Image processing method and device, electronic equipment and nonvolatile storage medium

Abstract

This application discloses an image processing method, apparatus, electronic device, and non-volatile storage medium. The method includes: acquiring an original core image of a target rock and determining a residual image corresponding to the original core image, wherein the residual image is used to supplement missing details in the original core image; performing a magnification operation on the original core image and the residual image; compensating and reconstructing the original core image of the target size based on the residual image of the target size to obtain a reconstructed core image; and performing threshold segmentation processing on the reconstructed core image to obtain the target core image, wherein the resolution of the target core image is higher than that of the original core image. This application solves the technical problems of insufficient resolution improvement and inaccurate detail reproduction in related technologies for core image super-resolution reconstruction when dealing with complex shale structures and micron-level cracks.

Description

Technical Field

[0001] This application relates to the field of three-dimensional digital rock technology, and more specifically, to an image processing method, apparatus, electronic device, and non-volatile storage medium. Background Technology

[0002] Digital rock technology is a key tool for studying the pore structure and physical properties of reservoir rocks, as well as simulating microscale flow in porous media. High-precision scanning equipment such as micro-CT, nano-CT, and FIB-SEM can acquire three-dimensional digital images of rock cores, providing a foundation for flow simulation and physical property calculations. However, due to limitations in equipment scanning costs, resolution, and external environmental interference, the acquired core images often suffer from insufficient resolution or loss of detail, severely impacting the accuracy of subsequent calculations of key physical properties such as porosity and permeability.

[0003] Image super-resolution technology is an effective method for improving image resolution and is widely used in the field of digital core imaging. However, core image super-resolution reconstruction technology is mostly applied to relatively homogeneous sandstone and carbonate rocks. For shale with strong heterogeneity and complex pore structure, as well as rocks containing hydraulic fracturing fractures, the technology often suffers from insufficient resolution improvement and inaccurate detail reproduction.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This application provides an image processing method, apparatus, electronic device, and non-volatile storage medium to at least solve the technical problems of insufficient resolution improvement and inaccurate detail reproduction in core image super-resolution reconstruction technology in related technologies when dealing with complex shale structures and micron-level cracks.

[0006] According to one aspect of the embodiments of this application, an image processing method is provided, comprising: acquiring an original core image of a target rock and determining a residual image corresponding to the original core image, wherein the residual image is used to supplement missing details in the original core image; performing a magnification operation on the original core image and the residual image, wherein the magnification operation is used to adjust the size of the original core image and the residual image to a target size; performing compensation reconstruction on the original core image of the target size based on the residual image of the target size to obtain a reconstructed core image; performing threshold segmentation processing on the reconstructed core image to obtain a target core image, wherein the threshold segmentation processing is used to distinguish between pores and rock parts in the image frame, and the resolution of the target core image is higher than the resolution of the original core image.

[0007] Optionally, obtaining the original core image of the target rock includes: scanning the target rock with a scanning device to obtain a three-dimensional scan image, wherein the scanning device includes at least one of the following: a focused ion beam scanning electron microscope and a micro CT scanner; extracting two-dimensional binary images in the target direction from the three-dimensional scan image, and determining the two-dimensional binary images as the original core image, wherein the target direction includes: the x direction, the y direction, and the z direction, the x direction, the y direction, and the z direction are three mutually orthogonal directions in three-dimensional space, and each target direction corresponds to a set of two-dimensional binary images.

[0008] Optionally, determining the residual image corresponding to the original core image includes: analyzing the original core image using a target prediction model to obtain the residual image, wherein a set of two-dimensional binary images for each target direction corresponds to a set of two-dimensional residual images; the training steps of the target prediction model include: obtaining a training dataset, wherein the training dataset includes: multiple first core image samples with a first resolution, and second core image samples with a second resolution corresponding to each first core image sample, the second resolution being higher than the first resolution; training the initial prediction model based on the training dataset to enable the model to learn how to recover the details of a high-resolution image from a low-resolution image, thereby obtaining the target prediction model.

[0009] Optionally, the magnification operation on the original core image and the residual image includes: using a bicubic interpolation algorithm to magnify the original core image in each target direction to obtain the original core image of the target size corresponding to each target direction; and using a bicubic interpolation algorithm to magnify the residual image in each target direction to obtain the residual image of the target size corresponding to each target direction.

[0010] Optionally, based on the residual image of the target size, the original core image of the target size is compensated and reconstructed to obtain the reconstructed core image. This includes: fusing the original core images of the target size corresponding to different target directions through arithmetic averaging to obtain a fused three-dimensional image; determining the weight coefficients corresponding to the residual images of the target size in different target directions, wherein the weight coefficients are used to characterize the degree of influence of the residual images of different target directions on image reconstruction, and each target direction corresponds to a weight coefficient; determining a three-dimensional compensation image based on the weight coefficients of different target directions and the residual image of the target size, and using the three-dimensional compensation image to compensate the fused three-dimensional image to obtain the reconstructed core image.

