and the mineral spatial distribution non-uniformity index, and then establish a comprehensive index of mineral non-uniformity of structural surfaces.

[0015] Furthermore, according to the aforementioned method for quantitative characterization of the inhomogeneity of natural structural surfaces, step 2 specifically includes the following steps:

[0016] Step 2.1: Based on the color differences of different mineral regions in the structural surface image, extract typical color feature parameters of various mineral regions;

[0017] Step 2.2: Based on the extracted typical color feature parameters, construct the typical color feature vector M corresponding to each type of mineral. i M i The expression is as follows:

[0018]

[0019] Among them, M i H represents the typical color feature vector of the i-th type of mineral; i S i B i These correspond to the hue, saturation, and brightness parameters extracted from the structural surface image of the i-th mineral region, respectively.

[0020] Step 2.3: Set the typical color feature vectors of various minerals as the base colors for color matching, and save them as separate mineral color matching parameter files;

[0021] Step 2.4: Number and store all mineral color matching parameter files in a preset order to form a mineral-color feature database.

[0022] Furthermore, according to the quantitative characterization method for the non-uniformity of natural structural surfaces, step 3, which involves performing mineral-by-mineral color matching and identification on the structural surface image to be analyzed based on the mineral-color feature database, includes: sequentially loading the color matching parameter files of each mineral in the mineral-color feature database, and then, for a single mineral, performing a matching determination based on the distance calculation result between the mineral color feature vector and the pixel color feature vector.

[0023] Furthermore, according to the aforementioned method for quantitative characterization of the inhomogeneity of natural structural surfaces, step 3 specifically includes the following steps:

[0024] Step 3.1: Convert the original structural surface image to a preset color space;

[0025] Step 3.2: Load the color matching parameter file for each type of mineral sequentially from the mineral-color feature database, and then, based on the color feature vector M of the currently loaded i-th type of mineral... i The pixels in the structural surface image are matched one by one with the color tolerance threshold T.

[0026] Step 3.3: Based on the color matching results, pixels that match the color characteristics of a type of mineral are spatially aggregated to form the distribution area of ​​that type of mineral. Then, the distribution area is processed to remove isolated noise pixels, and finally the segmented area of ​​that type of mineral is obtained.

[0027] Step 3.4: After completing the region segmentation processing corresponding to all mineral categories, assign the corresponding color to all mineral segmentation regions according to their corresponding mineral categories, and maintain the relative spatial position relationship of the mineral segmentation regions in the structural surface image to generate a structural surface mineral distribution reconstruction map. Different minerals are identified as multiple spatially independent patch regions on the structural surface, so that the spatial distribution characteristics of various minerals can be intuitively reflected.

[0028] Step 3.5: Compare and verify the mineral distribution reconstruction map generated in Step 3.4 with the original structural surface image. By comparing and verifying whether the spatial distribution and boundary morphology of each mineral region are basically consistent with the corresponding mineral region in the original structural surface image, it is determined whether there is oversegmentation or undersegmentation: if it is found that the same type of mineral region is oversegmented, it is determined that there is oversegmentation; if it is found that different mineral regions are mixed, it is determined that there is undersegmentation.

[0029] Step 3.6: Based on the comparison and verification results in Step 3.5, adjust the color tolerance threshold T: if oversegmentation is determined, increase the color tolerance threshold T to improve the connectivity of similar mineral regions; if undersegmentation is determined, decrease the color tolerance threshold T to enhance the ability to distinguish between different minerals.

[0030] The method for increasing or decreasing the color tolerance threshold T is as follows: the color tolerance threshold is increased or decreased step by step in an iterative manner, with each increase or decrease being a preset small percentage of the current threshold, preferably 1% to 10%;

[0031] Step 3.7: Repeat steps 3.2 to 3.5 until the mineral distribution reconstruction map and the mineral regions in the original structural surface image are basically consistent in terms of spatial distribution and boundary morphology.

[0032] Furthermore, according to the aforementioned method for quantitative characterization of the heterogeneity of natural structural surfaces, step 3.2, which involves using the color feature vector M of the currently loaded i-th type of mineral,... i The process involves matching and determining each pixel in the structural surface image, including:

[0033] For each pixel in the image, construct its color feature vector M. p And calculate its color feature vector M of the currently loaded i-th type of mineral. i The Euclidean distance D between them i :

[0034]

[0035] in These represent the hue, saturation, and brightness values ​​for each pixel.

