Segmentation method and device of compact sandstone SEM image

Abstract

The invention discloses a dense sandstone SEM image segmentation method and apparatus. The method comprises the steps of obtaining a dense sandstone SEM image; the SEM image is preprocessed, specifically, the abnormal gray value range of the pore edge is determined, and the abnormal gray value of the pore edge is corrected through the gray value of the non-pore component partially overlapped with the abnormal gray value range of the pore edge; and carrying out feature extraction on the preprocessed SEM image, and inputting the extracted feature value into the trained decision tree classifier so as to segment the component category of the compact sandstone. Through the technical scheme provided by the invention, the problem that the gray level of the pore edge in the SEM image is abnormal due to low surface flatness of the rock sample can be solved, and the image segmentation precision is improved.

Description

Background Art

[0002] With the increasing demand for clean energy, unconventional oil and gas resources (such as tight sandstone reservoirs) have attracted more and more attention. Accurate evaluation of the microstructure of tight sandstone is of great significance to improving the development efficiency of tight sandstone reservoirs.

[0003] SEM (Scanning Electron Microscope) is a commonly used method for characterizing the microstructure of materials. When SEM secondary electron imaging is used to test the microstructure of dense sandstone, ion beam polishing is used to create a large field of view and high-resolution MAPS image on the end face of the dense sandstone core sample, which provides an effective way to quantitatively analyze the various mineral components of dense sandstone. However, due to the large difference in hardness between the various components of dense sandstone, it is difficult to ensure that the end face of the polished sample is completely flat, especially the edge of the intersection of pores / cracks and minerals. Because it is relatively sharp, there will be abnormally high gray value areas, making it difficult to accurately segment the image. Summary of the invention

[0004] The embodiments of the present application provide a method and device for segmenting a dense sandstone SEM image, which can at least to a certain extent solve the problem of abnormal grayscale of pore edges in SEM images caused by the uneven surface flatness of the rock sample, thereby improving the image segmentation accuracy.

[0005] Other features and advantages of the present application will become apparent from the following detailed description, or may be learned in part by the practice of the present application.

[0006] According to a first aspect of an embodiment of the present application, a method for segmenting a dense sandstone SEM image is provided, comprising:

[0007] Obtain SEM images of dense sandstone;

[0008] Preprocessing the SEM image includes: determining the abnormal gray value range of the pore edge, and correcting the abnormal gray value of the pore edge by using the gray value of the non-porous component partially overlapping with the abnormal gray value range of the pore edge;

[0009] Feature extraction is performed on the preprocessed SEM image, and the extracted feature values ​​are input into the trained decision tree classifier to segment the component categories of the tight sandstone.

[0010] In some embodiments of the present application, based on the aforementioned scheme, the component categories of the tight sandstone include one or more of pores, potassium feldspar, quartz and kaolinite.

[0011] In some embodiments of the present application, based on the above scheme, the abnormal grayscale of the pore edge is corrected by using the grayscale value of the non-pore component that partially overlaps with the abnormal grayscale value range of the pore edge, including:

[0012] Screening out first target pixels that meet the grayscale value range of the pore edge abnormality from the SEM image, the first target pixels including pixels in the pore edge abnormal area and / or potassium feldspar pixels;

[0013] The first target pixel is used as the central pixel of the first matrix template. If the number of abnormal pixels in the first matrix template is less than a threshold, the average value of the normal pixels in the first matrix template is used as the new grayscale value of the abnormal pixels.

[0014] Among them, abnormal pixels are pixels whose grayscale values ​​are in the abnormal grayscale value range of the pore edge.

[0015] In some embodiments of the present application, based on the above solution, after taking the first target pixel as the central pixel of the first matrix template, the method further includes:

[0016] If the number of abnormal pixels in the first matrix template is not less than the threshold, the abnormal pixels are determined to be potassium feldspar pixels.

[0017] In some embodiments of the present application, based on the above scheme, feature extraction is performed on the preprocessed SEM image, including:

[0018] At least one filter function is used to perform filtering processing on the preprocessed SEM image to obtain at least one characteristic value based on pixel grayscale.

