Rock component modeling method based on image recognition

Abstract

The invention discloses a rock component modeling method based on image recognition, and belongs to the technical field of numerical simulation test, and the method comprises the following steps: S1, image gray processing and coordinate mapping; s2, carrying out gray scale statistics and threshold backstepping; s3, data import corresponds to the model; and S4, mineral grouping and attribute assignment. By fusing the XRD test result and the digital image processing technology, the mineral composition information of granite is accurately obtained, a modeling system conforming to actual rock microscopic grain morphology and distribution characteristics is constructed in discrete element software, the precision and calculation efficiency of a numerical model can be effectively improved, and the method is suitable for large-scale popularization and application. And the visualization capability of the rock fracture expansion process is enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of numerical simulation experiments, in particular to a rock component modeling method based on image recognition. BACKGROUND

[0002] In the field of rock mechanics and microstructure modeling, numerical modeling of samples is often dependent on mineral identification methods. However, existing methods lack correlation with real mineral composition and cannot accurately map image pixels to mineral categories. SUMMARY

[0003] In view of the technical problems in the background art, the present application provides a rock component modeling method based on image recognition, which combines XRD mineral identification and image grayscale identification for granite modeling, and combines PFC software for particle modeling, effectively improving the accuracy and automation level of microstructure modeling.

[0004] In a first aspect, the embodiments of the present application provide a modeling method of a rock component analysis model based on image recognition, including the following steps:

[0005] Obtain image data of a rock and rock component data of the rock;

[0006] Perform grayscale processing on the image data and obtain grayscale data corresponding to each pixel in the image data;

[0007] Derive grayscale threshold intervals corresponding to various components in the rock according to the rock component data and the grayscale data corresponding to each pixel;

[0008] Establish a discrete element model and associate the grayscale data corresponding to each pixel with spherical particles in the discrete element model;

[0009] Group the spherical particles in the discrete model based on the grayscale threshold intervals corresponding to various components, thereby establishing a rock component analysis model.

[0010] Further, in the present embodiment, the rock component data includes the rock components and rock component proportions, which are obtained based on X-ray diffraction experiments.

[0011] Further, in the present embodiment, deriving the grayscale threshold intervals corresponding to various components in the rock according to the rock component data and the grayscale data corresponding to each pixel includes:

[0012] Construct a histogram according to the grayscale data corresponding to each pixel, and calculate a cumulative distribution function by item-by-item accumulation;

[0013] The cumulative distribution function is normalized and matched with the rock component data to derive the gray threshold interval corresponding to each component in the rock.

[0014] Further, in the embodiment, the associating of the gray data of each pixel with the spherical particles in the discrete element model comprises:

[0015] projecting each pixel in the image data into the discrete element model through spatial mapping;

[0016] accumulating the gray data of each pixel into the spherical particles in the discrete element model based on a neighborhood search algorithm, and calculating the average gray value of the spherical particles according to the hit times of the spherical particles;

[0017] The average gray value is used for matching the gray threshold interval corresponding to each component by the spherical particles.

[0018] Further, in the embodiment, the projecting of each pixel in the image data into the discrete element model through spatial mapping comprises:

[0019] constructing a coordinate system of the image data, converting the position of each pixel in the image data into a coordinate in the coordinate system, and combining the coordinate of each pixel and the gray data corresponding to each pixel to form a triple data (X, Y, gray);

[0020] aligning the coordinate system with the discrete element model in coordinates, and mapping the triple data into the discrete element model.

[0021] Further, in the embodiment, the establishing of the rock component analysis model comprises:

[0022] matching the physical parameters of rock components according to the rock component data of the rock, and establishing the rock component analysis model according to the grouping of the spherical particles in the discrete model.

[0023] Further, in the embodiment, the image data of the rock is acquired by a camera.

