A method for analyzing and identifying thin sections of shale rocks using optical microscopy and field emission electron microscopy.

Abstract

This invention relates to the field of materials analysis technology, specifically to a method for analyzing and identifying shale thin sections using a combination of optical microscopy and field emission scanning electron microscopy. First, from an optical perspective, the method combines the brightness distribution of bright-field images of shale thin sections with the birefringence deviation of cross-polarized images to determine mineral discrimination indices representing atomic number proxies. Then, based on the distribution of the mineral discrimination indices and the surface undulations of the shale thin sections, different biases are applied, and the positional bias vector is determined by combining this with the gray-level gradient distribution. Finally, based on the physical distance scale of the image and the pixel coordinate transformation process involving rotation and translation, the positional bias vector is used to determine the electron microscope stage coordinates corresponding to each pixel point for more accurate cross-scale positioning. This improves the accuracy of cross-scale positioning using optical microscopy and field emission scanning electron microscopy, significantly enhancing the reliability of shale thin section analysis and identification.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of material analysis, in particular to a shale rock thin section analysis and identification method using optical microscopy and field emission electron microscopy. BACKGROUND

[0002] The method for analyzing and identifying shale rock thin sections by using an optical microscope mainly observes the morphology, color, cleavage, interference color, extinction characteristics and birefringence of minerals. However, the resolution of the optical microscope is limited, and it is difficult to observe the nanoscale microstructure and fine mineral composition. In addition, different shale minerals may have similar optical characteristics, and it is difficult to accurately distinguish these minerals relying only on observation experience under the optical microscope. When analyzing and identifying shale rock thin sections by using a field emission scanning electron microscope, the microstructure and element distribution characteristics of the rock surface can be clearly presented, but the field emission scanning electron microscope cannot continuously and dynamically observe the same field of view, and the information obtained is relatively single, which cannot fully reflect the comprehensive characteristics of the rock.

[0003] In the combination of the optical microscope and the field emission scanning electron microscope, cross-scale positioning is required. The prior art uses a mechanical coordinate positioning method to perform cross-scale positioning of the optical microscope and the field emission scanning electron microscope. However, because of the hardness difference of the minerals, a reset error occurs, which causes the actual landing point of the same target area in the electron microscope field of view and the optical microscope mark position to deviate, and the deviation presents a nonlinear distribution with the changes of each layer in the shale. Therefore, the traditional mechanical coordinate positioning cannot be accurately compensated, resulting in the loss of key areas and affecting the accuracy of element analysis. That is, the accuracy of the cross-scale positioning of the optical microscope and the field emission scanning electron microscope by the mechanical coordinate positioning method is low, which reduces the reliability of the shale rock thin section analysis and identification. SUMMARY

[0004] In order to solve the technical problem of low accuracy of cross-scale positioning of the optical microscope and the field emission scanning electron microscope by the mechanical coordinate positioning method, the purpose of the present application is to provide a shale rock thin section analysis and identification method using optical microscopy and field emission electron microscopy, and the technical solution adopted is as follows:

[0005] The first aspect of the present application provides a shale rock thin section analysis and identification method using optical microscopy and field emission electron microscopy, comprising:

[0006] Collecting a bright field image and a cross-polarization image of the shale rock thin section under the optical microscope;

[0007] Determining a mineral discrimination index of each pixel point based on the brightness distribution of the bright field image and the birefringence deviation of the cross-polarization image;

[0008] Determine a position bias vector of each pixel point based on distribution of the mineral identification index of each pixel point, surface relief of the shale rock thin section, and gradient distribution trend of the grayscale;

[0009] Determine an electron microscope stage coordinate of each pixel point based on the pixel coordinate conversion based on the image physical distance scale and the rotation and translation processing, and the position bias vector of each pixel point; and perform shale rock thin section analysis and identification according to the electron microscope stage coordinate.

[0010] Further, the mineral identification index acquisition process includes:

[0011] Determine a birefringence proxy value of each pixel point according to an average wavelength of a light source, a thickness of the shale rock thin section, and a pixel value of each pixel point on the crossed polarized image;

[0012] Determine the corresponding mineral identification index according to a sum value between the normalized value of the grayscale value of each pixel point on the bright field image and the birefringence proxy value.

[0013] Further, the birefringence proxy value acquisition process includes:

[0014] Determine a correction parameter according to a ratio between the average wavelength of the light source and the thickness of the shale rock thin section; and normalize a product between the pixel value of each pixel point on the crossed polarized image and the correction parameter to determine the birefringence proxy value of each pixel point.

[0015] Further, the position bias vector acquisition process includes:

[0016] Obtain a maximum depth difference of the shale rock thin section by a white light interferometer; and determine a bias amount of each pixel point according to the overall distribution of the mineral identification index size of each pixel point in combination with the maximum depth difference.

[0017] Determine the position bias vector of each pixel point according to the gradient amplitude distribution of all pixel points in each gradient direction angle on the bright field image in combination with the bias amount.

[0018] Further, the bias amount acquisition process includes:

[0019] The mineral discrimination index of all pixel points is subjected to k-means clustering analysis to determine three pixel point clustering clusters; the K value of the k-means clustering analysis is set to 3; the pixel points in the pixel point clustering cluster with the maximum mean value of the mineral discrimination index of all pixel points are taken as high-hardness pixel points; the pixel points in the pixel point clustering cluster with the minimum mean value of the mineral discrimination index of all pixel points are taken as low-hardness pixel points; the other pixel points except the high-hardness pixel points and the low-hardness pixel points are taken as medium-hardness pixel points; the bias of all high-hardness pixel points is determined according to the product between the maximum depth difference and a preset first coefficient; the bias of all medium-hardness pixel points is determined according to the product between the maximum depth difference and a preset second coefficient; the bias of all low-hardness pixel points is determined according to the product between the maximum depth difference and a preset third coefficient; wherein the preset first coefficient is less than the preset second coefficient, and the preset second coefficient is less than the preset third coefficient.

