A method for characterizing the crude oil occurrence state in continental high-clay shale reservoirs

Abstract

This application relates to the field of image filtering technology, specifically to a method for characterizing the crude oil occurrence state of continental high-clay shale reservoirs. The method includes: acquiring CT images of various shale reservoir samples; obtaining the geological complexity factor of each CT image based on edge and grayscale features; obtaining suspected pore contours in each CT image based on the size, shape, and grayscale features of each closed contour; obtaining the morphological influence coefficient of each pixel based on the size and grayscale features of each suspected pore contour, and combining this with the geological complexity factor of the CT image where each pixel is located, obtaining the window adjustment parameters for each pixel, and then obtaining the adjusted search window size for each pixel; subsequently, denoising each CT image and obtaining a standard mapping of the crude oil occurrence state. This application improves the accuracy of pore feature extraction from CT images by adaptively adjusting the search window size of each pixel in each CT image.

Description

Technical Field

[0001] This application relates to the field of image filtering technology, specifically to a method for characterizing the crude oil occurrence state in continental high-clay shale reservoirs. Background Technology

[0002] China's shale oil exploration and development has made significant breakthroughs in recent years, demonstrating promising exploration prospects. It has become a major replacement resource for increasing China's oil reserves and production after conventional oil, and to a certain extent, ensures stable and even increased crude oil production in China. Currently, research on shale reservoirs mainly focuses on measuring parameters such as porosity and permeability. By utilizing computer vision technology and high-resolution imaging technology to analyze the microscopic pore structure characteristics of continental high-clay shale reservoirs (hereinafter referred to as shale reservoirs), combined with fluid dynamics and chemical models, the occurrence state of crude oil can be more accurately characterized, improving the accuracy of reservoir evaluation.

[0003] Currently, the crude oil occurrence status of shale reservoirs is mainly determined through high-resolution imaging techniques (such as CT scans and scanning electron microscopy) to obtain three-dimensional pore structure images of reservoir cores. Image processing and analysis are then performed to identify the microscopic pore features of the reservoir. However, images obtained through these imaging techniques often contain significant noise, leading to insufficient accuracy in identifying the microscopic pore features. Existing techniques use filtering algorithms to eliminate artifacts left at the image center or edges due to factors such as non-localized X-ray sources or slight changes in sample position during scanning. After noise reduction, a watershed algorithm is used for threshold segmentation to identify pores in the image. However, due to differences in rock strata structure and composition, the features in CT scan images vary considerably. Traditional image processing methods are difficult to apply to CT images of shale reservoirs at different depths, resulting in the inability to extract accurate pore features from CT images of shale reservoirs at certain depths after processing. Ultimately, this leads to deviations in the characterization of the crude oil occurrence status of shale reservoirs. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides a method for characterizing the crude oil occurrence state in continental high-clay shale reservoirs, thereby resolving existing issues.

[0005] The method for characterizing the crude oil occurrence state in continental high-clay shale reservoirs proposed in this application adopts the following technical solution: One embodiment of this application provides a method for characterizing the crude oil occurrence state in continental high-clay shale reservoirs, the method comprising the following steps: Obtain CT images of samples from each shale reservoir; The edges and closed contours in each CT image are obtained; the geological complexity factor of each CT image is obtained based on the total number of edges in each CT image, the dispersion of gray values ​​of all pixels, and the difference in the number of edges in each closed contour; the porosity discrimination value of each closed contour is obtained based on the average level of the difference in the total number of pixels contained in each closed contour and all other closed contours in the CT image, the dispersion of curvature of all pixels on each closed contour, and the average level of gray values ​​of all pixels contained in each closed contour, and then the suspected porosity contours in each CT image are obtained. Based on the total number of pixels and average grayscale value of each suspected pore contour, as well as the number of pixels on each suspected pore contour, the morphological influence parameters of each suspected pore contour are obtained, and then the morphological influence coefficient of each pixel in each CT image is obtained. Based on the morphological influence coefficient of each pixel and the geological complexity factor of the CT image in which it is located, the window adjustment parameters of each pixel are obtained, and combined with the preset original search window size of each pixel, the adjusted search window size of each pixel is obtained. Then, the CT images are denoised, and the standard mapping of the crude oil occurrence state of the continental high-clay shale reservoir is obtained.