[0024] In a second aspect, the embodiments of the present application provide a computer readable storage medium, which is a non-volatile storage medium or a non-transitory storage medium, and has a computer program stored thereon, the computer program being run by a processor to perform the steps of the modeling method of the rock component analysis model based on image recognition.

[0025] In a third aspect, an embodiment of the present application provides a computer program product, comprising computer programs / instructions, which, when executed by a processor, implement the steps of the modeling method of the rock component analysis model based on image recognition according to any one of the above.

[0026] In a fourth aspect, an embodiment of the present application provides a rock component analysis model based on image recognition, comprising a memory and a processor, wherein the memory stores a computer program capable of running on the processor, and the processor executes the steps of the modeling method of the rock component analysis model based on image recognition according to any one of the above when running the computer program.

[0027] Beneficial effects: The present application combines the X-ray diffraction experimental results and the image gray distribution information to establish a rock component modeling method based on image recognition, combines the spatial coordinate mapping and the gray value assignment method in the discrete element method, and realizes the automatic grouping of the rock modeling at the particle level. The method can not only reflect the actual microstructure characteristics of the multi-mineral rock, but also has good adaptability and expansibility, and can be widely applied to numerical simulation under different rock types and engineering backgrounds. Compared with the traditional modeling method which depends on artificial grouping and assumption parameter setting, the present application improves the intelligent level and structure restoration degree of model construction, and provides a more real and reliable numerical basis for rock multi-scale fracture evolution simulation and rock mass engineering response analysis.

[0028] The above description is only a summary of the technical solutions of the present application, in order to enable the technical means of the present application to be implemented according to the content of the description, and in order to enable the above and other purposes, characteristics and advantages of the present application to be more obvious and easy to understand, the following specific embodiments of the present application are described. BRIEF DESCRIPTION OF DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the present application, the following will briefly introduce the drawings used in the present application. Obviously, the drawings described below are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0030] Figure 1 A flowchart of the rock component modeling method based on image recognition is provided for the embodiments of the present application. DETAILED DESCRIPTION

[0031] The embodiments of the technical solutions of the present application will be described in detail below with reference to the drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present application, and therefore only serve as examples, and cannot limit the protection scope of the present application.

[0032] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of this application; the use of the terms "including," "comprising," or "having" and variations thereof herein is intended to be broad and encompass the terms "consisting of" and "consisting essentially of" and variations thereof. Unless otherwise required by context, singular terms shall include pluralities and vice versa. Unless otherwise required by context, the use herein of the singular is also to be construed as a use of the plural and vice versa.

[0033] In the description of the embodiments of the present application, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is more than two, unless otherwise explicitly and specifically limited.

[0034] Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearance of the phrase in various places in the specification is not necessarily all referring to the same embodiment, or to a particular embodiment, or to a particular set of embodiments, and is not necessarily mutually exclusive of other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0035] In the description of the embodiments of the present application, the term "and / or" is only a description of the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent the three cases of A existing alone, A and B existing together, and B existing alone. In addition, the character " / " in this paper generally represents that the front and rear associated objects are a "or" relationship.

[0036] In the description of the embodiments of the present application, the term "multiple" refers to more than two (including two), and similarly, "multiple groups" refers to more than two groups (including two groups), and "multiple pieces" refers to more than two pieces (including two pieces).

[0037] In the description of the embodiments of the present application, the technical terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the embodiments of the present application and simplifying the description, and therefore cannot be understood as limiting the embodiments of the present application, which do not indicate or imply that the indicated devices or elements must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the embodiments of the present application.

[0038] In the description of the embodiments of the present application, unless otherwise explicitly specified and limited, the technical terms "mounting", "connection", "connecting", "fixing" and the like should be understood in a broad sense, for example, can be fixed connection, can also be detachable connection, or integral; can be mechanical connection, can also be electrical connection; can be direct connection, can also be indirect connection through an intermediate medium, can be internal communication of two elements or interaction relationship between two elements. For those skilled in the art, the specific meanings of the above terms in the embodiments of the present application can be understood according to the specific circumstances.