[0036]

[0037] Set a uniform color tolerance threshold T, and define that when D is satisfied... i When < T, the pixel is determined to meet the color matching condition, belongs to the i-th type of mineral, and is marked as a pixel of that type of mineral; when If the pixel does not belong to the current mineral, it will not be marked.

[0038] Furthermore, according to the aforementioned method for quantitative characterization of the non-uniformity of natural structural surfaces, step 4 specifically includes the following steps:

[0039] Step 4.1: Count the total number of pixels in the effective region of the structure surface image, denoted as N. total ;

[0040] Step 4.2: Count the number of pixels corresponding to each mineral, where the number of pixels corresponding to the i-th type of mineral is denoted as N. i ;

[0041] Step 4.3: Based on connected component analysis, spatial continuity is identified in the pixel distribution of various minerals. Spatially adjacent pixel sets belonging to the same mineral category are divided into independent regions, each defined as a mineral patch. The number of patch regions for the i-th mineral category is denoted as K. i ;

[0042] Step 4.4: Calculate the pixel area of ​​each patch region, where the pixel area of ​​the j-th patch region of the i-th mineral is denoted as A. ij ;

[0043] Step 4.5: In the patch regions of various minerals, identify the patch region with the largest area, and denote its pixel area as A. max,i .

[0044] Furthermore, according to the aforementioned method for quantitative characterization of the inhomogeneity of natural structural surfaces, step 5 specifically includes the following steps:

[0045] Step 5.1: Calculate the area ratio of each type of mineral on the structural surface, which is used to characterize the mineral composition ratio of each type of mineral on the structural surface. The specific calculation formula is as follows:

[0046]

[0047] Among them, P i N represents the area percentage of the i-th type of mineral; i This represents the number of pixels in the effective region of the i-th type of mineral in the structural surface image; This represents the total number of pixels in the effective region of the structural surface image;

[0048] Step 5.2: Calculate the percentage of the largest patchy area for each type of mineral, which characterizes the degree of concentration of minerals on the structural surface. The specific calculation formula is as follows:

[0049]

[0050] in, This represents the percentage of the largest patch area for mineral type i; This represents the area of ​​the patch with the largest area of ​​mineral type i;

[0051] Step 5.3: Calculate the scale discrete characteristic parameters of various mineral patch regions to characterize the degree of dispersion of mineral patch scale distribution;

[0052] Furthermore, according to the quantitative characterization method for the non-uniformity of natural structural surfaces, step 5.3, calculating the scale discrete characteristic parameters of various mineral patch regions, includes:

[0053] First, calculate the average area and standard deviation of various mineral patch regions. Then, calculate the coefficient of variation of the area using the following formula:

[0054] Average area:

[0055]

[0056] Area standard deviation:

[0057]

[0058] Area variation coefficient:

[0059]

[0060] in, A represents the average area of ​​the i-th type of mineral patch region; ij K represents the area of ​​the j-th patch region in the i-th mineral class; i σ represents the number of patchy regions identified for the i-th mineral class; i This represents the standard deviation of the area of ​​the i-th type of mineral patch. This represents the coefficient of variation of the area of ​​mineral patches of type i.

[0061] The area variation coefficient is used to eliminate the influence of dimensions and characterize the relative dispersion of patch size. The larger the value, the more significant the difference in patch size and the more uneven the spatial distribution of this type of mineral.

[0062] Furthermore, according to the quantitative characterization method for the non-uniformity of natural structural surfaces, in step 6:

[0063] First, a non-uniformity index of mineral composition in structural planes is constructed based on the area proportion of various minerals. This is used to characterize the degree of difference in the compositional proportions of different minerals on the structural plane. The specific calculation formula is as follows:

[0064]

[0065] Where m is the number of mineral categories identified in the structural plane; P i This represents the area percentage of the i-th type of mineral on the structural surface;

[0066] Subsequently, the spatial concentration of minerals was coupled with the patch-scale dispersion to construct a mineral spatial distribution non-uniformity index. The specific calculation formula is as follows:

[0067]

[0068] Among them, L i This represents the percentage of the largest patch area for mineral type i.

[0069] Based on this, a comprehensive index of mineral heterogeneity on structural surfaces is constructed using a product-type coupling function. The specific calculation formula is as follows:

[0070] .