[0020] Further, the position bias vector acquisition process comprises:

[0021] In the bright field image, the gradient amplitude and the gradient direction angle of each pixel point are obtained by a sobel operator, and the gradient direction angle is rounded to an integer;

[0022] According to the gradient amplitude distribution of the corresponding pixel points under each gradient direction angle, an overall direction angle is screened out; the overall direction angle is subjected to unit vectorization to determine a corresponding two-dimensional unit vector; the position bias vector of each pixel point is determined according to the product between the bias of each pixel point and the two-dimensional unit vector.

[0023] Further, the overall direction angle acquisition process comprises:

[0024] The pixel points with equal gradient direction angles are divided into a pixel point set, and all pixel point sets are determined; the accumulated value of the gradient amplitudes of all pixel points in each pixel point set is calculated to determine a corresponding gradient weight; the gradient direction angle corresponding to the pixel point set with the maximum gradient weight is taken as the overall direction angle.

[0025] Further, the electron microscope stage coordinate acquisition process comprises:

[0026] An electron microscope image of a shale rock slice is collected by a field emission scanning electron microscope; in the bright field image, the physical length corresponding to each pixel point is determined according to the ratio between the camera pixel size and the objective lens magnification; based on the feature point matching of each diamond polishing powder particle on the shale rock slice in the electron microscope image and in the bright field image, the electron microscope reference coordinates of each pixel point are determined;

[0027] The position bias vector is represented by coordinates, and corresponding bias coordinates are determined; the bias coordinates are added to corresponding coordinate values of the electron microscope reference coordinates to determine the electron microscope stage coordinates of each pixel point.

[0028] Further, the acquisition process of the electron microscope reference coordinates comprises:

[0029] The electron microscope physical coordinates corresponding to each pixel point in the electron microscope image are acquired;

[0030] According to the product of each coordinate value of the pixel coordinates of each pixel point in the bright field image and the corresponding physical length, the physical coordinate reference value corresponding to each coordinate value is determined; according to the two physical coordinate reference values corresponding to each pixel point, the optical physical coordinates corresponding to each pixel point in the bright field image are determined; based on the acquisition principle of the optical physical coordinates, the electron microscope physical coordinates corresponding to each pixel point in the electron microscope image are determined; based on the SIFT feature matching algorithm, the optical feature points corresponding to each diamond polishing powder particle on the shale rock thin section in the bright field image and the electron microscope feature points corresponding to each diamond polishing powder particle in the electron microscope image are determined;

[0031] According to the coordinate correspondence relationship between the optical feature points and the electron microscope feature points, the electron microscope reference coordinates corresponding to each pixel point are determined.

[0032] Further, the process of determining the electron microscope reference coordinates corresponding to each pixel point according to the coordinate correspondence relationship between the optical feature points and the electron microscope feature points comprises:

[0033] A similarity transformation model is established; the output of the similarity transformation model is the electron microscope physical coordinates, and the input is the optical physical coordinates; the similarity transformation model includes a magnification error correction coefficient, an orthogonal rotation matrix and a translation vector; the magnification error correction coefficient, the orthogonal rotation matrix and the translation vector are determined by a coordinate data set combined with the least square method; the coordinate data set includes the electron microscope physical coordinates of each electron microscope feature point and the optical physical coordinates of the corresponding optical feature point;

[0034] The optical physical coordinates of each pixel point in the bright field image are input into the similarity transformation model, and the electron microscope reference coordinates of each pixel point are output.

[0035] In a second aspect, the present application provides a shale rock thin section analysis and identification system for optical microscope-field emission electron microscope combination, the system comprising:

[0036] A data acquisition and preprocessing module is used to acquire the bright field image and the cross polarized image of the shale rock thin section under the optical microscope;

[0037] The first determining module is configured to determine a mineral identification index of each pixel point based on the brightness distribution of the bright field image and the birefringence deviation of the cross polarized image.

[0038] The second determining module is configured to determine a position bias vector of each pixel point based on the distribution of the mineral identification index of each pixel point, the surface relief of the shale rock thin section, and the gray scale gradient distribution trend.

[0039] The analysis and identification module is configured to determine an electron microscope stage coordinate of each pixel point based on the image physical distance scale, the rotation and translation processing, and the position bias vector of each pixel point, and perform shale rock thin section analysis and identification according to the electron microscope stage coordinate.

[0040] In a third aspect, the present application provides a computer device, comprising a memory and a processor. The memory is configured to store computer program code, and the processor is configured to call and run the computer program code from the memory to execute the method of the first aspect or any embodiment of the first aspect of the present application.

[0041] In a fourth aspect, the present application provides a computer program product, comprising computer program code, which, when executed, performs the method of the first aspect or any embodiment of the first aspect of the present application.

[0042] In a fifth aspect, the present application provides a computer readable storage medium, which stores computer program code, which, when executed, performs the method of the first aspect or any embodiment of the first aspect of the present application.

[0043] The present application has the following beneficial effects:

[0044] The present application first determines a mineral identification index representing atomic number proxy based on the brightness distribution of the bright field image and the birefringence deviation of the cross polarized image of the shale rock thin section from the perspective of optics. Then, different biases are applied based on the distribution of the mineral identification index and the surface relief of the shale rock thin section, and the position bias vector is determined based on the gray scale gradient distribution, so as to effectively compensate for the nonlinear repositioning error caused by the difference in mineral hardness based on the position bias vector. Finally, the position bias vector is combined during the process of pixel coordinate conversion based on the image physical distance scale and the rotation and translation processing, so as to determine the electron microscope stage coordinate corresponding to each pixel point with more accurate cross-scale positioning, so that the accuracy of the cross-scale positioning of the optical microscope and the field emission scanning electron microscope is higher, and the reliability of the shale rock thin section analysis and identification is significantly improved. BRIEF DESCRIPTION OF DRAWINGS

[0045] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0046] Figure 1 A flow chart of a shale rock thin section analysis and identification method combined with optical microscopy and field emission electron microscopy provided by an embodiment of the present application;

[0047] Figure 2 A structural diagram of a shale rock thin section analysis and identification system combined with optical microscopy and field emission electron microscopy provided by an embodiment of the present application;

[0048] Figure 3 A structural diagram of a computer device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0049] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined purpose, the specific embodiments, structures, characteristics and effects of the shale rock thin section analysis and identification method combined with optical microscopy and field emission electron microscopy according to the present application are described in detail as follows. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment, and the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form. In addition, the terms "first", "second" are only used for description purposes, and cannot be understood as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Therefore, the features with "first", "second" can be explicitly or implicitly included one or more features.