[0019] In some embodiments of the present application, based on the above scheme, the training process of the decision tree classifier is as follows:

[0020] The preprocessed SEM images are labeled according to the component categories of the dense sandstone to obtain labeled training pixels;

[0021] Using at least one filter function to filter the labeled training pixels, obtaining at least one feature value based on the pixel grayscale, and forming a training data set;

[0022] The decision tree classifier is trained using a training data set.

[0023] In some embodiments of the present application, based on the above scheme, when marking according to the component category of the tight sandstone, it also includes:

[0024] The grayscale abnormal area at the pore edge is divided into the non-pore component category adjacent to the abnormal area.

[0025] In some embodiments of the present application, based on the above solution, after the component categories of the tight sandstone are segmented, the following is further included:

[0026] The component category segmentation results are optimized by using the gray value range of non-porous components that are close to the gray value of pores.

[0027] In some embodiments of the present application, based on the above scheme, the component category segmentation result is optimized by using the gray value range of the non-porous component close to the pore gray value, including:

[0028] A second target pixel is selected according to the gray value range of kaolinite, and the second target pixel is used as the central pixel of the second matrix template;

[0029] If the number of pixels in the second matrix template that meet the kaolin gray value range is less than a second threshold, the pixels less than the second threshold are classified as pores.

[0030] According to a second aspect of an embodiment of the present application, a device for segmenting a dense sandstone SEM image is provided, comprising:

[0031] An image acquisition module, used to acquire SEM images of dense sandstone;

[0032] An image preprocessing module is used to preprocess the SEM image, including: determining the abnormal gray value range of the pore edge, and correcting the abnormal gray value of the pore edge by using the gray value of the non-pore component that partially overlaps with the abnormal gray value range of the pore edge;

[0033] The image segmentation module is used to extract features from the preprocessed SEM image and input the extracted feature values ​​into the trained decision tree classifier to segment the component categories of the dense sandstone.

[0034] According to a third aspect of an embodiment of the present application, a computer-readable storage medium is provided, in which at least one program code is stored. The at least one program code is loaded and executed by a processor to implement the operations performed by the method described in any one of the first aspects.

[0035] According to a fourth aspect of an embodiment of the present application, an electronic device is provided, comprising one or more processors and one or more memories, wherein at least one program code is stored in the one or more memories, and the at least one program code is loaded and executed by the one or more processors to implement the operations performed by the method described in any one of the first aspects above.

[0036] The technical solution proposed in the present application, before segmenting the SEM image, determines the abnormal grayscale value range of the pore edge, and uses the grayscale value of the non-porous component that partially overlaps with the abnormal grayscale value range of the pore edge to correct the abnormal grayscale of the pore edge, thereby solving the problem of abnormal grayscale of the pore edge in the SEM secondary electron grayscale image caused by the low surface flatness of the rock sample, eliminating the influence of the abnormal pore edge area on the subsequent image segmentation accuracy, and improving the image segmentation accuracy of dense sandstone.

[0037] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0039] Figure 1 A flow chart showing a method for segmenting a dense sandstone SEM image according to an embodiment of the present application is shown;

[0040] Figure 2 shows an SEM magnified image of dense sandstone according to one embodiment of the present application;

[0041] Figure 3 A detailed enlarged view of a local pore edge brightness anomaly according to an embodiment of the present application is shown;

[0042] Figure 4 A diagram showing the processing result of the abnormal pore edge region according to one embodiment of the present application is shown;

[0043] Figure 5 A schematic diagram of labeled training pixels according to an embodiment of the present application is shown;

[0044] Figure 6 A schematic diagram showing the result of image segmentation according to an embodiment of the present application is shown;

[0045] Figure 7 A schematic diagram showing an image segmentation result after optimization according to an embodiment of the present application is shown;

[0046] Figure 8 A structural block diagram of a device for segmenting a dense sandstone SEM image according to an embodiment of the present application is shown;

[0047] Fig. 9A schematic diagram of a computer-readable storage medium according to an embodiment of the present application is shown;

[0048] Fig.10 A schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present application is shown. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0050] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present application. However, those skilled in the art will appreciate that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, known methods, devices, realizations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0051] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0052] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0053] It should also be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the objects used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those shown or described.