[0071] The beneficial effects of adopting the above technical solution are as follows:

[0072] (1) The method of this application identifies and comprehensively analyzes the mineral composition and spatial distribution characteristics of the structural plane, constructs a multi-index evaluation system, and realizes the quantitative characterization of the non-uniformity characteristics of the structural plane;

[0073] (2) Compared with the traditional method of simplifying the structural plane into a homogeneous interface, the method of this application can characterize the differences in composition and spatial distribution of different minerals, effectively reflect the non-uniform characteristics inside the structural plane, and provide a more reasonable characterization basis for the analysis of the mechanical characteristics of the structural plane.

[0074] (3) Compared with the limitations of traditional mineralogical analysis methods that only focus on mineral composition, the method of this application can further characterize the spatial distribution characteristics of minerals on structural surfaces, and realize a holistic description of the distribution morphology and aggregation characteristics of minerals;

[0075] (4) The method of this application is based on image recognition to realize mineral region extraction and parameter calculation. The test process is relatively simple and the analysis efficiency is high. It is suitable for rapid quantitative analysis of structural surface samples in engineering.

[0076] (5) The quantitative characterization results of non-uniformity established by the method of this application can provide basic parameter support for the analysis of the mechanical behavior of structural surfaces and the evaluation of engineering stability, and have good engineering application value. Attached Figure Description

[0077] Figure 1 This is a flowchart illustrating the quantitative characterization method for the non-uniformity of natural structural surfaces in this embodiment.

[0078] Figure 2 This is a schematic diagram of the structural surface image acquisition in this embodiment;

[0079] Figure 3 This is a schematic diagram of the structural plane of this embodiment;

[0080] Figure 4 This is a schematic diagram of the mineral-color feature database construction process in this embodiment;

[0081] Figure 5 This is a schematic diagram of the mineral-by-mineral segmentation process for structural surfaces based on color features in this embodiment.

[0082] Figure 6 This is a schematic diagram of the mineral segmentation results and their spatial distribution areas in this embodiment, wherein (a) is a schematic diagram of the segmented area of ​​mineral A; (b) is a schematic diagram of the segmented area of ​​mineral B; and (c) is a schematic diagram of the segmented area of ​​mineral C.

[0083] Figure 7 This is a schematic diagram of the reconstructed mineral distribution on the structural plane in this embodiment;

[0084] Figure 8 This is a schematic diagram illustrating the process of constructing the comprehensive index of mineral heterogeneity on the structural plane in this embodiment;

[0085] The following are explanations of the labels in the attached diagram:

[0086] 1—Imaging equipment, 2—Photo studio, 3—Structural surface, 4—Stable light source, 5—Bearing platform, 6—Scale calibration reference object. Detailed Implementation

[0087] To facilitate understanding of this application, a more comprehensive description of this application will be provided below with reference to the accompanying drawings.

[0088] Figure 1 This is a schematic diagram of the overall process of the quantitative characterization method for the non-uniformity of natural structural surfaces in this embodiment, covering the main steps such as structural surface image acquisition, mineral identification and segmentation, spatial distribution feature extraction, and non-uniformity index construction.

[0089] This application first standardizes the acquisition of structural surface images. Based on this, according to the preset mineral color features, the pixels in the image are matched and judged one by one to identify various minerals and extract their spatial distribution information. Then, a multi-index comprehensive evaluation system is constructed to realize the quantitative characterization of the mineral composition of the structural surface and its spatial non-uniform distribution characteristics.

[0090] It should be noted that the structural plane referred to is the observation interface formed after the original structural planes in the rock mass are opened through shearing or splitting. This interface corresponds to the location of the structural plane itself and can reflect the mineral composition and spatial distribution characteristics inside the structural plane, rather than a newly formed random fracture surface.

[0091] This embodiment of the method for quantitative characterization of the non-uniformity of natural structural surfaces includes the following steps.

[0092] Step 1: Under unified and standardized image acquisition conditions, acquire structural surface images, and uniformly store and manage the acquired images to establish a structural surface image dataset for subsequent mineral identification and spatial distribution analysis;

[0093] Step 1.1: Establish unified and standardized image acquisition conditions, including: using the same model of imaging equipment; setting sufficient, stable and uniform lighting conditions; unifying shooting parameters, including exposure time, focal length, white balance and image resolution, and the image resolution should meet the accuracy requirements of mineral identification and spatial distribution analysis; and maintaining a consistent shooting distance H between the lens and the structural surface.

[0094] Step 1.2: Under the standardized image acquisition conditions described above, acquire high-resolution images of each structural surface. During the acquisition process, a scale calibration reference can be set on the structural surface support platform to assist in obtaining image scale information. (Reference) Figure 2 The diagram shown illustrates the image acquisition of the structural surface in this embodiment. Figure 3 This is a schematic diagram of a single structural plane.