[0050] 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 the present application belongs.

[0051] The specific scheme of the shale rock thin section analysis and identification method combined with optical microscopy and field emission electron microscopy provided by the present application will be specifically described below with reference to the drawings.

[0052] The present application provides a shale rock thin section analysis and identification method combined with optical microscopy and field emission electron microscopy. Please refer to Figure 1 which shows a flow chart of a shale rock thin section analysis and identification method combined with optical microscopy and field emission electron microscopy provided by an embodiment of the present application. The method comprises:

[0053] Step S101: Collect the bright field image and crossed polarized image of the shale rock thin section under the optical microscope.

[0054] A shale sample to be analyzed is selected from the shale reservoir of the oilfield to be analyzed, and the sample is cut into a block with a size of 25 mm x 25 mm x 3 mm using a cutting machine. The cut shale rock thin section is placed on a coarse grinding disc, and the surface of the coarse grinding disc is coated with diamond abrasive with a particle size of 200 mesh. The coarse grinding equipment is turned on, the speed is set to 150 r / min, a stainless steel clamping ring is used, a constant force of 4 N is set, the shale rock thin section is made to move in a circular motion on the grinding disc, and the process is continued until the thickness of the shale rock thin section reaches 0.05 mm. During this process, the abrasive and coolant are constantly added to ensure the cutting ability of the abrasive and prevent overheating of the sample.

[0055] Then, the particle size of the diamond abrasive is changed to 1200 mesh, and the coarse ground shale rock thin section is transferred to a fine grinding disc. The speed of the fine grinding equipment is adjusted to 100 r / min, the shale rock thin section is also made to move in a circular motion on the grinding disc, and uniform pressure is applied. The grinding process is continued until the thickness of the thin section reaches 0.03 mm.

[0056] The fine ground shale rock thin section is placed on the polishing disc of the polishing machine, the polishing disc surface is covered with polishing cloth, and cerium oxide polishing liquid is applied. The polishing machine is set to a speed of 80 r / min, a stainless steel clamping ring is used, a constant force of 4 N is set, the shale rock thin section is made to move in a circular motion on the polishing disc, and the polishing time is 15 min until the surface of the shale rock thin section reaches a mirror effect. The polishing process requires constant replenishment of the polishing liquid to ensure the polishing effect.

[0057] The polished shale rock thin section is fixed to the surface of the conductive glass using an epoxy resin adhesive. A thin layer of adhesive is evenly applied to the conductive glass, then the shale rock thin section is placed on the adhesive, and the position of the thin section is adjusted so that it is centered on the conductive glass. The conductive glass with the fixed sample is placed in a constant temperature drying box and cured at 60°C for 2 hours to allow the adhesive to fully cure and ensure that the sample is firmly bonded to the conductive glass.

[0058] The polished shale rock thin section is placed on a positionable rotary stage, and a non-immersion transmission light path is used with an LED white light source. First, the 4 objective lens is pre-focused and the field of view is selected, then the 10 or 20 objective lens is switched, and a 8-bit grayscale image is collected using a CMOS camera under completely coaxial illumination conditions. The exposure time is set to a fixed 5 ms to ensure consistent brightness response of different shale thin sections. Then, a flat field correction white board is used to eliminate background noise and normalize to obtain the bright field image.

[0059] Under the vertical coaxial illumination, the incident light enters the objective lens after reflecting on the mineral surface, and the reflectivity is affected by the refractive index, extinction coefficient and surface microtopography, so the brightness is related to the physical properties of the rock hardness, such as average atomic coefficient, organic matter content and porosity.

[0060] The positions of the slices and the field of view of the bright field are completely consistent, the upper and lower polarizers (polarizer and analyzer) are inserted in the optical path and adjusted to the extinction position, and the angle between the crystal axis and the x axis at the 0° scale of the turntable is taken as the reference, the sample is rotated counterclockwise by 45° around the axis (microscope optical axis) by using the rotation function of the stage, that is, the angle between the polarizer transmission direction and the x axis, to obtain the maximum interference brightness. The same objective lens and camera parameters as the bright field are used, and the exposure time is only extended to 8 ms to compensate for the light intensity attenuation, an 8-bit grayscale image is collected, and after dark current subtraction and normalization, the cross-polarization image is obtained.

[0061] The shale rock slice is placed in a vacuum evaporation instrument, and the film thickness is controlled at 10 nm by adjusting the sputtering current and sputtering time. The sputtering time is 3 min, and the sputtering material is selected to be gold wire. During the sputtering process, the sample is uniformly rotated by the rotating device in the sample chamber, and the rotation rate is set to 5 r / min to ensure the uniformity of the film.

[0062] During the polishing process of the shale rock slice, the difference in mineral hardness directly determines the cutting resistance of each grain. High-hardness mineral phases such as quartz and pyrite are not easy to cut, and the surface will have protrusions, while low-hardness mineral phases such as organic matter, clay and carbonate are easily excavated first, resulting in surface depressions, and thus forming a rolling microtopography on the shale surface.

[0063] Step S102: Based on the brightness distribution of the bright field image and the birefringence deviation of the cross-polarization image, the mineral discrimination index of each pixel point is determined.