[0054] See also Figure 1 , shows a flow chart of a method for segmenting a dense sandstone SEM image according to an embodiment of the present application.

[0055] like Figure 1 As shown, according to the first aspect of the embodiment of the present application, a method for segmenting a dense sandstone SEM image is provided, including but not limited to being implemented by steps S1 to S3:

[0056] Step S1. Obtaining a SEM image of dense sandstone;

[0057] Specifically, the embodiment of the present application uses a wide ion beam to polish the core end face of the dense sandstone, and uses the secondary electron mode imaging of the SEM to obtain a SEM secondary electron grayscale image of the dense sandstone. Preferably, the SEM secondary electron grayscale image contains 2048×1768 pixels.

[0058] In some embodiments of the present application, based on the aforementioned scheme, the component categories of the tight sandstone include one or more of pores, potassium feldspar, quartz and kaolinite.

[0059] See also Figure 2 , shows an SEM magnified image of dense sandstone according to an embodiment of the present application. Figure 2 As shown in the figure, the area with the smallest gray value (darkest) is the pore or crack, and the other areas in the figure are

[0060] The images of the domain are kaolinite, quartz, and potassium feldspar in descending order according to the grayscale value. It should be noted that since the grayscale value ranges of pores and cracks, quartz and albite, potassium feldspar and chlorite in the SEM secondary electron grayscale image are basically the same, when the dense sandstone is segmented into component categories based on the grayscale value, pores and cracks are divided into one category, quartz and albite are divided into one category, and potassium feldspar and chlorite are divided into one category. For example, when the grayscale value range is identified as the grayscale value range of pores / cracks, the component category is directly divided into pores / cracks without distinguishing whether it is a pore or a crack.

[0061] Step S2. Preprocessing the SEM image, including: determining the abnormal gray value range of the pore edge, and correcting the abnormal gray value of the pore edge using the gray value of the non-porous component that partially overlaps with the abnormal gray value range of the pore edge;

[0062] See also Figure 3 , showing a detailed enlarged view of a local pore edge brightness anomaly according to an embodiment of the present application.

[0063] Specifically, Figure 3shows an enlarged image of pores in quartz of dense sandstone. It can be seen from the figure that the grayscale at the edge of the enlarged pore is abnormal, that is, there is a highlighted area around the pore, and its grayscale value is significantly higher than the surrounding quartz. Since the abnormal grayscale value of the pore edge overlaps greatly with the grayscale value range of potassium feldspar or chlorite, the embodiment of the present application preferably uses the grayscale value range of potassium feldspar or chlorite to correct the abnormal grayscale of the pore edge.

[0064] In some embodiments of step S2, based on the above scheme, the abnormal grayscale of the pore edge is corrected by using the grayscale value of the non-pore component that partially overlaps with the abnormal grayscale value range of the pore edge, including:

[0065] Step S21. Screening out first target pixels that meet the grayscale value range of pore edge abnormality from the SEM image, the first target pixels including pixels of pore edge abnormal region and / or potassium feldspar pixels;

[0066] Specifically, by locating the grayscale abnormal region of the pore edge in the SEM secondary grayscale image, the grayscale value range of the region can be determined. Preferably, the grayscale value range of the pore edge abnormality in the embodiment of the present application is 233≤Gp≤255. After the grayscale value range of the pore edge abnormality is determined, all pixels that meet the range are screened out from the SEM image. At this time, these pixels may be pixels of the grayscale abnormal region of the pore edge, or they may be potassium feldspar pixels or chlorite pixels.

[0067] Step S22. Taking the first target pixel as the central pixel of the first matrix template, if the number of abnormal pixels in the first matrix template is less than the threshold, taking the average value of the normal pixels in the first matrix template as the new grayscale value of the abnormal pixels;

[0068] Among them, the abnormal pixels are pixels whose grayscale values ​​are in the abnormal grayscale value range of the pore edge.