[0011] Optionally, threshold segmentation processing is performed on the reconstructed core image to obtain the target core image, including: determining the porosity of the target rock and determining the target grayscale threshold based on the porosity; traversing each pixel in the reconstructed core image and segmenting the pixels according to the target grayscale threshold to obtain the target core image, wherein, during the segmentation process, pixels with grayscale values ​​not less than the target grayscale threshold are marked as pores, and pixels with grayscale values ​​less than the target grayscale threshold are marked as rock.

[0012] Optionally, the method further includes: importing the target core image into a simulation environment and setting simulation parameters and conditions; performing fluid dynamics simulation in the simulation environment according to the simulation parameters and conditions to obtain fluid flow data in the pore network corresponding to the target core image during the simulation process, wherein the flow data includes at least one of the following: velocity data and pressure distribution data; and visualizing the flow data based on the target core image to obtain a target distribution image, wherein the distribution image includes at least one of the following: a three-dimensional pressure distribution image and a three-dimensional velocity distribution image.

[0013] According to another aspect of the embodiments of this application, an image processing apparatus is also provided, comprising: a residual determination module, configured to acquire an original core image of a target rock and determine a residual image corresponding to the original core image, wherein the residual image is used to supplement missing details in the original core image; a size adjustment module, configured to perform a magnification operation on the original core image and the residual image, wherein the magnification operation is used to adjust the size of the original core image and the residual image to a target size; an image reconstruction module, configured to compensate and reconstruct the original core image of the target size based on the residual image of the target size, to obtain a reconstructed core image; and a threshold segmentation module, configured to perform threshold segmentation processing on the reconstructed core image to obtain a target core image, wherein the threshold segmentation processing is used to distinguish between pores and rock parts in the image frame, and the resolution of the target core image is higher than the resolution of the original core image.

[0014] According to another aspect of the embodiments of this application, an electronic device is also provided, including: a memory and a processor, the processor being configured to run a program stored in the memory, wherein the program executes an image processing method during runtime.

[0015] According to another aspect of the embodiments of this application, a non-volatile storage medium is also provided, the non-volatile storage medium including a stored computer program, wherein the device where the non-volatile storage medium is located executes an image processing method by running the computer program.

[0016] According to another aspect of the embodiments of this application, a computer program product is also provided, including a computer program that, when executed by a processor, implements the steps of an image processing method.

[0017] In this embodiment, the original core image of the target rock is acquired, and a residual image corresponding to the original core image is determined. The residual image is used to supplement the missing details in the original core image. The original core image and the residual image are magnified to adjust their sizes to a target size. Based on the residual image of the target size, the original core image of the target size is reconstructed to obtain a reconstructed core image. The reconstructed core image is then subjected to threshold segmentation to obtain the target core image. The threshold segmentation distinguishes between pores and rock parts in the image frame. The target core image has a higher resolution than the original core image. By combining deep learning technology with residual image compensation, the resolution and detail restoration capabilities of digital core images are effectively improved. This solves the technical problems of insufficient resolution improvement and inaccurate detail reproduction in related technologies for core image super-resolution reconstruction when dealing with complex shale structures and micron-level cracks. Attached Figure Description

[0018] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0019] Figure 1 This is a hardware structure block diagram of a computer terminal (or electronic device) for implementing an image processing method according to an embodiment of this application;

[0020] Figure 2 This is a schematic diagram of an image processing method flow according to an embodiment of this application;

[0021] Figure 3 This is a schematic diagram of a scanned original binary image provided according to an embodiment of this application;

[0022] Figure 4 This is a schematic diagram of a reconstructed VDSR image and a thresholded segmented image provided according to an embodiment of this application;

[0023] Figure 5 This is a schematic diagram of a three-dimensional pressure distribution image provided according to an embodiment of this application;

[0024] Figure 6 This is a schematic diagram of a three-dimensional velocity distribution image provided according to an embodiment of this application;

[0025] Figure 7 This is a schematic diagram of the structure of an image processing apparatus provided according to an embodiment of this application. Detailed Implementation

[0026] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0027] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0028] Image super-resolution technology is an effective method to improve image resolution. While traditional methods in this field are simple and easy to use, their effectiveness is limited, often leading to blurred edges and loss of detail. In recent years, deep learning methods, such as convolutional neural networks, residual networks, and generative adversarial networks, have been applied to the field of digital core analysis. However, these technologies are mostly applied to relatively homogeneous sandstone and carbonate rocks, with limited application to highly heterogeneous shale with complex pore structures, as well as rocks containing hydraulic fracturing fractures. Nanoscale shale matrix and micron-scale hydraulic fracturing fractures, although different in scale, both belong to highly sensitive porous and fractured media: minute distortions at nanopore throats and fracture boundaries can significantly alter pore connectivity and the main seepage channels, thereby amplifying errors in the calculation of physical properties such as permeability.