[0064] When the shale slice is transferred from the optical microscope (reflected light path is not sensitive to height) to the electron microscope (high vacuum, secondary electron or backscattered signal is sensitive to surface topography), the coordinates recorded by the optical microscope only correspond to the projection of the polished plane. When the electron microscope stage is reset, the actual landing point of the electron beam slides to the grain with higher hardness and protrusion due to the rolling, resulting in a deviation, and the deviation amount is nonlinearly accumulated with the hardness gradient in the shale bedding, so the traditional mechanical coordinates cannot compensate, resulting in the loss of the target area and the deviation of the element analysis result; therefore, in order to perform more accurate cross-scale positioning, the hardness information needs to be analyzed.

[0065] In shale, there exist components with similar brightness in bright-field images but different crystal structures, such as quartz and calcite, dolomite and magnesite. Therefore, cross-polarized images are acquired to obtain the birefringence information of the minerals. When linearly polarized light passes through anisotropic grains, it decomposes into two beams of polarized light with a phase difference. When the upper and lower polarizers are orthogonal, the phase difference determines the transmission intensity. The relationship between this intensity and the birefringence of the mineral is used to distinguish the hardness of different crystal structures. Therefore, this embodiment of the invention determines the mineral discrimination index for each pixel based on the brightness distribution of the bright-field image and the birefringence deviation of the cross-polarized image.

[0066] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the mineral discrimination index includes:

[0067] The birefringence surrogate value of each pixel is determined based on the average wavelength of the light source, the thickness of the shale rock sheet, and the pixel value of each pixel on the cross-polarized image.

[0068] In one specific implementation of this invention, the process of obtaining the birefringence surrogate value includes: determining the corresponding correction parameter based on the ratio between the average wavelength of the light source and the thickness of the shale rock sheet; and normalizing the product between the pixel value of each pixel on the cross-polarized image and the correction parameter to determine the birefringence surrogate value of each pixel.

[0069] In one specific implementation of this invention, the process of obtaining the birefringence surrogate value is expressed by the following formula: ; For pixels Birefringence proxy value; The average wavelength of the light source; The thickness of a shale rock thin section; For pixels The pixel values ​​on the cross-polarized image; wherein, the thickness of the shale rock slice is set to 0.03 mm after fine grinding; the average wavelength of the light source is set to the average between the maximum (600 nm) and minimum (400 nm) wavelengths of the LED white light source in this embodiment of the invention, i.e., 500 nm. It should be noted that the unit of each parameter is not considered when calculating the birefringence surrogate value, and the calculated value is used for subsequent sorting, so ignoring the unit does not affect the subsequent calculation and analysis process.

[0070] The process of determining the birefringence surrogate value of each pixel based on the average wavelength of the light source, the thickness of the shale rock sheet, and the pixel value of each pixel in the cross-polarized image can be explained as follows: the pixel intensity (pixel value) given by the cross-polarized image. and phase delay The relationship between the linearly polarized light and the transmitted light is a sinusoidal square relationship: According to Malus's law, the relationship between the linearly polarized light transmitted through the analyzer and the transmitted light is a cosine square relationship. Since the polarizer and analyzer are orthogonal, the optical axis of the grain and the direction of the polarizer are perpendicular. Angular, linearly polarized light The amplitude is decomposed at the front surface of the grain into ordinary light in the direction perpendicular to the optical axis. and unusual light along the optical axis The two propagate at different speeds after entering the grain, and the cumulative phase difference is the phase delay. .like When the reference phase is 0, then , where i represents the imaginary unit.

[0071] Since the analyzer only allows the y-direction component to pass through, the two components are projected onto the analyzer direction and superimposed. In optics, light intensity (i.e., the strength of light) is proportional to the square of the amplitude, i.e. ,in, For the incident light intensity, The intensity of transmitted light. The amplitude component in the direction of the analyzer. .so ,Right now In the embodiments of the present invention Therefore, the transmitted light intensity and They satisfy a sinusoidal square relationship, that is... (1), therefore, under the small delay approximation And because of the delay With birefringence and shale rock thin section thickness There is a linear relationship between them. (2) is a well-known optical phase delay formula, which will not be elaborated further here. Combining (1) and (2), we get ,in Indicates the incident wavelength In this embodiment of the invention, d represents the average wavelength of the light source; d is the propagation distance of light in the mineral crystal, and its value is related to the thickness of the shale rock sheet. Similarly, the ratio of the two is used as a normalization weight to convert the retardation into a difference in birefringence per unit thickness. .

[0072] To avoid amplifying the noise caused by the square root, therefore... To represent pixels The local birefringence difference, considering that the mineral discrimination index calculated based on the birefringence difference is used for relative ranking, therefore, to avoid the influence of the value range on the magnitude of the discrimination index, in Based on the normalization, the range of the obtained birefringence surrogate value is limited to 0 to 1, and the final birefringence surrogate value is determined. It is directly related to the crystal structure of the mineral, thereby introducing additional distinguishability between mineral phases with similar bright field brightness but different crystallographic properties. It is used to characterize the material hardness during cross-scale resetting and to calculate the bias along the polishing direction, thereby compensating for surface undulations caused by the difference in cutting rates between soft and hard minerals.

[0073] In bright-field images, a higher grayscale value indicates a smoother, polished surface, which better matches the polishing characteristics of high-hardness minerals. Furthermore, due to the smoothness of the surface, the offset between the actual position under an electron microscope and the coordinates under an optical microscope is usually smaller, thus requiring a smaller bias vector and assigning a larger mineral discrimination index. Therefore, this embodiment of the invention, based on correlation, further determines the corresponding mineral discrimination index according to the sum of the normalized grayscale value and the birefringence surrogate value of each pixel in the bright-field image; this allows for adjustment with a larger bias when the mineral discrimination index is small. In one specific implementation of this invention, the method for normalizing the grayscale value of each pixel in the bright-field image uses a minimum-maximum normalization method, which can be adjusted according to the specific implementation environment and will not be further elaborated here.

[0074] Step S103: Based on the distribution of mineral discrimination indicators of each pixel, the surface undulation of shale rock thin sections, and the gray-scale gradient distribution trend, determine the position offset vector of each pixel.