[0069] In some embodiments of the present application, based on the above solution, after taking the first target pixel as the central pixel of the first matrix template, the method further includes:

[0070] If the number of abnormal pixels in the first matrix template is not less than the threshold, the abnormal pixels are determined to be potassium feldspar pixels.

[0071] Specifically, in order to filter out pixels in the grayscale abnormal area of ​​the pore edge from the first target pixel, the first matrix template that matches the distribution range of the grayscale abnormal area of ​​the pore edge is selected. Preferably, a 3×3 matrix is ​​set as the kernel size of the template, and the first target pixel is used as the central pixel of the first matrix template, thereby determining the number of pixels contained in the first matrix template.

[0072] When the number of abnormal pixels in the first matrix template is less than the threshold, that is, the pixels with grayscale values ​​in the abnormal grayscale value range of the pore edge are less than the threshold, for example, the number of pixels with grayscale values ​​in 233≤Gp≤255 is less than 4, then it means that the abnormal pixels in the first matrix template are pixels in the abnormal area of ​​the pore or crack edge. In order to eliminate the abnormality, the embodiment of the present application uses the average value of the normal pixels in the first matrix template as the new grayscale value of the abnormal pixels, that is, the grayscale value of the pixels in the abnormal area of ​​the pore edge is processed as the grayscale value of the quartz stone pixels adjacent to the pore, thereby eliminating the interference caused by the pore edge abnormality to the subsequent component category segmentation and improving the accuracy of image segmentation.

[0073] Of course, it is understandable that when the number of abnormal pixels in the first matrix template is not less than the threshold, for example, the number of pixels with a grayscale value within 233≤Gp≤255 is greater than or equal to 4, the abnormal pixels in the template are considered to be potassium feldspar or chlorite pixels, and no processing is required on the pixel at this time.

[0074] Preferably, in order to ensure that the abnormal grayscale of the pore / crack edge can be effectively suppressed, step S2 can be repeated multiple times until the grayscale value of the pore / crack edge is normal. Figure 4 As shown, a processing result diagram of the abnormal area of ​​the pore edge according to an embodiment of the present application is shown. It can be seen from the figure that the abnormal grayscale of the pore edge is effectively suppressed.

[0075] Step S3: extract features from the preprocessed SEM image, and input the extracted feature values ​​into the trained decision tree classifier to segment the component categories of the tight sandstone.

[0076] In some embodiments of step S3, based on the above scheme, the training process of the decision tree classifier is as follows:

[0077] The preprocessed SEM images are labeled according to the component categories of the dense sandstone to obtain labeled training pixels;

[0078] Figure 5 A schematic diagram of labeled training pixels according to an embodiment of the present application is shown.

[0079] like Figure 5 As shown, it can be understood that the preprocessing method of the SEM image here adopts the preprocessing method described in step S2, and the obtained is a SEM secondary electron grayscale image with the pore edge anomaly eliminated, and then a plurality of image areas containing the grayscale characteristics of each component category are selected as training pixels. The selected training pixels are dispersedly arranged, including the grayscale characteristics of each component and the grayscale anomaly of the pore edge; the category labels in the training pixels are annotated to obtain the annotated training pixels.

[0080] In some embodiments of the present application, based on the above scheme, when marking according to the component category of the tight sandstone, it also includes:

[0081] The grayscale abnormal area at the edge of the pore is divided into the non-pore component category adjacent to the abnormal area; for example, the grayscale abnormal area at the edge of the pore is divided into quartz stone, thereby eliminating the influence of the abnormal area on the component category segmentation.

[0082] Using at least one filter function to filter the labeled training pixels, obtaining at least one feature value based on the pixel grayscale, and forming a training data set;

[0083] The decision tree classifier is trained using a training data set.

[0084] Specifically, the training data set uses an unbalanced data set, and the parameters of the decision tree classifier are set as follows: the size of the number of pixels processed in each batch is 100-500, preferably 200; the number of classification trees is 100-500, preferably 200; the number of features used randomly is 5-12, preferably 5; the maximum depth is 4-10, preferably 5.

[0085] In some embodiments of step S3, based on the above scheme, feature extraction is performed on the preprocessed SEM image, including:

[0086] At least one filter function is used to perform filtering processing on the preprocessed SEM image to obtain at least one characteristic value based on pixel grayscale.