[0095] Step 1.3: Number, archive, and standardize the acquired structural surface images, and record the corresponding sample number and acquisition parameter information to construct a structural surface image dataset for subsequent mineral identification, spatial distribution analysis, and calculation of structural surface non-uniformity characterization indicators.

[0096] Step 2: Based on the structural surface images in the structural surface image dataset, extract typical color feature parameters of various mineral regions, and perform parametric representation and vectorized modeling of the typical color features to construct a mineral-color feature database;

[0097] Step 2.1: Based on the color differences of different mineral regions in the structural surface image, extract typical color feature parameters of various mineral regions;

[0098] Step 2.2: Based on the extracted typical color feature parameters, construct the typical color feature vector M corresponding to each type of mineral. i Its expression is as follows:

[0099]

[0100] Among them, M i H represents the typical color feature vector of the i-th type of mineral; i S i B i These correspond to the hue, saturation, and brightness parameters extracted from the structural surface image of the i-th mineral region, respectively.

[0101] The typical color feature parameters are vectorized and then uniformly used in subsequent calculations in the form of color feature vectors.

[0102] Step 2.3: Set the typical color feature vectors of various minerals as the base colors for color matching, and save them as separate mineral color matching parameter files;

[0103] Step 2.4: Number all mineral color matching parameter files according to a preset order as i = 1,2,…,n, and store them uniformly to form a mineral-color feature database; (Refer to...) Figure 4 The diagram shows the process of constructing a mineral-color feature database.

[0104] Step 3: Obtain the structural surface image to be analyzed from the structural surface image dataset, and carry out mineral identification and spatial distribution analysis based on the image.

[0105] refer to Figure 5 The mineral-by-mineral color matching segmentation process shown loads the color matching parameter files of each mineral in the mineral-color feature database in sequence. It then performs mineral-by-mineral color matching and recognition on the structural surface image to be analyzed. Specifically, for a single mineral, it performs matching and judgment based on the distance calculation results between the mineral color feature vector and the pixel color feature vector, identifies and extracts the segmentation regions of various minerals, reconstructs the mineral distribution map of the structural surface, and then iteratively optimizes the segmentation parameters by comparing and verifying the reconstructed mineral distribution map of the structural surface with the original image until the comparison and verification results of the reconstructed mineral distribution map of the structural surface and the original image are consistent.

[0106] Step 3.1: Convert the original structural surface image to the preset HSB color space to improve the stability and distinguishability of different mineral colors;

[0107] Step 3.2: Load the color matching parameter file for each type of mineral sequentially from the mineral-color feature database, and then, based on the color feature vector M of the currently loaded i-th type of mineral... i The pixels in the structural surface image are matched and determined one by one.

[0108] For each pixel in the image, construct its color feature vector M. p And calculate its color feature vector M of the currently loaded i-th type of mineral. i The Euclidean distance D between them i :

[0109]

[0110] in These represent the hue, saturation, and brightness values ​​for each pixel.

[0111]

[0112] Set a uniform color tolerance threshold T, and define that when D is satisfied... i When < T, the pixel is determined to meet the color matching condition, belongs to the i-th type of mineral, and is marked as a pixel of that type of mineral; when If the pixel does not belong to the current mineral, it will not be marked.

[0113] This implementation method adopts a mineral-by-mineral traversal matching method, that is, pixel matching is performed only for the currently loaded single mineral each time.

[0114] Step 3.3: Based on the color matching results, pixels that match the color characteristics of this type of mineral are spatially aggregated to form the distribution area of ​​this type of mineral; the distribution area can be one or more spatially independent regions. Further image processing is performed on the distribution area to remove isolated noise pixels, ultimately obtaining the segmented region of this type of mineral.

[0115] Step 3.4: After completing the segmentation of all mineral categories, assign corresponding identifier colors to all segmented mineral regions according to their respective mineral categories, and maintain the relative spatial position of the segmented mineral regions in the structural surface image to generate a reconstructed mineral distribution map of the structural surface (e.g., Figure 7 (As shown). This figure illustrates the spatial distribution of each mineral type on the structural plane and provides visualization support for subsequent analysis. Reference Figure 6 The mineral segmentation results and their spatial distribution diagrams shown illustrate that different minerals are identified as multiple spatially independent patch regions on the structural surface, which can intuitively reflect the spatial distribution characteristics of various minerals. It should be noted that the color markings are used to distinguish different mineral categories and are only for visualization; they are not involved in mineral identification or subsequent parameter calculations.