[0039] In the field of rock mechanics and microstructure modeling, numerical modeling of samples is often performed by relying on mineral identification means. However, the existing method lacks association with real mineral composition and cannot achieve accurate mapping of image pixels to mineral categories.

[0040] In order to solve the technical problems that the existing method lacks association with real mineral composition and cannot achieve accurate mapping of image pixels to mineral categories, the present application provides a rock component modeling method based on image recognition. The rock component modeling method based on image recognition fuses a granite modeling method of XRD mineral identification and image gray identification, and combines PFC software for particle modeling, effectively improving the accuracy and automation level of microstructure modeling.

[0041] The embodiment of the present application provides a rock component modeling method based on image recognition, which specifically comprises the following steps:

[0042] S1, obtaining image data of a rock and rock component data of the rock;

[0043] S2, performing gray scale processing on the image data, and obtaining gray scale data corresponding to each pixel in the image data;

[0044] S3, deriving gray scale threshold intervals corresponding to various components in the rock according to the rock component data and the gray scale data corresponding to each pixel;

[0045] S4, establishing a discrete element model, and associating the gray scale data corresponding to each pixel with spherical particles in the discrete element model;

[0046] S5, grouping the spherical particles in the discrete model based on the gray scale threshold intervals corresponding to various components, so as to establish a rock component analysis model.

[0047] Specifically, in the embodiment, the rock sample is subjected to image acquisition and component analysis by the image acquisition device and the rock component analysis device, so as to obtain image data and rock component data of the rock sample, the image data is subjected to gray scale processing, and gray scale data corresponding to each pixel in the image data is obtained; the gray scale threshold interval corresponding to each component in the rock is derived according to the rock component data and the gray scale data corresponding to each pixel; a discrete element model is established, and the gray scale data corresponding to each pixel is associated with the spherical particles in the discrete element model; the spherical particles in the discrete model are grouped based on the gray scale threshold interval corresponding to each component, so as to establish a rock component analysis model; in the embodiment, an image recognition mechanism for automatically deducing the gray scale threshold is established by combining the image data and the rock component data of the rock, the spatial coordinate mapping and the gray scale assignment method in the discrete element method are combined, the rock modeling automatic grouping at the particle level is realized, the actual microstructure characteristics of the multi-mineral rock such as granite can be reflected, the adaptability and expansibility are good, and the method can be widely applied to numerical simulation under different rock types and engineering backgrounds, and provides a more real and reliable numerical basis for rock multi-scale fracture evolution simulation and rock mass engineering response analysis.

[0048] As shown in Figure 1 Figure 1 A flowchart of a rock component modeling method based on image recognition is provided for the embodiments of the application, and specifically includes the following steps:

[0049] S1, obtaining image data of a rock and rock component data of the rock;

[0050] For example, in the embodiment, a high-definition digital camera is used to take a picture of the rock sample, so as to obtain high-definition rock image data;

[0051] When the rock component data of the rock is obtained, the components of the rock sample can be analyzed by X-ray diffraction analysis experiment (XRD), so as to obtain rock component data such as rock components and rock component proportions of the rock sample. For example, after being analyzed by X-ray diffraction analysis experiment (XRD), it is confirmed that the rock sample is mainly composed of Mica, Quartz and Feldspar, wherein the proportion of Mica is 6.6%, the proportion of Quartz is 23.8%, and the proportion of Feldspar is 69.6%.

[0052] S2, the image data is subjected to gray scale processing, and gray scale data corresponding to each pixel in the image data is obtained;

[0053] ​Exemplarily, the image data acquired by the high-definition digital camera is generally a color image, and in the embodiment, the image data is mapped from the RGB color space to a single-channel gray value by the average method or weighted method, and then the image is grayed, and the pixels of the grayed image are split to obtain the gray data corresponding to each pixel in the image data.