[0087] It should be noted that in the embodiment of the present application, both in the classifier training stage and in the image segmentation stage using the classifier, a filtering function is used to filter the preprocessed SEM image for feature extraction, wherein the filtering functions used include but are not limited to membrane projection filtering, bilateral filtering, Lipschitz filtering, and entropy filtering functions.

[0088] Among them, when the original image is processed by the membrane projection filter function, the membrane structure of the image is enhanced by directional filtering. The initial kernel for this operation is hard-coded as a 19×19 zero matrix with the middle column entry set to 1. Multiple kernels are created by rotating the original kernel by 6 degrees until the total rotation is 180 degrees, resulting in 30 kernels. Each kernel is convolved with the image, and then the set Z of 30 images is projected into a single image using 6 methods.

[0089] When a bilateral filter is applied to the original image, it is possible to better preserve edges while averaging / blurring the image. The filter does this by averaging only the values ​​of the color values ​​around the pixel that are close to the current pixel. The "closeness" of the other neighboring pixels to the current pixel is determined by the specified threshold. That is, for a value of 10, every pixel that contributes to the current average must be within 10 values ​​of the current pixel. In this example, a spatial radius of 5 and 10 is combined, as well as a range radius of 50 and 100.

[0090] When the original image is processed with a Lipschitz filter, it is equivalent to a cone-shaped grayscale opening. The Lipschitz cover can be applied to eliminate slowly varying image backgrounds by subtracting the lower Lipschitz cover (top-hat process). A combination of bottom-hat and top-hat is used with a sequential double-scan algorithm with slopes s=5, 10, 15, 20, 25.

[0091] When the entropy filter function is used to process the original image, a circle with a radius of r is drawn around each pixel; the histogram of the circle divided into numBins blocks is obtained; and the entropy in the histogram is calculated as ∑ pin histogram -p*log2(p), where p is the probability of each bin in the histogram. numBins is equal to 32, 64, 128, 256. r is equal to σ.

[0092] Figure 6 A schematic diagram of the result of image segmentation according to an embodiment of the present application is shown.

[0093] Will Figure 6 and Figure 2 By comparison, it can be seen that the decision tree classifier of the embodiment of the present application can automatically identify and classify pores / cracks, kaolinite, quartz / albite, and potassium feldspar / chlorite. There is no abnormal grayscale phenomenon at the pore edge in the segmented image, and the area with abnormal grayscale at the pore edge has been accurately classified.

[0094] In some embodiments of the present application, based on the above solution, after the component categories of the tight sandstone are segmented, the following is further included:

[0095] Step S4. Optimize the component category segmentation result by using the gray value range of the non-porous components close to the gray value of the pores.

[0096] In some embodiments of step S4, based on the above scheme, the component category segmentation result is optimized by using the gray value range of the non-porous component close to the gray value of the pore, including:

[0097] Step S41. Screening out a second target pixel according to the gray value range of kaolinite, and using the second target pixel as the center pixel of the second matrix template;

[0098] Step S42: If the number of pixels in the second matrix template that meet the kaolin gray value range is less than a second threshold, the pixels less than the second threshold are classified as pores.

[0099] It should be noted that, since the grayscale values ​​of some small pores and fine cracks are relatively high and close to the grayscale values ​​of kaolinite, some pores may be classified as kaolinite during identification by the classifier, resulting in classification errors. The embodiment of the present application performs morphological optimization for the above-mentioned problems, thereby improving the image segmentation accuracy. Specifically, the second target pixel is screened out according to the grayscale value range characteristics (99-194) of kaolinite, and the second target pixel is used as the central pixel of the second matrix template. Preferably, the second matrix template adopts a kernel size of 7×7. The number of pixels that meet the grayscale value range characteristics (99-194) inside the second matrix template is calculated. When the number of pixels whose grayscale values ​​meet the grayscale characteristics in the second matrix template is less than the second threshold value, for example, less than 24, it is considered that the component category corresponding to the pixel should be pores / cracks, thereby optimizing the image component category segmentation and further improving the segmentation accuracy.