[0075] First, based on step S102, it is known that the smaller the mineral discrimination index, the larger the bias is usually generated, so the amplitude of the corresponding position bias vector should be larger. The amplitude of the bias needs to be set as a benchmark. Considering that the absolute depth of the surface undulation may be different for different thin slices and different polishing batches, using the absolute bias amount as a benchmark will lead to poor robustness in the process of obtaining the position bias vector. Since the polishing machine's rotation speed and polishing time determine the total cutting energy input per unit area, and the Vickers hardness of the material determines the volume that this energy can remove, the difference in volume that can be removed between the soft and hard phases, after being normalized by the sheet thickness, corresponds to the maximum depth of surface undulations. This parameter quantifies the limit distance at which the electron beam slips due to changes in surface morphology. Therefore, when obtaining the position offset vector, it is also necessary to consider the surface undulations of the shale rock sheet. In addition, determining the position offset vector also requires determining the orientation angle. The grayscale field of the polishing scratch under a light microscope can be regarded as a one-dimensional curve unfolding along a single direction. On the planed surface perpendicular to the scratch direction, the scratch area presents a bright fine peak, which rapidly drops to the matrix grayscale on both sides, while the grayscale changes slowly along the scratch direction. Therefore, it is only necessary to detect the peak position and direction within this profile, and use one-dimensional extreme value tracking to replace two-dimensional complex texture analysis, thereby quickly extracting the main orientation angle. This angle represents the dominant direction of relative motion between the polishing disk and the shale section during polishing. As input to the subsequent bias vector, it ensures that surface undulations caused by differences in cutting rates between soft and hard minerals are accurately compensated along this direction, thus pulling the electron microscope landing point back from geometric coordinates to the true target position after morphology correction. Therefore, based on the distribution of mineral discrimination indices for each pixel, the surface undulations of the shale section, and the grayscale gradient distribution trend, the position bias vector for each pixel is further determined.

[0076] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the position offset vector includes:

[0077] Considering that the maximum depth of the undulations on the shale rock thin section quantifies the limit distance of electron beam slippage due to changes in surface morphology, in order to improve the accuracy and robustness of the subsequently determined position offset vector, the maximum depth difference of the shale rock thin section is further obtained by a white light interferometer; based on the overall distribution of the mineral discrimination index of each pixel and the maximum depth difference, the offset of each pixel is determined.

[0078] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the bias amount includes:

[0079] K-means clustering analysis was performed on the mineral discrimination index of all pixels to determine three pixel clusters; the K value of the k-means clustering analysis was set to 3. Pixels in the cluster with the largest mean of the mineral discrimination index were identified as high-hardness pixels; pixels in the cluster with the smallest mean of the mineral discrimination index were identified as low-hardness pixels; all other pixels besides the high-hardness and low-hardness pixels were identified as medium-hardness pixels. The bias of all high-hardness pixels was determined by the product of the maximum depth difference and a preset first coefficient; the bias of all medium-hardness pixels was determined by the product of the maximum depth difference and a preset second coefficient; and the bias of all low-hardness pixels was determined by the product of the maximum depth difference and a preset third coefficient. The preset first coefficient was less than the preset second coefficient, and the preset second coefficient was less than the preset third coefficient.

[0080] In one specific implementation of this invention, the first preset coefficient is set to 0, the second preset coefficient is set to 1, and the third preset coefficient is set to 1.5, which can be adjusted according to the specific implementation environment. Based on the relative size of the mineral discrimination index, a lower bias is assigned to the pixel with greater hardness at the corresponding position, and a larger bias is assigned to the pixel with less hardness at the corresponding position.

[0081] Based on the gradient magnitude distribution of all pixels in the brightfield image at each gradient direction angle, combined with the bias, the position bias vector of each pixel is determined; preferably, in some possible implementations of this invention, the process of obtaining the position bias vector includes:

[0082] In a bright-field image, the gradient magnitude and gradient direction angle of each pixel are obtained using the Sobel operator, with the gradient direction angle rounded to the nearest integer. Based on the distribution of gradient magnitudes of the corresponding pixels under each gradient direction angle, the overall direction angle is selected. Specifically, pixels with equal gradient direction angles are grouped into a set of pixels, and all such sets are determined. The cumulative gradient magnitude of all pixels in each set is calculated to determine the corresponding gradient weight. The gradient direction angle corresponding to the set of pixels with the largest gradient weight is taken as the overall direction angle.

[0083] High-hardness minerals have low cutting rates during polishing and relatively convex surfaces, so they are given small or even negative biases to pull their electron microscope landing points back to the concave areas. Conversely, low-hardness minerals have high cutting rates and relatively concave surfaces, requiring larger positive biases to extrapolate the electron beam and dynamically compensate for micron-level morphological errors caused by differences in hardness along the polishing direction. Therefore, for each gradient direction angle, the larger the overall gradient amplitude of all corresponding pixels, the more it matches the dominant direction of relative motion between the polishing disk and the shale section during polishing. Using the overall direction angle as the input for subsequent bias vectors ensures that surface undulations caused by differences in cutting rates between hard and soft minerals are accurately compensated along this direction, thus pulling the electron microscope landing point from geometric coordinates back to the true target position after morphological correction. Therefore, the overall direction angle is further unit-vectorized to determine the corresponding two-dimensional unit vector; the position bias vector of each pixel is determined by the product of the bias of each pixel and the two-dimensional unit vector.

[0084] Step S104: After pixel coordinate transformation based on image physical distance scale and rotation and translation processing, the position offset vector of each pixel is combined to determine the electron microscope stage coordinates of each pixel; shale rock thin section analysis and identification are performed based on the electron microscope stage coordinates.

[0085] Before introducing the bias, it is necessary to perform cross-scale matching between the optical image and the field emission scanning electron microscope image. After pixel coordinate transformation based on the physical distance scale of the image and rotation and translation processing, the position bias vector of each pixel is combined to determine the electron microscope stage coordinates of each pixel.