[0116] Step 3.5: Compare and verify the reconstructed mineral distribution map generated in Step 3.4 with the original structural surface image to determine whether the location, extent, and boundary morphology of each mineral region correspond to the spatial distribution and boundary morphology of the original structural surface image. If it is found that similar mineral regions are over-segmented or different mineral regions are mixed, it is judged as over-segmentation or under-segmentation. Among them, the situation where similar mineral regions are divided into multiple discrete small regions is defined as over-segmentation, and the situation where different mineral regions are merged or mixed is defined as under-segmentation.

[0117] Step 3.6: Based on the comparison and verification results of Step 3.5, if it is determined to be an oversegmentation phenomenon, increase the color tolerance threshold to improve the connectivity of similar mineral regions; if it is an undersegmentation phenomenon, decrease the color tolerance threshold to enhance the ability to distinguish between different minerals.

[0118] In this embodiment, the method for increasing or decreasing the color tolerance threshold T is as follows: the color tolerance threshold is increased or decreased in a step-by-step iterative manner, with each increase or decrease being a preset small percentage of the current threshold, preferably 1% to 10%.

[0119] Step 3.7: Repeat steps 3.2 to 3.5 until the segmentation result is basically consistent with the mineral region in the original image in terms of spatial distribution and boundary morphology. The segmentation result is considered to meet the requirements.

[0120] Step 4: Based on the mineral segmentation region obtained after iterative optimization in Step 3, the pixel-spatial distribution information of each mineral in the structural surface image is statistically analyzed and extracted to obtain the mineral pixel-spatial distribution parameters;

[0121] Step 4.1: Count the total number of pixels in the effective region of the structure surface image, denoted as N. total ;

[0122] Step 4.2: Count the number of pixels corresponding to each mineral, where the number of pixels corresponding to the i-th type of mineral is denoted as N. i ;

[0123] Step 4.3: Based on connected component analysis, spatial continuity is identified in the pixel distribution of various minerals. Spatially adjacent pixel sets belonging to the same mineral category are divided into independent regions, each defined as a mineral patch. The number of patch regions for the i-th mineral category is denoted as K. i ;

[0124] Step 4.4: Calculate the pixel area of ​​each patch region, where the pixel area of ​​the j-th patch region of the i-th mineral is denoted as A. ij ;

[0125] Step 4.5: In the patch regions of various minerals, identify the patch region with the largest area, and denote its pixel area as A. max,i ;

[0126] The above parameters are used to characterize the spatial distribution of minerals on structural surfaces, providing basic data for subsequent heterogeneity evaluation.

[0127] Step 5: Based on the mineral pixel-spatial distribution parameters obtained in Step 4 (including N) total N i K i A ij and A max,i ), calculate the characteristic parameters used to characterize the mineral composition ratio, spatial concentration and patch-scale distribution characteristics, as the basis parameters for subsequent non-uniformity index calculation;

[0128] The area of ​​each patch region refers to the pixel area of ​​the corresponding region.

[0129] Step 5.1: Calculate the area ratio of each type of mineral on the structural surface, which is used to characterize the mineral composition ratio of each type of mineral on the structural surface. The specific calculation formula is as follows:

[0130]

[0131] Among them, P i N represents the area percentage of the i-th type of mineral; i This represents the number of pixels in the effective region of the i-th type of mineral in the structural surface image; This represents the total number of pixels in the effective region of the structural surface image;

[0132] Step 5.2: Calculate the percentage of the largest patchy area for each type of mineral, which characterizes the degree of concentration of minerals on the structural surface. The specific calculation formula is as follows:

[0133]

[0134] in, This represents the percentage of the largest patch area for mineral type i; This represents the area of ​​the patch with the largest area of ​​mineral type i;

[0135] Step 5.3: Calculate the scale dispersion characteristic parameters of various mineral patch regions to characterize the degree of dispersion in the scale distribution of mineral patches. First, calculate the average area and area standard deviation of various mineral patch regions. Then, further calculate the area variation coefficient. The specific calculation formula is as follows:

[0136] Average area:

[0137]

[0138] Area standard deviation:

[0139]

[0140] Area variation coefficient:

[0141]

[0142] in, A represents the average area of ​​the i-th type of mineral patch region; ij K represents the area of ​​the j-th patch region in the i-th mineral class; i σ represents the number of patchy regions identified for the i-th mineral class; i This represents the standard deviation of the area of ​​the i-th type of mineral patch. This represents the coefficient of variation of the area of ​​mineral patches of type i.