[0100] like Figure 7 FIG. 1 shows a schematic diagram of optimizing the image segmentation result according to an embodiment of the present application, and FIG. 2 shows a schematic diagram of optimizing the image segmentation result according to an embodiment of the present application, and FIG. 3 shows a schematic diagram of optimizing the image segmentation result according to an embodiment of the present application, and FIG. 4 shows a schematic diagram of optimizing the image segmentation result according to an embodiment of the present invention, and FIG. Figure 6 It can be seen that the tiny cracks in the quartz are accurately divided.

[0101] Based on the above disclosed content, before segmenting the SEM image, the embodiment of the present application corrects the abnormal grayscale of the pore edge by determining the abnormal grayscale value range of the pore edge, using the grayscale value of the non-porous component that partially overlaps with the abnormal grayscale value range of the pore edge, thereby solving the problem of abnormal grayscale of the pore edge in the SEM secondary electron grayscale image caused by the low surface flatness of the rock sample, eliminating the influence of the abnormal pore edge area on the subsequent image segmentation accuracy, and improving the image segmentation accuracy of dense sandstone. In addition, the grayscale image segmentation is converted into a pixel classification operation, using Gaussian filtering function, bilateral filtering function, mean migration filtering function and minimum filtering function, each pixel and its neighboring area in the original image are described in multiple ways based on space and scale-related information, and multiple grayscale-based feature values ​​of each pixel are obtained, an efficient training data set is constructed, and the accuracy of the trained decision tree classifier is improved.

[0102] The following describes an embodiment of the device of the present application, which can be used to execute the method in the above embodiment of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the above embodiment of the method of the present application.

[0103] See also Figure 8, shows a structural block diagram of a dense sandstone SEM image segmentation device according to an embodiment of the present application.

[0104] like Figure 8 As shown, according to a second aspect of an embodiment of the present application, a device for segmenting a dense sandstone SEM image is provided, comprising:

[0105] An image acquisition module 201 is used to acquire a SEM image of dense sandstone;

[0106] The image preprocessing module 202 is used to preprocess the SEM image, including: determining the abnormal gray value range of the pore edge, and correcting the abnormal gray value of the pore edge by using the gray value of the non-pore component that partially overlaps with the abnormal gray value range of the pore edge;

[0107] The image segmentation module 203 is used to extract features from the preprocessed SEM image and input the extracted feature values ​​into the trained decision tree classifier to segment the component categories of the tight sandstone.

[0108] like Fig. 9 As shown, based on the same inventive concept, an embodiment of the present application provides a computer-readable storage medium 300, in which at least one computer program instruction is stored, and the at least one computer program instruction is loaded and executed by a processor to implement the operations performed by the method described in any one of the second aspects.

[0109] The computer readable storage medium 300 may be a portable compact disk read-only memory (CD-ROM) and include program code, and may be run on a terminal device, such as a personal computer. However, the computer readable storage medium 300 of the present application is not limited thereto. In the present application, the computer readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0110] The readable storage medium 300 may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof. More specific examples (a non-exhaustive list) of the readable storage medium 300 include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0111] Program code for performing the operations of the present application may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., using an Internet service provider to connect through the Internet).

[0112] As another aspect, the present application further provides an electronic device 400 capable of implementing the method described in the first aspect.

[0113] Those skilled in the art will appreciate that various aspects of the present application may be implemented as a system, method or program product. Therefore, various aspects of the present application may be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to as "circuit", "module" or "system" herein.

[0114] Refer to the following Fig.10 The electronic device 400 according to this embodiment of the present application is described. Fig.10 The electronic device 400 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0115] like Fig.10 As shown, the electronic device 400 is in the form of a general computing device. The components of the electronic device 400 may include but are not limited to: at least one processing unit 410, at least one storage unit 420, and a bus 430 connecting different system components (including the storage unit 420 and the processing unit 410).

[0116] The storage unit stores program codes, which can be executed by the processing unit 410, so that the processing unit 410 executes the steps described in the above “Example Method” section of this specification according to various exemplary implementations of the present application.