[0086] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the coordinates of the electron microscope stage includes:

[0087] Electron microscopy (EM) images of shale thin sections were acquired using a field emission scanning electron microscope (FET). Specifically, after recording the pixel coordinates of each pixel in the bright-field image, the shale thin sections were coated with a conductive film and then transferred to the FET. A panoramic EEM image of the shale thin sections was acquired in low-magnification (100×) backscatter mode. The backscatter mode is sensitive to atomic number, ensuring that diamond (high atomic number) appears as a bright spot on the conductive glass (low atomic number) substrate, consistent with the bright spot characteristics under an optical microscope, making the subsequent feature extraction and feature matching process more accurate.

[0088] In bright-field images, the physical length of each pixel is determined based on the ratio between the camera pixel size and the objective lens magnification. CMOS pixels can only provide dimensionless row and column indices, while the compensation for polishing morphology caused by differences in mineral hardness, the subsequent calibration of rotation matrices and translation vectors, and the travel of the electron microscope stage all require real distances as a reference. Therefore, the pixel scale is first mapped to physical distance using a physical length acquisition method to ensure dimensional consistency.

[0089] Based on the feature point matching of each diamond polishing powder particle on the shale thin section in the electron microscope image and the bright-field image, the electron microscope reference coordinates of each pixel are determined. Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the electron microscope reference coordinates includes: obtaining the electron microscope physical coordinates corresponding to each pixel in the electron microscope image; determining the physical coordinate reference value corresponding to each coordinate value based on the product between each coordinate value of the pixel coordinates in the bright-field image and the corresponding physical length; determining the optical physical coordinates corresponding to each pixel in the bright-field image based on the two physical coordinate reference values ​​corresponding to each pixel; determining the electron microscope physical coordinates corresponding to each pixel in the electron microscope image based on the principle of obtaining optical physical coordinates; and determining the optical feature points corresponding to each diamond polishing powder particle on the shale thin section in the bright-field image and the electron microscope feature points corresponding to each diamond polishing powder particle in the electron microscope image based on the SIFT feature matching algorithm.

[0090] By analyzing optical and electron microscope (EM) physical coordinates, pixels in both the EM and bright-field images can be analyzed within the same physical reference frame, reducing mapping errors caused by different reference frames. Since diamond polishing powder particles exhibit bright spot characteristics distinct from other pixels in both bright-field and EM images, the SIFT feature matching algorithm can be used to extract the corresponding feature points of each diamond polishing powder particle in both images. These feature points are then matched to ultimately determine the optical feature points corresponding to each diamond polishing powder particle on the shale thin section in the bright-field image and the electron microscope feature points in the EM image. In one specific implementation of this invention, 20 diamond polishing powder particles are randomly placed at different locations on the shale thin section. The number of diamond polishing powder particles can be adjusted according to the specific implementation environment, or other high-brightness feature particles such as gold particles can be used as substitutes; further details are omitted here. It should be noted that the SIFT feature matching algorithm is a well-known technique in the art and will not be elaborated further here.

[0091] By establishing the positional correspondence between optical and electron microscope feature points, the positional scale correspondence of pixels in bright-field images (optical domain) and electron microscope images (electron microscope domain) can be determined, thus providing the fundamental cross-scale mapping basis for calculating the positional bias vectors of each pixel. In this embodiment, the electron microscope reference coordinates corresponding to each pixel are determined based on the coordinate correspondence between optical and electron microscope feature points.

[0092] Preferably, in some possible implementations of the embodiments of the present invention, the process of determining the electron microscope reference coordinates corresponding to each pixel point based on the coordinate correspondence between optical feature points and electron microscope feature points includes:

[0093] A similarity transformation model is established; the output of the similarity transformation model is the electron microscope physical coordinates, and the input is the optical physical coordinates; the similarity transformation model includes magnification error correction coefficients, orthogonal rotation matrices, and translation vectors; the magnification error correction coefficients, orthogonal rotation matrices, and translation vectors are determined by combining the coordinate data set with the least squares method; the coordinate data set includes the electron microscope physical coordinates of each electron microscope feature point and the corresponding optical physical coordinates of the optical feature point; the optical physical coordinates of each pixel in the brightfield image are input into the similarity transformation model, and the electron microscope reference coordinates of each pixel are output.

[0094] In one specific implementation of this invention, a similarity transformation model is first established in an initial state where the magnification error correction coefficient, orthogonal rotation matrix, and translation vector are all unknowns: ;in, The output of the similarity transformation model, This serves as the input to the similarity transformation model; This is the magnification error correction factor; It is an orthogonal rotation matrix; The translation vector is used as the input. The optical physical coordinates of each optical feature point are taken as input, and the electron microscope physical coordinates of the corresponding electron microscope feature point are taken as output. The least squares method is used to solve for the magnification error correction coefficient, orthogonal rotation matrix, and translation vector that minimize the reprojection error. The obtained values ​​of the magnification error correction coefficient, orthogonal rotation matrix, and translation vector are then input into the initial similarity transformation model to obtain the final similarity transformation model. By using the known mapping error magnitudes of each feature point, the magnification error correction coefficient, orthogonal rotation matrix, and translation vector with the smallest error are selected for the transformation, resulting in a more accurate similarity transformation model. It should be noted that the least squares method is a technique well-known to those skilled in the art and will not be elaborated further here.

[0095] The electron microscope reference coordinates are the physical coordinates of each pixel in the bright-field image without considering the offset. To determine the electron microscope stage coordinates, which more accurately represent the physical coordinates, a position offset vector is superimposed on the electron microscope reference coordinates to determine the electron microscope stage coordinates. Specifically, the position offset vector is represented by coordinates to determine the corresponding offset coordinates; the offset coordinates are added to the corresponding coordinate values ​​of the electron microscope reference coordinates to determine the electron microscope stage coordinates of each pixel.