[0143] The area variation coefficient is used to eliminate the influence of dimensions and characterize the relative dispersion of patch size. The larger the value, the more significant the difference in patch size and the more uneven the spatial distribution.

[0144] Step 6: Based on the mineral spatial distribution characteristic parameters obtained in Step 5, construct the mineral composition heterogeneity index and the mineral spatial distribution heterogeneity index, and further establish a comprehensive index for mineral heterogeneity on structural surfaces, forming a two-dimensional heterogeneity evaluation system of "mineral composition-spatial distribution". The construction process is as follows: Figure 8 As shown.

[0145] First, a mineral composition non-uniformity index is constructed based on the area proportion of various minerals on the structural surface to characterize the degree of difference in the compositional proportion of different minerals on the structural surface. The specific calculation formula is as follows:

[0146]

[0147] in, The structural plane represents the non-uniformity index of mineral composition; m is the number of mineral categories identified in the structural plane; P i denoted as the area percentage of the i-th type of mineral on the structural surface.

[0148] Subsequently, the spatial concentration of minerals is coupled with the dispersion at the patch scale to construct a mineral spatial distribution non-uniformity index, the specific calculation formula of which is as follows:

[0149]

[0150] in, L is the index of non-uniformity in the spatial distribution of minerals on structural planes. i Indicates the percentage of the largest patchy area of ​​mineral type i; CV i denoted as the area variation coefficient of the i-th type of mineral patch region.

[0151] Based on this, a product-type coupling function is used to construct a comprehensive index of mineral heterogeneity on the structural surface. The specific calculation formula is as follows:

[0152]

[0153] Among them, I H This represents the comprehensive index of mineral heterogeneity on structural planes. I represents the index of non-uniformity in mineral composition of structural planes. S This indicates the non-uniformity index of the spatial distribution of minerals on structural planes.

[0154] This coupling method allows both mineral composition differences and spatial distribution differences to participate in the evaluation. When only a single dimension of non-uniformity exists, the comprehensive index response is limited; when both types of non-uniformity are enhanced simultaneously, the comprehensive index shows a synergistic increasing trend, thereby avoiding a single factor dominating the evaluation results and achieving a comprehensive quantitative characterization of mineral non-uniformity on structural surfaces.

[0155] Since the mineral heterogeneity of structural planes affects the mechanical response characteristics of structural planes, a comprehensive index I based on the mineral heterogeneity of structural planes is used. H Furthermore, a structural plane stability influence coefficient can be constructed, which can serve as an auxiliary evaluation parameter to reflect the degree of influence of mineral heterogeneity on the mechanical stability of the structural plane. Specifically, after obtaining the comprehensive index I of mineral heterogeneity of the structural plane... H Then, the structural surface stability influence coefficient can be constructed, and its calculation formula is as follows:

[0156]

[0157] Where η is the structural surface stability influence coefficient; I H is the comprehensive index of mineral heterogeneity of the structural plane; k is the sensitivity coefficient of heterogeneity, used to characterize the sensitivity of the degree of influence of mineral heterogeneity on the stability of the structural plane. The sensitivity coefficient k can be determined based on the lithological assemblage of the structural plane, the differences in mineral mechanical properties, or engineering experience, or it can be calibrated through structural plane shear tests or numerical simulation analysis.

[0158] It should be noted that the structural surface stability influence coefficient η can be constructed using a linear or nonlinear function; in this embodiment, a linear relationship is used as an example. The structural surface stability influence coefficient η can be used as a non-uniformity correction parameter in structural surface stability analysis. The value of η characterizes the degree of influence of mineral non-uniformity on structural surface stability; the larger the value, the more significant the influence of mineral non-uniform distribution on the mechanical response of the structural surface.

[0159] Through the above steps, the present invention achieves quantitative identification and comprehensive characterization of the mineral composition and spatial distribution characteristics of structural planes, and provides non-uniformity parameter support for structural plane stability analysis.

[0160] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the present invention. Various modifications or equivalent substitutions made by those skilled in the art without departing from the technical concept of the present invention should fall within the scope of protection of the present invention.