[0029] In summary, digital core image super-resolution reconstruction technology has several drawbacks when dealing with complex shale structures and micron-level fractures, including insufficient resolution enhancement, inaccurate detail reproduction, and failure to effectively preserve fracture connectivity and matrix anisotropy.

[0030] To address the aforementioned issues, this application provides a solution: a digital rock depth super-resolution (VDSR) reconstruction method. This method employs residual learning to recover high-frequency details and extracts and fuses structural information in the x, y, and z directions. It can take into account both matrix anisotropy and fracture orientation characteristics, avoiding edge blurring, fracture breakage, or pseudo-connectivity problems caused by traditional interpolation. This method can efficiently reconstruct high-resolution, high-fidelity three-dimensional digital core images from low-resolution scan images, highly restoring the fine structure of nanoscale shale matrix and micron-scale fractures, and significantly improving the calculation accuracy of physical properties such as permeability. A detailed description follows.

[0031] According to an embodiment of this application, an image processing method embodiment is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0032] The methods and embodiments provided in this application can be executed on mobile terminals, computer terminals, or similar computing devices. Figure 1 A hardware block diagram of a computer terminal (or electronic device) for implementing an image processing method is shown. Figure 1 As shown, the computer terminal 10 (or electronic device) may include one or more processors 102 (shown as 102a, 102b, ..., 102n in the figure) 102 (processor 102 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0033] It should be noted that the aforementioned one or more processors 102 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10 (or electronic device). As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0034] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the image processing method in this embodiment. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the above-mentioned image processing method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0035] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0036] The display may be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10 (or electronic device).

[0037] Under the above operating environment, this application provides an image processing method. Figure 2 This is a schematic diagram of an image processing method flow according to an embodiment of this application, such as... Figure 2 As shown, the method includes the following steps:

[0038] Step S202: Obtain the original core image of the target rock and determine the residual image corresponding to the original core image, wherein the residual image is used to supplement the missing details in the original core image;

[0039] Step S204: Perform a magnification operation on the original core image and the residual image, wherein the magnification operation is used to adjust the size of the original core image and the residual image to the target size;

[0040] Step S206: Based on the residual image of the target size, the original core image of the target size is compensated and reconstructed to obtain the reconstructed core image;

[0041] Step S208: Threshold segmentation is performed on the reconstructed core image to obtain the target core image. The threshold segmentation is used to distinguish the pores and rock parts of the image frame. The resolution of the target core image is higher than that of the original core image.

[0042] Through the above steps, by combining deep learning technology with residual image compensation, the resolution and detail reproduction capabilities of digital core images are effectively improved. This technology is specifically designed to generate high-resolution three-dimensional digital rock images of nanoscale shale matrix and micron-scale hydraulic fracturing, making the reconstructed images highly similar to the actual images, reducing errors in permeability calculations, and thus solving the technical problems of insufficient resolution improvement and inaccurate detail reproduction in core image super-resolution reconstruction technology when dealing with complex shale structures and micron-scale fractures.

[0043] The image processing method in steps S202 to S208 of the embodiments of this application will be further described below.

[0044] First, obtain the original core image of the target rock (taking shale as an example). The specific steps are as follows.

[0045] In some embodiments of this application, obtaining the original core image of the target rock includes the following steps: scanning the target rock with a scanning device to obtain a three-dimensional scan image, wherein the scanning device includes at least one of the following: a focused ion beam scanning electron microscope and a micro CT scanner; extracting two-dimensional binary images in the target direction from the three-dimensional scan image, and determining the two-dimensional binary images as the original core image, wherein the target direction includes: the x direction, the y direction, and the z direction, the x direction, the y direction, and the z direction are three mutually orthogonal directions in three-dimensional space, and each target direction corresponds to a set of two-dimensional binary images.

[0046] Specifically, high-precision scanning equipment such as focused ion beam scanning electron microscopy (FIB-SEM) or microscopic CT scanners can be used to comprehensively scan the target rock, obtaining three-dimensional scan images. Subsequently, specific two-dimensional binary images in the x, y, and z directions are selected from the three-dimensional scan images as the basis for the original core images. The x, y, and z directions represent three mutually perpendicular dimensions in three-dimensional space, and each set of directions is associated with a set of two-dimensional binary images. This selective extraction of two-dimensional images not only simplifies the data processing flow but also ensures that information in each direction is fully preserved, providing comprehensive data support for subsequent deep learning training. Through precise two-dimensional binary image extraction, the subtle features of anisotropic rocks can be fully captured during the model training stage, especially for nanoscale pore throats and micron-scale cracks in shale. This allows for higher-precision image detail recovery in the final super-resolution reconstruction process, reducing errors in physical property calculations and improving the accuracy of rock image reconstruction.