[0096] Finally, the shale thin section analysis and identification are performed based on the electron microscope stage coordinates. In one specific implementation of this invention, the electron microscope stage coordinates corresponding to each pixel are converted into millimeter units and then written into the electron microscope stage for control, thereby enabling more accurate cross-scale positioning of optical microscope and field emission scanning electron microscope, which results in lower reliability of shale thin section analysis and identification.

[0097] The rock thin sections were further placed on the field emission electron microscope (FET) stage. The center coordinates of each section were used as the center of the field of view, and the size of the field of view was determined according to the resolution of the electron microscope. The electron beam was scanned point-by-point from the upper left corner. The accelerating voltage was set to 15 kV, the beam current to 0.4 nA, and the acquisition time to 60 s to ensure sufficient counts were obtained for accurate identification of elements and their content. Observation began at low magnification (e.g., 100×), gradually increasing the magnification (e.g., 1000×, 5000×, 10000× and above) to perform high-resolution observation of the marked areas, acquiring information such as the microscopic morphology of the rock surface, mineral grain boundaries, microcrack development characteristics, and pore structure. After acquisition, the spectra were processed using the analysis software built into the energy dispersive spectrometer (EDS). The software automatically identified the elements present in the sample by comparing them with a standard elemental spectrum database and calculated the relative content of each element based on the intensity of characteristic X-rays. Elemental analysis of the observed area was performed using the energy dispersive spectrometer (EDS) on the FET to determine the elemental composition of the minerals. By combining the observation of mineral optical properties under an optical microscope, the mineral types and contents can be more accurately identified. Adjust the microscope's focal length and begin observation with a low-power objective (e.g., 4×) to comprehensively observe the overall structure, mineral distribution, and macroscopic structural features of the rock thin section. Record the observed rock type, main mineral composition, grain morphology, and structural features, and take low-power micrographs to identify the area of ​​interest. Gradually switch to high-power objectives (e.g., 10×, 20×, 40×) for detailed observation of the area of ​​interest, further observing the mineral's optical properties, such as interference colors, extinction characteristics, and birefringence. These optical properties are used to determine the mineral type and content.

[0098] In summary, a method for analyzing and identifying shale thin sections using a combination of optical microscopy and field emission scanning electron microscopy (FETS) first determines mineral discrimination indices representing atomic number surrogates by combining the brightness distribution of bright-field images and the birefringence deviation of cross-polarized images from an optical perspective. Then, based on the distribution of the mineral discrimination indices and the surface undulations of the shale thin sections, different biases are applied, and the positional bias vector is determined by combining the gray-level gradient distribution. This positional bias vector effectively compensates for nonlinear resetting errors caused by differences in mineral hardness. Finally, by combining the positional bias vector with the pixel coordinate transformation process based on the image physical distance scale and rotation / translation processing, the electron microscope stage coordinates corresponding to each pixel are determined for more accurate cross-scale positioning. This improves the accuracy of cross-scale positioning for both optical microscopy and FETS, significantly enhancing the reliability of shale thin section analysis and identification.

[0099] This application also provides a shale thin section analysis and identification system using optical microscopy and field emission electron microscopy. Please refer to [link to relevant documentation]. Figure 2 The diagram shows a structural diagram of a shale thin section analysis and identification system using optical microscopy and field emission electron microscopy according to an embodiment of the present invention. The system includes: a data acquisition and preprocessing module 201, a first determination module 202, a second determination module 203, and an analysis and identification module 204.

[0100] The data acquisition and preprocessing module 201 is used to acquire bright-field images and cross-polarized images of shale rock thin sections under an optical microscope;

[0101] The first determining module 202 is used to determine the mineral discrimination index of each pixel based on the brightness distribution of the bright field image and the birefringence deviation of the cross-polarized image.

[0102] The second determining module 203 is used to determine the position offset vector of each pixel based on the distribution of mineral discrimination indicators of each pixel, the surface undulation of shale rock thin section and gray-scale gradient distribution trend.

[0103] The analysis and identification module 204 is used to determine the electron microscope stage coordinates of each pixel by performing pixel coordinate transformation based on the physical distance scale of the image and rotation and translation processing, combined with the position offset vector of each pixel; and to perform shale rock thin section analysis and identification based on the electron microscope stage coordinates.

[0104] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the shale thin section analysis and identification system using optical microscopy-field emission electron microscopy and the shale thin section analysis and identification method using optical microscopy-field emission electron microscopy provided in the above embodiments belong to the same concept, and the specific implementation process is detailed in the method embodiments, which will not be repeated here.

[0105] This application also provides a computer device; please refer to [link / reference]. Figure 3 The illustration shows a schematic diagram of a computer device structure according to an embodiment of the present invention. The computer device includes a memory 301, a processor 302, and a computer program 303 stored in the memory 301 and running on the processor 302. When the processor 302 executes the computer program 303, the computer device can execute any of the aforementioned methods for analyzing and identifying thin sections of shale rocks using optical microscopy and field emission electron microscopy.

[0106] This application also provides a computer program product that, when run on a computer device, enables the computer device to execute any of the aforementioned methods for analyzing and identifying thin sections of shale rocks using a combination of optical microscopy and field emission electron microscopy.

[0107] This application also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer device, the computer device can execute any of the aforementioned methods for analyzing and identifying thin sections of shale rocks using optical microscopy and field emission electron microscopy.

[0108] In the embodiments provided in this application, it should be understood that the computer device, computer program product and computer-readable storage medium provided are all used to perform the corresponding methods provided above, and therefore the beneficial effects they can achieve can be referred to the beneficial effects of the methods provided above, which will not be repeated here.