[0117] The storage unit 420 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 421 and / or a cache memory unit 422 , and may further include a read-only memory unit (ROM) 423 .

[0118] The storage unit 420 may also include a program / utility 424 having a set (at least one) of program modules 425, such program modules 425 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0119] Bus 430 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0120] The electronic device 400 may also communicate with one or more external devices 500 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 400, and / or communicate with any device that enables the electronic device 400 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 450. Furthermore, the electronic device 400 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 460. As shown, the network adapter 460 communicates with other modules of the electronic device 400 via a bus 430. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 400, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0121] The above description is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present application shall be included in the scope of the claims of the present application.

Claims

1. A method for segmenting a dense sandstone SEM image, characterized in that: include: Obtain SEM images of dense sandstone; Preprocessing the SEM image includes: determining the abnormal gray value range of the pore edge, and correcting the abnormal gray value of the pore edge by using the gray value of the non-porous component partially overlapping with the abnormal gray value range of the pore edge; Feature extraction is performed on the preprocessed SEM image, and the extracted feature values ​​are input into the trained decision tree classifier to segment the component categories of the tight sandstone.

2. The method according to claim 1, characterized in that The compositional categories of tight sandstone include one or more of pores, potassium feldspar, quartz and kaolinite.

3. The method according to claim 2, characterized in that The abnormal gray value at the pore edge is corrected by using the gray value of the non-pore component that partially overlaps with the abnormal gray value range of the pore edge, including: Screening out first target pixels that meet the grayscale value range of the pore edge abnormality from the SEM image, the first target pixels including pixels in the pore edge abnormal area and / or potassium feldspar pixels; The first target pixel is used as the central pixel of the first matrix template. If the number of abnormal pixels in the first matrix template is less than a threshold, the average value of the normal pixels in the first matrix template is used as the new grayscale value of the abnormal pixels. Among them, abnormal pixels are pixels whose grayscale values ​​are in the abnormal grayscale value range of the pore edge.

4. The method according to claim 3, characterized in that After taking the first target pixel as the central pixel of the first matrix template, the method further includes: If the number of abnormal pixels in the first matrix template is not less than the threshold, the abnormal pixels are determined to be potassium feldspar pixels.

5. The method according to claim 1, characterized in that Feature extraction is performed on the preprocessed SEM image, including: At least one filter function is used to perform filtering processing on the preprocessed SEM image to obtain at least one characteristic value based on pixel grayscale.

6. The method according to claim 1, characterized in that The training process of the decision tree classifier is as follows: The preprocessed SEM images are labeled according to the component categories of the dense sandstone to obtain labeled training pixels; Using at least one filter function to filter the labeled training pixels, obtaining at least one feature value based on the pixel grayscale, and forming a training data set; The decision tree classifier is trained using a training data set.

7. The method according to claim 6, characterized in that When labeling according to the compositional category of tight sandstone, it also includes: The grayscale abnormal area at the pore edge is divided into the non-pore component category adjacent to the abnormal area.

8. The method according to any one of claims 1 to 7, characterized in that: After segmenting the compositional categories of tight sandstone, it also includes: The component category segmentation results are optimized by using the gray value range of non-porous components that are close to the gray value of pores.

9. The method according to claim 8, characterized in that The gray value range of non-porous components close to the gray value of pores is used to optimize the component category segmentation results, including: A second target pixel is selected according to the gray value range of kaolinite, and the second target pixel is used as the central pixel of the second matrix template; If the number of pixels in the second matrix template that meet the kaolin gray value range is less than a second threshold, the pixels less than the second threshold are classified as pores.

10. A device for segmenting a dense sandstone SEM image, characterized in that: include: An image acquisition module, used to acquire SEM images of dense sandstone; An image preprocessing module is used to preprocess the SEM image, including: determining the abnormal gray value range of the pore edge, and correcting the abnormal gray value of the pore edge by using the gray value of the non-pore component that partially overlaps with the abnormal gray value range of the pore edge; The image segmentation module is used to extract features from the preprocessed SEM image and input the extracted feature values ​​into the trained decision tree classifier to segment the component categories of the dense sandstone.

Images