Claims

1. A method for quantitatively characterizing the non-uniformity of natural structural surfaces, characterized in that, The method includes the following steps: Step 1: Acquire structural surface images and establish a structural surface image dataset; Step 2: Based on the structural surface images in the structural surface image dataset, extract typical color feature parameters of various mineral regions, and perform parametric representation and vectorized modeling of the typical color features to construct a mineral-color feature database; Step 3: Obtain the structural surface image to be analyzed from the structural surface image dataset. Based on the mineral-color feature database, perform mineral-by-mineral color matching and recognition on the structural surface image to be analyzed, identify and extract the segmentation regions of various minerals, reconstruct the mineral distribution map of the structural surface, and iteratively optimize the segmentation parameters by comparing and verifying the reconstructed mineral distribution map of the structural surface with the original image until the comparison and verification results of the reconstructed mineral distribution map of the structural surface and the original image are consistent. Step 4: Based on the segmented regions of each mineral, the pixel-spatial distribution information of each mineral in the structural surface image to be analyzed is statistically analyzed and extracted to obtain the mineral pixel-spatial distribution parameters; Step 5: Based on the mineral pixel-spatial distribution parameters, calculate the mineral spatial distribution characteristic parameters used to characterize the mineral composition ratio, spatial concentration degree, and patch-scale distribution characteristics; Step 6: Based on the spatial distribution characteristic parameters of minerals, construct the mineral composition non-uniformity index and the mineral spatial distribution non-uniformity index, and then establish a comprehensive index of mineral non-uniformity of structural surfaces.

2. The method for quantitative characterization of the non-uniformity of natural structural surfaces according to claim 1, characterized in that, Step 2 specifically includes the following steps: Step 2.1: Based on the color differences of different mineral regions in the structural surface image, extract typical color feature parameters of various mineral regions; Step 2.2: Based on the extracted typical color feature parameters, construct the typical color feature vector M corresponding to each type of mineral. i M i The expression is as follows: ； Among them, M i H represents the typical color feature vector of the i-th type of mineral; i S i B i These correspond to the hue, saturation, and brightness parameters extracted from the structural surface image of the i-th mineral region, respectively. Step 2.3: Set the typical color feature vectors of various minerals as the base colors for color matching, and save them as separate mineral color matching parameter files; Step 2.4: Number and store all mineral color matching parameter files in a preset order to form a mineral-color feature database.

3. The method for quantitative characterization of the non-uniformity of natural structural surfaces according to claim 2, characterized in that, Step 3, which describes performing mineral-by-mineral color matching and recognition on the structural surface image to be analyzed based on the mineral-color feature database, includes: sequentially loading the color matching parameter files of each mineral in the mineral-color feature database, and then, for a single mineral, performing a matching determination based on the distance calculation results between the mineral color feature vector and the pixel color feature vector.

4. The method for quantitative characterization of the non-uniformity of natural structural surfaces according to claim 3, characterized in that, Step 3 specifically includes the following steps: Step 3.1: Convert the original structural surface image to a preset color space; Step 3.2: Load the color matching parameter file for each type of mineral sequentially from the mineral-color feature database, and then, based on the color feature vector M of the currently loaded i-th type of mineral... i The pixels in the structural surface image are matched one by one with the color tolerance threshold T. Step 3.3: Based on the color matching results, pixels that match the color characteristics of a type of mineral are spatially aggregated to form the distribution area of ​​that type of mineral. Then, the distribution area is processed to remove isolated noise pixels, and finally the segmented area of ​​that type of mineral is obtained. Step 3.4: After completing the region segmentation processing corresponding to all mineral categories, assign the corresponding color to all mineral segmentation regions according to their corresponding mineral categories, and maintain the relative spatial position relationship of the mineral segmentation regions in the structural surface image to generate a structural surface mineral distribution reconstruction map. Different minerals are identified as multiple spatially independent patch regions on the structural surface, so that the spatial distribution characteristics of various minerals can be intuitively reflected. Step 3.5: Compare and verify the mineral distribution reconstruction map generated in Step 3.4 with the original structural surface image. By comparing and verifying whether the spatial distribution and boundary morphology of each mineral region are basically consistent with the corresponding mineral region in the original structural surface image, it is determined whether there is oversegmentation or undersegmentation. Step 3.6: Based on the comparison and verification results in Step 3.5, adjust the color tolerance threshold T: if oversegmentation is determined, increase the color tolerance threshold T to improve the connectivity of similar mineral regions; if undersegmentation is determined, decrease the color tolerance threshold T to enhance the ability to distinguish between different minerals. The method for increasing or decreasing the color tolerance threshold T is as follows: the color tolerance threshold is increased or decreased step by step in an iterative manner, with each increase or decrease being a preset small percentage of the current threshold, preferably 1% to 10%; Step 3.7: Repeat steps 3.2 to 3.5 until the mineral distribution reconstruction map and the mineral regions in the original structural surface image are basically consistent in terms of spatial distribution and boundary morphology.