[0006] Preferably, the formula for calculating the geological complexity factor of each CT image is as follows: In the formula, Let be the geological complexity factor of the i-th CT image. It is the product of the total number of edges in the i-th CT image and the variance of the gray values ​​of all pixels. It is the mean of the differences in the total number of edges between any two closed contours in the i-th CT image.

[0007] Preferably, the formula for calculating the pore discrimination value of each closed contour is: In the formula, Let j be the porosity discrimination value of the j-th closed profile. Let be the average of the differences between the j-th closed contour and the total number of pixels contained within all other closed contours in the CT image. Let be the variance of the curvature of all pixels on the j-th closed contour. is the average gray value of the pixels contained within the j-th closed contour.

[0008] Preferably, the suspected pore contours in each CT image refer to closed contours in each CT image whose pore discrimination value is greater than or equal to a preset segmentation threshold.

[0009] Preferably, the calculation formula for the morphological influence parameters of each suspected pore profile is as follows: In the formula, Let be the morphological influence parameter of the j-th suspected pore profile. Let be the total number of pixels contained within the j-th suspected pore contour. Let J be the total number of pixels on the j-th suspected pore contour. is the average gray value of all pixels contained within the j-th suspected pore contour.

[0010] Preferably, the specific process of obtaining the morphological influence coefficient of each pixel in each CT image is as follows: the morphological influence parameter of each suspected pore contour in each CT image is used as the morphological influence parameter of all pixels contained in the corresponding suspected pore contour, and the morphological influence parameter of the remaining pixels in each CT image is set as a preset constant, and the value of the preset constant is greater than the maximum value of the morphological influence parameter of all suspected pore contours.

[0011] Preferably, the window adjustment parameter for each pixel refers to the ratio of the geological complexity factor of the CT image to its morphological influence parameter.

[0012] Preferably, the formula for calculating the size of the search window after adjusting each pixel is: In the formula, The size of the search window after adjusting for the v-th pixel. Adjust parameters for the window at the v-th pixel. The preset original search window size for the v-th pixel. To take an odd function.

[0013] Preferably, the specific process of denoising each CT image is as follows: each CT image and the adjusted search window size of each pixel in each CT image are used as input to the NLM algorithm to obtain each denoised CT image.

[0014] Preferably, the specific process for obtaining the standard mapping of crude oil occurrence state in continental high-clay shale reservoirs is as follows: using a geological knowledge base and a neural network model, pore features are extracted from each denoised CT image, and then a crude oil occurrence state model is established using the flow element method; pre-acquired well logging data is used as input to the neural network algorithm to obtain well logging curves; a mapping relationship is established between the well logging curves and the crude oil occurrence state model; the standard of various flow elements is determined using the Bayesian discriminant method, and a standard mapping of crude oil occurrence state in continental high-clay shale reservoirs is formed through the global optimal solution.

[0015] This application has at least the following beneficial effects: This application addresses the issue that when using the NLM algorithm to filter and denoise CT images of shale reservoir samples, it may smooth out pores, leading to deviations in subsequent pore feature extraction. By evaluating the impact of the complexity of geological components and the morphology of suspected pore contours in CT images of shale reservoir samples at different depths, the search window size of the NLM algorithm is adjusted according to different situations. This achieves high-quality denoising of CT images of shale reservoir samples at different depths using the NLM algorithm, improves the accuracy of pore feature extraction, and ultimately enhances the accuracy of characterizing the crude oil occurrence state of shale reservoirs. Attached Figure Description

[0016] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart illustrating the steps of a method for characterizing the crude oil occurrence state in terrestrial high-clay shale reservoirs provided in this application; Figure 2 A flowchart illustrating the process of obtaining the adjusted search window size for each pixel provided in this application. Detailed Implementation

[0018] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a method for characterizing the crude oil occurrence state of continental high-clay shale reservoirs proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0019] 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 pertains.

[0020] The following description, in conjunction with the accompanying drawings, details the specific scheme of the method for characterizing the crude oil occurrence state in terrestrial high-clay shale reservoirs provided in this application.

[0021] This application provides a method for characterizing the crude oil occurrence state in continental high-clay shale reservoirs, specifically, the following method is provided. Please refer to [link to relevant documentation]. Figure 1 The method includes the following steps: Step 1: Obtain CT images of each shale reservoir sample.