[0054] S3, deriving the gray threshold interval corresponding to each component in the rock according to the rock component data and the gray data corresponding to each pixel;

[0055] Exemplarily, in the embodiment, the gray data corresponding to each pixel in the image data is counted to construct a complete gray histogram with a gray value range of 0-255, so as to reflect the number of pixels corresponding to each gray level.

[0056] The above gray histogram is accumulated item by item to obtain a cumulative distribution function (CDF), which represents the proportion of the number of pixels with a gray value less than or equal to a certain value in the total number of pixels; and in order to facilitate calculation and demarcation, the CDF is normalized to limit its value range to 0-1.

[0057] The cumulative distribution function is matched with the rock component data, the position where the cumulative distribution function is first greater than or equal to the target proportion is searched, the gray value corresponding to the position is determined as the demarcation threshold, and then the gray threshold interval corresponding to each component in the rock is derived by back deduction.

[0058] S4, establishing a discrete element model and associating the gray data corresponding to each pixel with the spherical particles in the discrete element model;

[0059] Exemplarily, in the embodiment, step S4 specifically includes the following steps:

[0060] S41, establishing a discrete element model, which can be established by means of PFC (Particle Flow Code) software in the embodiment. In the PFC modeling environment, the spatial range of the modeling domain and the particle parameters are defined, the random particles are set, the model boundary wall is generated, and the spherical particles are distributed in the specified box region according to the target porosity and the particle radius range; the density and damping coefficient are assigned to all spherical particles, the material constitutive relationship is initialized as a linear contact model, the Young's modulus, the normal and tangential stiffness ratio are set, and then the discrete element model is established.

[0061] S42, projecting each pixel in the image data into the discrete element model by spatial mapping;

[0062] Specifically, the image resolution is acquired to determine the height and width of the image, an X-Y physical space coordinate system of the image data is constructed with the center of the image as the coordinate origin, and the positions of the pixels in the image data and the number of image pixels are generated into corresponding coordinate points, and then the corresponding coordinates of each pixel point of the image in the two-dimensional space are obtained, and the coordinates of each pixel and the corresponding gray data of each pixel are combined to form a three-tuple data (X, Y, gray);

[0063] The coordinate system is aligned with the discrete element model in terms of coordinates, so as to map the three-tuple data into the discrete element model, and ensure that the physical coordinates of the image pixels are consistent with the spatial range of the discrete element model, avoiding data misplacement caused by errors in the coordinate system or scale.

[0064] S43, based on the neighborhood search algorithm, the gray data of each pixel is accumulated into the spherical particles in the discrete element model, and the average gray value of the gray threshold interval corresponding to the matching of various components by the spherical particles is calculated according to the hit number of the spherical particles;

[0065] Specifically, after each pixel in the image data is projected into the discrete element model by spatial mapping, the spherical particles are matched based on spatial proximity, each coordinate point of each pixel is associated with the nearest spherical particle, and when the spherical particle is matched with the pixel point, the gray value of the pixel point is matched into the spherical particle, and the matching number of the spherical particle is recorded. The average gray value of each spherical particle is recorded by using the gray value accumulation and counting method.

[0066] S5, based on the gray threshold interval corresponding to various components, the spherical particles in the discrete model are grouped to establish a rock component analysis model.

[0067] Specifically, in this embodiment, according to the gray threshold interval obtained by the foregoing back calculation, the spherical particles in different gray ranges are divided into corresponding mineral category groups and marked for grouping, and the physical parameters of the rock components are matched according to the rock component data of the rock, so as to assign different contact models, friction coefficients, fracture strengths and other physical parameters to different spherical particles, thereby establishing a rock component analysis model.