[0109] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0110] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for identifying shale rock thin section by optical microscope-field emission scanning electron microscope combination, characterized in that, The method comprises: Collecting a bright field image and a crossed polarized image of a shale rock thin section under an optical microscope; Determining a mineral identification index of each pixel point based on the brightness distribution of the bright field image and the birefringence deviation of the crossed polarized image; Determining a position bias vector of each pixel point based on the distribution of the mineral identification index of each pixel point, the surface relief of the shale rock thin section, and the gray scale gradient distribution trend; Converting the pixel coordinates based on the image physical distance scale and the rotation and translation processing, combining the position bias vector of each pixel point, determining the electron microscope stage coordinates of each pixel point, and performing shale rock thin section analysis and identification according to the electron microscope stage coordinates; The mineral identification index acquisition process comprises: Determining a birefringence proxy value of each pixel point according to the average wavelength of the light source, the thickness of the shale rock thin section, and the pixel value of each pixel point on the crossed polarized image; Determining the corresponding mineral identification index according to the sum value between the normalized value of the gray value of each pixel point on the bright field image and the birefringence proxy value; The birefringence proxy value acquisition process comprises: Determining a correction parameter according to the ratio between the average wavelength of the light source and the thickness of the shale rock thin section, and normalizing the product between the pixel value of each pixel point on the crossed polarized image and the correction parameter to determine the birefringence proxy value of each pixel point; The position bias vector acquisition process comprises: Obtaining the maximum depth difference of the shale rock thin section, and determining the bias amount of each pixel point based on the overall distribution of the mineral identification index size of each pixel point and the maximum depth difference; Determining the position bias vector of each pixel point based on the gradient amplitude distribution of all pixel points in each gradient direction angle on the bright field image and the bias amount.

2. The method according to claim 1, wherein the method is characterized by, The maximum depth difference acquisition process comprises: Obtaining the maximum depth difference of the shale rock thin section by a white light interferometer.

3. The method according to claim 1, wherein the method is characterized by, The bias amount acquisition process comprises: Performing k-means clustering analysis on the mineral identification indexes of all pixel points to determine three pixel point clusters; wherein the K value of the k-means clustering analysis is set to 3; the pixel points in the pixel point cluster with the maximum mean value of the mineral identification indexes of all pixel points are taken as high-hardness pixel points; the pixel points in the pixel point cluster with the minimum mean value of the mineral identification indexes of all pixel points are taken as low-hardness pixel points; the other pixel points except the high-hardness pixel points and the low-hardness pixel points are taken as medium-hardness pixel points; the bias amount of all high-hardness pixel points is determined according to the product between the maximum depth difference and a preset first coefficient; the bias amount of all medium-hardness pixel points is determined according to the product between the maximum depth difference and a preset second coefficient; the bias amount of all low-hardness pixel points is determined according to the product between the maximum depth difference and a preset third coefficient; wherein the preset first coefficient is less than the preset second coefficient, and the preset second coefficient is less than the preset third coefficient.

4. The method according to claim 1, wherein the method is characterized by, The position bias vector acquisition process comprises: In the bright field image, the gradient amplitude and the gradient direction angle of each pixel point are obtained by a sobel operator, and the gradient direction angle is rounded to an integer; According to the gradient amplitude distribution of the corresponding pixel points under each gradient direction angle, an overall direction angle is screened out; the overall direction angle is unit vectorized to determine a corresponding two-dimensional unit vector; and a position bias vector of each pixel point is determined according to the product between the bias of each pixel point and the two-dimensional unit vector.

5. The method according to claim 4, wherein the method is characterized by, The overall direction angle is obtained by: The pixel points with the same gradient direction angle are divided into a pixel point set, and all pixel point sets are determined; the accumulated value of the gradient amplitudes of all pixel points in each pixel point set is calculated to determine the corresponding gradient weight; and the gradient direction angle corresponding to the pixel point set with the maximum gradient weight is taken as the overall direction angle.

6. The method according to claim 1, wherein the method is characterized by, The acquisition process of the electron microscope stage coordinates includes: The electron microscope image of the shale rock slice is collected by a field emission scanning electron microscope; in the bright field image, the physical length corresponding to each pixel point is determined according to the ratio between the camera pixel size and the objective lens magnification; and the electron microscope reference coordinates of each pixel point are determined based on the feature point matching of each diamond polishing powder particle on the shale rock slice in the electron microscope image and in the bright field image. The position bias vector is represented by coordinates to determine the corresponding bias coordinates; and the bias coordinates and the electron microscope reference coordinates are added to determine the electron microscope stage coordinates of each pixel point.

7. The method according to claim 6, wherein the method is characterized by, The acquisition process of the electron microscope reference coordinates includes: The electron microscope physical coordinates corresponding to each pixel point in the electron microscope image are obtained; The physical coordinate reference value corresponding to each coordinate value is determined according to the product between each coordinate value of the pixel coordinates of each pixel point in the bright field image and the corresponding physical length; the optical physical coordinates corresponding to each pixel point in the bright field image are determined according to the two physical coordinate reference values of each pixel point; the electron microscope physical coordinates corresponding to each pixel point in the electron microscope image are determined based on the acquisition principle of the optical physical coordinates; and the optical feature points corresponding to each diamond polishing powder particle on the shale rock slice in the bright field image and the electron microscope feature points corresponding to each diamond polishing powder particle in the electron microscope image are determined based on the SIFT feature matching algorithm; The electron microscope reference coordinates corresponding to each pixel point are determined according to the coordinate correspondence relationship between the optical feature points and the electron microscope feature points.

8. The method according to claim 7, wherein the method is characterized by, The process of determining the electron microscope reference coordinates corresponding to each pixel point according to the coordinate correspondence relationship between the optical feature points and the electron microscope feature points includes: A similarity transformation model is established; the output of the similarity transformation model is the electron microscope physical coordinates, and the input is the optical physical coordinates; the similarity transformation model includes a magnification error correction coefficient, an orthogonal rotation matrix and a translation vector; the magnification error correction coefficient, the orthogonal rotation matrix and the translation vector are determined by a coordinate data set combined with a least square method; and the coordinate data set includes the electron microscope physical coordinates of each electron microscope feature point and the optical physical coordinates of the corresponding optical feature points. The optical physical coordinates of each pixel point in the bright field image are input into the similarity transformation model, and the electron microscope reference coordinates of each pixel point are output.

Images