[0022] In this application, the scanning equipment used is a nanoVoxel-3000 CT scanner. The basic principle of CT scanning is that the intensity of X-rays passing through an object follows Beer's law, and the two exhibit an exponential decay relationship. The shale reservoir depth ranges from 1756.63m to 3912.03m, from which N shale reservoir samples are selected. CT images of each shale reservoir sample are acquired using the CT scanner. It should be noted that the value of N ranges from 10 to 1000 to avoid randomness in the data processing results. Implementers can choose different values ​​while ensuring this premise; in this embodiment, the value of N is 50.

[0023] Step 2: Obtain the edges and closed contours in each CT image; based on the total number of edges in each CT image, the dispersion of gray values ​​of all pixels, and the difference in the number of edges in each closed contour, obtain the geological complexity factor of each CT image; based on the average level of the difference in the total number of pixels contained in each closed contour and all other closed contours in the CT image, the dispersion of the curvature of all pixels on each closed contour, and the average level of gray values ​​of all pixels contained in each closed contour, obtain the porosity discrimination value of each closed contour, and then obtain the suspected porosity contours in each CT image.

[0024] Noise in CT images includes quantum noise, electronic noise, and reconstruction noise. The presence of these noises can lead to insufficient accuracy in identifying reservoir micropore features, thus necessitating denoising of CT images. When using the NLM algorithm to filter and denoise CT scan images, as shale depth increases, compaction reduces the number of pores in the sample, worsening pore connectivity. Furthermore, significant differences exist in bedding and lithofacies at different depths. Consequently, when using the NLM algorithm to filter and denoise CT images of shale reservoir samples, it may smooth out the pores, causing biases in subsequent pore feature extraction. This application aims to optimize the parameters of the NLM algorithm to achieve better denoising results when processing CT images of samples at different depths.

[0025] Typically, CT images of shale reservoir samples contain regions with different geological compositions, such as different rocks and clays. These different geological compositions vary in distribution and content, resulting in multiple regions of different sizes and color intensities within the image. During image denoising, the influence of these multiple different regions can easily lead to distortion after denoising. Therefore, the complex geological composition surrounding the pores must also be considered during the denoising process.

[0026] First, the edges and closed contours in each CT image are obtained. This application does not limit the method for obtaining the edges and closed contours in the image; in this embodiment, the Canny edge detection operator and contour tracking algorithm, which are known technologies, are used to obtain the edges and closed contours in each CT image.

[0027] When the geological composition of shale reservoir samples is complex, the morphological differences between regions are greater, and the differences in texture between regions are also greater. Clay regions contain more textures, and the regional boundaries are less regular. Under the compaction of minerals, they mainly exhibit an approximately parallel bedding structure with low gray values ​​and a banded distribution. Regions containing different rock compositions contain relatively smaller textures, are generally smooth, have more regular regional boundaries, and are densely distributed in sheet-like patterns.

[0028] In a preferred embodiment, the geological complexity factor of each CT image is obtained based on the total number of edges in each CT image, the dispersion of gray values ​​of all pixels, and the difference in the number of edges in each closed contour. This factor is used to characterize the geological complexity of the shale reservoir sample represented by each CT image.

[0029] In this embodiment, the geological complexity factor of the i-th CT image is denoted as... Its specific expression is: In the formula, Let be the geological complexity factor of the i-th CT image. It is the product of the total number of edges in the i-th CT image and the variance of the gray values ​​of all pixels. It is the mean of the differences in the total number of edges between any two closed contours in the i-th CT image.

[0030] The meaning of this expression is: The larger the value, the more edges there are in the i-th CT image, the more complex the pixel gray value (i.e., the overall color of the image), and the more obvious the total number of edges between closed contours in the image (i.e., the texture difference). This indicates that the geological complexity of the shale reservoir sample corresponding to the CT image is higher, and the more geological components it contains, the more complex it is.

[0031] When denoising CT images with complex geological conditions, a larger search window should be used to avoid image distortion after denoising, which could lead to deviations in the analysis of porosity.

[0032] Furthermore, the location, number, and size of pores in shale are not fixed at different depths. Generally, as the depth increases, the pores in shale become smaller under the action of compaction. They are easily smoothed out as noise during the noise reduction process. Therefore, in addition to considering the complexity of the geological composition in shale, the impact of noise reduction on pores should also be considered.

[0033] In shale, pores typically exhibit closed outlines and have a low density, resulting in lower grayscale values ​​in CT images. In contrast, other geological components with higher densities show higher grayscale values ​​in CT images. Furthermore, pore morphology is highly variable, with tortuous pore edges that are generally smaller than those formed by other geological components.