[0068] For example, this embodiment combines Python code to explain in detail the specific implementation of the rock image gray value extraction and coordinate mapping process in the present application:

[0069] The image processing library OpenCV (cv2), the scientific computing library NumPy (np), and the data structure library Pandas (pd) are imported to provide a basis for subsequent image reading, matrix operation, and data export;

[0070] Read an image called granite.jpg of granite by the cv2.imread() function, and specify to import in grayscale mode, convert to single-channel grayscale image;

[0071] Get the height and width of the image for subsequent coordinate system construction;

[0072] Define the actual range of the image mapping to the physical space, where the X-axis range is ±0.025 and the Y-axis range is ±0.5, and the unit length can be adjusted according to the actual modeling size;

[0073] Generate corresponding coordinate points in X and Y directions according to the number of image pixels, so that each pixel point of the image has corresponding coordinates in two-dimensional space;

[0074] Use np.meshgrid() to build X and Y coordinate matrices, and convert the image grayscale value to a floating-point Z matrix representing grayscale intensity;

[0075] Expand the three-dimensional data (X, Y, grayscale Z) into one dimension and build it into a structured DataFrame format;

[0076] Export the processed data to an Excel file as an input file for subsequent import into PFC software. The final output format is three columns (X, Y, grayscale), with each row representing the physical location and grayscale value of a pixel point in the modeling space;

[0077] In the PFC modeling environment, first define the spatial range of the modeling domain and the particle parameters. Set the model geometry range by the following FISH script, define the upper and lower limits of the particle radius and the target porosity;

[0078] Set the random seed and generate the model boundary wall, and distribute spherical particles in the specified box area according to the target porosity and particle radius range;

[0079] Assign density and damping coefficient to all spheres, and initialize the material constitutive relationship to linear contact model, set Young's modulus and normal to shear stiffness ratio;

[0080] Perform preliminary relaxation of the system through a short loop (calm), then use the density scaling method to reduce the time step to accelerate convergence, and finally use solve aratio 1e-4 to seek the equilibrium state;

[0081] To ensure good contact between particles in the subsequent process, call the function to slightly enlarge the "suspended particles" with contact number less than 3;

[0082] In PFC software, read the above file content, and perform spatial matching and accumulation of grayscale values;

[0083] The X, Y coordinates and gray value of each line are extracted from the file, and the ball.near() function is called to search for the nearest ball near the coordinate position;

[0084] If the search is successful, the gray value is accumulated in ball.extra(1) and the count is recorded in ball.extra(2);

[0085] All balls are traversed, and the average gray value is calculated and saved in ball.extra(3);

[0086] The average gray value is classified using the dynamic gray threshold value obtained from the XRD composition and image CDF in step two. According to the interval to which the gray value belongs, the balls are automatically grouped;

[0087] The average gray value in the first interval (e.g. 0-A) is grouped as "Mica";

[0088] The average gray value in the second interval (e.g. A-B) is grouped as "Quartz";

[0089] The average gray value in the third interval (e.g. B-255) is grouped as "Feldspar";

[0090] The ball.group() function is called to add a group label to each particle, establishing the spatial distribution identification of mineral components in the model;

[0091] Subsequent different physical parameters such as contact model, friction coefficient, and fracture strength can be assigned to different particles according to the grouping information, for simulating the rock mechanical response behavior under different mineral compositions, including crack propagation, stress concentration, and fracture process.

[0092] In a second aspect, the embodiments of the present application provide a computer readable storage medium, which is a non-volatile storage medium or a non-transitory storage medium, and has stored thereon a computer program. The computer program is run by a processor to perform the steps of the modeling method of the rock component analysis model based on image recognition according to any one of the above aspects.

[0093] In a third aspect, the embodiments of the present application provide a computer program product, which includes computer programs / instructions. The computer programs / instructions are executed by a processor to implement the steps of the modeling method of the rock component analysis model based on image recognition according to any one of the above aspects.

[0094] In a fourth aspect, the embodiments of the present application provide a rock component analysis model based on image recognition, comprising a memory and a processor, the memory storing a computer program capable of running on the processor, and the processor executing the steps of the modeling method of the rock component analysis model based on image recognition of any one of the above aspects when running the computer program.