5. The method for quantitative characterization of the non-uniformity of natural structural surfaces according to claim 4, characterized in that, Step 3.2 describes using the color feature vector M of the currently loaded i-th type of mineral. i The process involves matching and determining each pixel in the structural surface image, including: For each pixel in the image, construct its color feature vector M. p And calculate its color feature vector M of the currently loaded i-th type of mineral. i The Euclidean distance D between them i : ； in These are the hue, saturation, and brightness values ​​for each pixel; ； Set a uniform color tolerance threshold T, and define that when D is satisfied... i When < T, the pixel is determined to meet the color matching condition, belongs to the i-th type of mineral, and is marked as a pixel of that type of mineral; when If the pixel does not belong to the current mineral, it will not be marked.

6. The method for quantitative characterization of the non-uniformity of natural structural surfaces according to claim 4, characterized in that, Step 4 specifically includes the following steps: Step 4.1: Count the total number of pixels in the effective region of the structure surface image, denoted as N. total ; Step 4.2: Count the number of pixels corresponding to each mineral, where the number of pixels corresponding to the i-th type of mineral is denoted as N. i ; Step 4.3: Based on connected component analysis, spatial continuity is identified in the pixel distribution of various minerals. Spatially adjacent pixel sets belonging to the same mineral category are divided into independent regions, each defined as a mineral patch. The number of patch regions for the i-th mineral category is denoted as K. i ; Step 4.4: Calculate the pixel area of ​​each patch region, where the pixel area of ​​the j-th patch region of the i-th mineral is denoted as A. ij ; Step 4.5: In the patch regions of various minerals, identify the patch region with the largest area, and denote its pixel area as A. max,i .

7. The method for quantitative characterization of the non-uniformity of natural structural surfaces according to claim 1, characterized in that, Step 5 specifically includes the following steps: Step 5.1: Calculate the area ratio of each type of mineral on the structural surface, which is used to characterize the mineral composition ratio of each type of mineral on the structural surface. The specific calculation formula is as follows: ； Among them, P i N represents the area percentage of the i-th type of mineral; i This represents the number of pixels in the effective region of the i-th type of mineral in the structural surface image; This represents the total number of pixels in the effective region of the structural surface image; Step 5.2: Calculate the percentage of the largest patchy area for each type of mineral, which characterizes the degree of concentration of minerals on the structural surface. The specific calculation formula is as follows: ； in, This represents the percentage of the largest patch area for mineral type i; This represents the area of ​​the patch with the largest area of ​​mineral type i; Step 5.3: Calculate the scale discrete characteristic parameters of various mineral patch regions to characterize the degree of dispersion of mineral patch scale distribution.

8. The method for quantitative characterization of the non-uniformity of natural structural surfaces according to claim 7, characterized in that, Step 5.3, which involves calculating the scale-discrete characteristic parameters of various mineral patch regions, includes: First, calculate the average area and standard deviation of various mineral patch regions. Then, calculate the coefficient of variation of the area using the following formula: Average area: ； Area standard deviation: ； Area variation coefficient: ； in, A represents the average area of ​​the i-th type of mineral patch region; ij K represents the area of ​​the j-th patch region in the i-th mineral class; i σ represents the number of patchy regions identified for the i-th mineral class; i This represents the standard deviation of the area of ​​the i-th type of mineral patch. The area variation coefficient represents the area variation coefficient of mineral patch region of type i; the area variation coefficient is used to eliminate the influence of dimensions and characterize the relative dispersion of patch size. The larger the value, the more significant the difference in patch size and the more uneven the spatial distribution.

9. The method for quantitative characterization of the non-uniformity of natural structural surfaces according to claim 8, characterized in that, In step 6: First, a non-uniformity index of mineral composition in structural planes is constructed based on the area proportion of various minerals. This is used to characterize the degree of difference in the compositional proportions of different minerals on the structural plane. The specific calculation formula is as follows: ； Where m is the number of mineral categories identified in the structural plane; P i This represents the area percentage of the i-th type of mineral on the structural surface; Subsequently, the spatial concentration of minerals was coupled with the patch-scale dispersion to construct a mineral spatial distribution non-uniformity index. The specific calculation formula is as follows: ； Among them, L i This represents the proportion of the largest patchy area of ​​the i-th mineral; based on this, a product-type coupling function is used to construct a comprehensive index of mineral heterogeneity on the structural surface. The specific calculation formula is as follows: 。

Images