[0034] In a preferred embodiment, a pore discrimination value for each closed contour is obtained based on the average difference between the total number of pixels contained in each closed contour and all other closed contours in the CT image, the dispersion of the curvature of all pixels on each closed contour, and the average gray value of all pixels contained in each closed contour. This value is used to characterize the probability that each closed contour belongs to the edge of a pore.

[0035] In this embodiment, the pore discrimination value of the j-th closed contour is denoted as Its expression is: In the formula, Let j be the porosity discrimination value of the j-th closed profile. Let be the average of the differences between the j-th closed contour and the total number of pixels contained within all other closed contours in the CT image. Let be the variance of the curvature of all pixels on the j-th closed contour. Let be the average grayscale value of the pixels contained within the j-th closed contour. The method for calculating the curvature of each pixel within the contour is a well-known technique, and the specific process will not be elaborated further.

[0036] The larger the value, the larger the area covered by the j-th closed contour in the CT image is compared to the area covered by the other closed contours. The less regular the shape of the closed contour and the darker the color, the more likely the edge is to be a pore edge in the shale.

[0037] A threshold acquisition algorithm is used to obtain the segmentation threshold of the pore discrimination value of all closed contours in a single CT image, which is recorded as the preset segmentation threshold. Closed contours in the CT image with pore discrimination values ​​greater than or equal to the preset segmentation threshold are considered as suspected pore contours. Known threshold acquisition techniques include, but are not limited to, Otsu's thresholding method, Kriging method, and threshold iteration method. In one embodiment of this application, Otsu's thresholding method is used for threshold acquisition.

[0038] Thus, by using the above method, all suspected pore contours in each CT image can be obtained.

[0039] Step 3: Based on the total number of all pixels contained within each suspected pore contour and the average grayscale value, as well as the number of pixels on each suspected pore contour, obtain the morphological influence parameters of each suspected pore contour, and then obtain the morphological influence coefficient of each pixel in each CT image; based on the morphological influence coefficient of each pixel and the geological complexity factor of the CT image in which it is located, obtain the window adjustment parameters of each pixel, and combine them with the preset original search window size of each pixel to obtain the adjusted search window size of each pixel, and then denoise each CT image and obtain the standard mapping of the crude oil occurrence state of the continental high-clay shale reservoir.

[0040] Furthermore, when denoising suspected pore contours, the influence of their overall shape on denoising should be considered. The smaller the area covered by the suspected pore contour, the narrower the width, and the lighter the overall color, the more likely it is to be smoothed during the denoising process.

[0041] Therefore, based on the above feature analysis of the suspected pore contours, as a preferred implementation, the morphological influence parameters of each suspected pore contour are obtained according to the total number of all pixels contained in each suspected pore contour, the average gray level, and the number of pixels on each suspected pore contour. These parameters are used to characterize the possibility that each suspected pore contour will be smoothed during the denoising process.

[0042] In this embodiment, the morphological influence parameter of the j-th suspected pore profile is denoted as... Its specific expression is: In the formula, Let be the morphological influence parameter of the j-th suspected pore profile. Let be the total number of pixels contained within the j-th suspected pore contour. Let J be the total number of pixels on the j-th suspected pore contour. is the average grayscale value of all pixels contained within the j-th suspected pore contour. This morphological influence parameter means that the smaller the area covered by the suspected pore contour in the shale reservoir sample, the narrower the edge width, and the lighter the overall color, the more likely it is to be treated as noise during the denoising process and ultimately smoothed, thus affecting the acquisition of the porosity of the shale reservoir sample.

[0043] Thus, the morphological influence parameters of each suspected pore contour in each CT image can be obtained through the above method.

[0044] Furthermore, each shale reservoir sample CT image has a corresponding geological complexity factor. The higher the geological complexity of the CT image, the larger the search window should be used when denoising the pixels in the CT image using the NLM algorithm. Additionally, each suspected pore contour in the CT image of each shale reservoir sample has a morphological influence parameter. The larger the morphological influence parameter of the suspected pore contour, the smaller the search window should be used when denoising the pixels in the suspected pore contour using the NLM algorithm.