[0095] It should be noted that the present application is not limited to the above-described embodiments. The above-described embodiments are merely examples, and embodiments having substantially the same configuration and exerting the same effects as the technical idea within the scope of the technical solutions of the present application are all included in the technical scope of the present application. Furthermore, within the scope of the gist of the present application, various modifications that can be thought of by those skilled in the art, other modes constructed by combining part of the elements of the embodiments are also included in the scope of the present application.

Claims

1. A modeling method for a rock component analysis model based on image recognition, characterized in that: The following steps are involved: Acquiring image data of the rock and rock composition data of the rock; Performing grayscale processing on the image data and obtaining grayscale data corresponding to each pixel in the image data; Derived grayscale threshold intervals corresponding to various components in the rock according to the rock component data and the grayscale data corresponding to each pixel; Establishing a discrete element model, and associating the grayscale data corresponding to each pixel with the spherical particles in the discrete element model; The spherical particles in the discrete element model are grouped based on the grayscale threshold intervals corresponding to various components, thereby establishing a rock component analysis model.

2. The modeling method of the rock component analysis model based on image recognition according to claim 1 is characterized in that: The rock component data includes the rock components and rock component ratios, and the rock components and rock component ratios are obtained based on X-ray diffraction experiments.

3. The modeling method of the rock component analysis model based on image recognition according to claim 1 is characterized in that: The grayscale threshold intervals corresponding to various components in the rock are derived based on the rock component data and the grayscale data corresponding to each pixel, including: Constructing a histogram according to the grayscale data corresponding to each pixel, and calculating a cumulative distribution function by accumulating item by item; The cumulative distribution function is normalized and matched with rock component data to derive grayscale threshold intervals corresponding to various components in the rock.

4. The modeling method of the rock component analysis model based on image recognition according to claim 1 is characterized in that: The associating the grayscale data corresponding to each pixel with the spherical particles in the discrete element model includes: Projecting each pixel in the image data into the discrete element model by spatial mapping; Accumulating the grayscale data of each pixel into the spherical particles in the discrete element model based on a neighborhood search algorithm, and calculating the average grayscale value of the spherical particles according to the number of hits of the spherical particles; The average grayscale value is used to match the grayscale threshold intervals corresponding to various components of the spherical particles.

5. The modeling method of the rock component analysis model based on image recognition according to claim 4 is characterized in that: Projecting each pixel in the image data into the discrete element model by spatial mapping includes: Constructing a coordinate system for the image data, converting the position of each pixel in the image data into coordinates in the coordinate system, and combining the coordinates of each pixel and the grayscale data corresponding to each pixel to form a triplet of data (X, Y, grayscale); The coordinate system is aligned with the discrete element model, and the triplet data is mapped into the discrete element model.

6. The modeling method of the rock component analysis model based on image recognition according to claim 1, characterized in that: The establishment of the rock component analysis model comprises: The physical parameters of the rock components are matched according to the rock component data of the rock; and the rock component analysis model is established according to the grouping of the spherical particles in the discrete model.

7. The modeling method of the rock component analysis model based on image recognition according to claim 1 is characterized in that: Image data of the rock is acquired by a camera.

8. A computer-readable storage medium, wherein the computer-readable storage medium is a non-volatile storage medium or a non-transient storage medium, and a computer program is stored thereon, wherein: When the computer program is executed by a processor, the steps of the modeling method of the rock component analysis model based on image recognition according to any one of claims 1 to 7 are executed.

9. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the modeling method of the rock component analysis model based on image recognition according to any one of claims 1 to 7 are implemented.

10. A rock component analysis model based on image recognition, comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, characterized in that: When the processor runs the computer program, the processor executes the steps of the modeling method of the rock component analysis model based on image recognition according to any one of claims 1 to 7.

Images