[0045] In summary, the geological complexity factor of a single shale reservoir sample CT image can be assigned to all pixels in that image; that is, each pixel corresponds to the geological complexity factor of the shale reservoir sample CT image. Similarly, the morphological influence parameter of each suspected pore contour in each CT image is used as the morphological influence parameter of all pixels contained on and within the corresponding suspected pore contour. The morphological influence parameters of the remaining pixels in the CT image are set to preset constants, and the value of the preset constants should be greater than the maximum value of the morphological influence parameters of all suspected pore contours. In this embodiment, the preset parameter is set to 1. Thus, each pixel in a single shale reservoir sample CT image corresponds to a geological complexity factor and a morphological influence parameter.

[0046] Based on the morphological influence coefficient of each pixel and the geological complexity factor of the CT image in which it is located, the window adjustment parameters for each pixel are obtained. In this embodiment, the window adjustment parameter of the v-th pixel is denoted as... Its expression is: In the formula, Adjust parameters for the window at the v-th pixel. Let v be the geological complexity factor of the CT image containing the v-th pixel. This refers to the shape-affecting parameters for the v-th pixel. A larger window adjustment parameter for each pixel indicates that the search window size for that pixel should be set larger.

[0047] Furthermore, the preset original search window size for each pixel in each CT image is adjusted. The formula for calculating the adjusted search window size for each pixel is as follows: In the formula, The size of the search window after adjusting for the v-th pixel. Adjust parameters for the window at the v-th pixel. The preset original search window size for the v-th pixel. This is an odd-number function used to extract odd numbers from the input data. The process for obtaining the adjusted search window size for each pixel is as follows: Figure 2 As shown.

[0048] In the above manner, the more complex the geological composition of a single shale reservoir sample, the larger the search window needs to be. However, the pixels currently being denoised using the NLM algorithm are suspected pore contours and have narrow and short edges, so the search window needs to be reduced.

[0049] Each CT image and the adjusted search window size for each pixel in each CT image are used as input to the NLM algorithm to obtain the denoised CT images.

[0050] Thus, by adaptively adjusting the search window size in the NLM algorithm according to the actual situation, the denoising effect of the NLM algorithm on CT images of different depths and geological compositions can be improved, ultimately obtaining high-quality CT images of shale reservoir samples.

[0051] Using a geological knowledge base and a neural network model, pore features are extracted from each denoised CT image. These pore features include total pore volume, total pore surface area, total throat volume, and total throat surface area. The neural network used in the embodiments of this application is a convolutional neural network (CNN). Implementers can choose other neural network models while ensuring recognition accuracy.

[0052] A crude oil storage state model was established using the flow element method, including parameters such as effective porosity, pore throat geometry factor, and pore throat specific surface area. The specific model formulas are as follows: In the formula, S represents the crude oil occurrence state parameter. For effective porosity, For the pore throat geometry factor, The specific surface area of ​​the pore throat is given. This model fully considers the influence of micropore structure on the occurrence state of crude oil.

[0053] in, The calculation method is as follows: In the formula, For effective porosity, The total pore volume, This represents the total volume of the shale sample. The calculation method is as follows: In the formula, For the pore throat geometry factor, The sum of the volumes of the pores and throats, It is the sum of the surface areas of the pores and throats. The calculation method is as follows: In the formula, The specific surface area of ​​the pore throat. The sum of the surface areas of pores and throats Let be the sum of the volumes of the pores and throats.

[0054] Based on the established crude oil occurrence state model, the flow unit method is used for evaluation. Specifically, neural network technology is used to analyze and process the pre-acquired logging data to obtain logging curves and establish a mapping relationship between logging curves and crude oil occurrence state parameters. Bayesian discrimination method is used to determine the standards of various flow units, and the standard mapping of crude oil occurrence state in continental high-clay shale reservoirs is formed through global optimal solution.

[0055] The calculated crude oil occurrence state was verified using reservoir core analysis data. Cross-plot comparison showed good agreement between the calculated results and the core analysis results, with the absolute error controlled within one order of magnitude, meeting production requirements.

[0056] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments of this specification have been described above. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0057] 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.

[0058] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them; modifications to the technical solutions described in the foregoing embodiments, or equivalent substitutions of some of the technical features, do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for characterizing the crude oil occurrence state in continental high-clay shale reservoirs, characterized in that, The method includes the following steps: Obtain CT images of samples from each shale reservoir; The edges and closed contours in each CT image are obtained; the geological complexity factor of each CT image is obtained based on the total number of edges in each CT image, the dispersion of gray values ​​of all pixels, and the difference in the number of edges in each closed contour; the porosity discrimination value of each closed contour is obtained based on the average level of the difference in the total number of pixels contained in each closed contour and all other closed contours in the CT image, the dispersion of curvature of all pixels on each closed contour, and the average level of gray values ​​of all pixels contained in each closed contour, and then the suspected porosity contours in each CT image are obtained. Based on the total number of all pixels contained within each suspected pore contour and the average gray level, as well as the number of pixels on each suspected pore contour, the morphological influence parameters of each suspected pore contour are obtained, and then the morphological influence coefficient of each pixel in each CT image is obtained; based on the morphological influence coefficient of each pixel and the geological complexity factor of the CT image in which it is located, the window adjustment parameters of each pixel are obtained, and combined with the preset original search window size of each pixel, the adjusted search window size of each pixel is obtained, and then the CT images are denoised, and the standard mapping of the crude oil occurrence state of the continental high clay shale reservoir is obtained. The formula for calculating the geological complexity factor of each CT image is as follows: In the formula, Let be the geological complexity factor of the i-th CT image. It is the product of the total number of edges in the i-th CT image and the variance of the gray values ​​of all pixels. It is the mean of the differences in the total number of edges between any two closed contours in the i-th CT image; The formula for calculating the pore discrimination value of each closed contour is as follows: In the formula, Let j be the porosity discrimination value of the closed profile. Let be the average of the differences between the j-th closed contour and the total number of pixels contained within all other closed contours in the CT image. Let be the variance of the curvature of all pixels on the j-th closed contour. The average grayscale value of the pixels contained within the j-th closed contour; The calculation formula for the morphological influence parameters of each suspected pore profile is as follows: In the formula, Let be the morphological influence parameter of the j-th suspected pore profile. Let be the total number of pixels contained within the j-th suspected pore contour. Let J be the total number of pixels on the j-th suspected pore contour. is the average gray value of all pixels contained within the j-th suspected pore contour.

2. The method for characterizing the crude oil occurrence state of continental high-clay shale reservoirs as described in claim 1, characterized in that, The suspected pore contours in each CT image refer to closed contours in each CT image whose pore discrimination value is greater than or equal to a preset segmentation threshold.

3. The method for characterizing the crude oil occurrence state of continental high-clay shale reservoirs as described in claim 1, characterized in that, The specific process for obtaining the morphological influence coefficient of each pixel in each CT image is as follows: the morphological influence parameter of each suspected pore contour in each CT image is used as the morphological influence parameter of all pixels contained in the corresponding suspected pore contour, and the morphological influence parameter of the remaining pixels in each CT image is set as a preset constant, and the value of the preset constant is greater than the maximum value of the morphological influence parameter of all suspected pore contours.

4. The method for characterizing the crude oil occurrence state in continental high-clay shale reservoirs as described in claim 1, characterized in that, The window adjustment parameter for each pixel refers to the ratio of the geological complexity factor of the CT image to its morphological influence parameter.

5. The method for characterizing the crude oil occurrence state of continental high-clay shale reservoirs as described in claim 1, characterized in that, The formula for calculating the size of the search window after adjusting each pixel is as follows: In the formula, The size of the search window after adjusting for the v-th pixel. Adjust parameters for the window at the v-th pixel. The preset original search window size for the v-th pixel. To take an odd function.

6. The method for characterizing the crude oil occurrence state of continental high-clay shale reservoirs as described in claim 1, characterized in that, The specific process of denoising each CT image is as follows: each CT image and the adjusted search window size of each pixel in each CT image are used as input to the NLM algorithm to obtain each denoised CT image.

7. The method for characterizing the crude oil occurrence state of continental high-clay shale reservoirs as described in claim 1, characterized in that, The specific process for obtaining the standard mapping of crude oil occurrence state in continental high-clay shale reservoirs is as follows: using a geological knowledge base and a neural network model, pore features are extracted from each denoised CT image, and then a crude oil occurrence state model is established using the flow element method; pre-acquired well logging data is used as input to the neural network algorithm to obtain well logging curves; a mapping relationship is established between the well logging curves and the crude oil occurrence state model; the standard of various flow elements is determined using the Bayesian discriminant method, and a standard mapping of crude oil occurrence state in continental high-clay shale reservoirs is formed through the global optimal solution.

Images