[0047] Then, a deep learning model can be used to determine the residual image corresponding to the original core image. The specific steps are as follows.

[0048] In some embodiments of this application, determining the residual image corresponding to the original core image includes: analyzing the original core image using a target prediction model to obtain a residual image, wherein a set of two-dimensional binary images for each target direction corresponds to a set of two-dimensional residual images; the training step of the target prediction model includes: obtaining a training dataset, wherein the training dataset includes: multiple first core image samples with a first resolution, and second core image samples with a second resolution corresponding to each first core image sample, the second resolution being higher than the first resolution; training the initial prediction model based on the training dataset to enable the model to learn how to recover the details of a high-resolution image from a low-resolution image, thereby obtaining the target prediction model.

[0049] Specifically, a target prediction model can be used to analyze the original core images and generate a two-dimensional residual image matching each target direction. The training process of this model is based on a dataset containing core image samples at different resolution levels. Each set of low-resolution first core image samples is paired with a corresponding high-resolution second core image sample, with the second resolution being significantly higher than the first. Through comparative learning with these samples, the initial prediction model is gradually optimized, mastering the ability to recover high-resolution image details from low-resolution images, and ultimately evolving into a target prediction model.

[0050] The deep learning-based technical solution in this application can intelligently capture and reproduce subtle features in images, especially for complex structures such as shale matrix and hydraulic fracturing fractures, significantly improving the accuracy and reliability of image super-resolution reconstruction. In practical applications, the trained model can effectively extract and reconstruct the detailed information of high-resolution images from low-resolution core images, reducing errors in physical property analysis caused by insufficient image resolution, and providing strong support for the accurate calculation of key parameters such as porosity and permeability. In subsequent embodiments, the training and use of the model can be further extended to more dimensions or different types of rock samples to adapt to a wider range of application scenarios.

[0051] After obtaining the original core image and the corresponding two-dimensional residual image, the image size of both can be adjusted in one step, as follows.

[0052] In some embodiments of this application, the magnification operation of the original core image and the residual image includes: using a bicubic interpolation algorithm to magnify the original core image in each target direction to obtain the original core image of the target size corresponding to each target direction; and using a bicubic interpolation algorithm to magnify the residual image in each target direction to obtain the residual image of the target size corresponding to each target direction.

[0053] Specifically, a bicubic interpolation algorithm is used to magnify the original core image and the residual image. This includes magnifying the original core image in each target direction to obtain an image corresponding to the target size in each direction; similarly, the residual image in each target direction is magnified to obtain a residual image corresponding to the target size in each direction.

[0054] This step ensures that the image maintains good edge smoothness and detail integrity during the magnification process, making the magnified image visually and structurally closer to the original image. The bicubic interpolation algorithm, with its excellent performance in image magnification, not only effectively preserves the texture features of the original image but also reduces blockiness and jagged edges during magnification, thus achieving high-quality image size expansion in the x, y, and z directions, providing a high-quality base image for subsequent fusion.

[0055] After the image size adjustment is completed, the difference between the reconstructed image and the real image can be minimized by fusing the residual image with the base image. The specific steps are as follows.

[0056] In some embodiments of this application, the process of compensating and reconstructing the original core image of the target size based on the residual image of the target size to obtain the reconstructed core image includes: fusing the original core images of the target size corresponding to different target directions through arithmetic averaging to obtain a fused three-dimensional image; determining the weight coefficients corresponding to the residual images of the target size in different target directions, wherein the weight coefficients are used to characterize the degree of influence of the residual images of different target directions on image reconstruction, and each target direction corresponds to a weight coefficient; determining a three-dimensional compensation image based on the weight coefficients of different target directions and the residual image of the target size, and using the three-dimensional compensation image to compensate the fused three-dimensional image to obtain the reconstructed core image.

[0057] Specifically, the original core images of different target sizes corresponding to different target directions can be fused using an arithmetic average process to generate an initial fused 3D image, as shown in the following formula:

[0058]

[0059] in, This is a bicubic interpolated image (i.e., the fused 3D image mentioned above). , and These represent bitriplexed images in the x, y, and z directions (i.e., original core images of the target size corresponding to different target directions).

[0060] Subsequently, directional correction coefficients for three-directional residual fusion can be introduced, which are weighting coefficients for the residual images of different target sizes in different target directions. These weighting coefficients quantify the contribution of the residual images in each direction to the image reconstruction process. They can be determined by minimizing the reconstruction error on a calibration dataset or by automatic learning through optimization algorithms. Using these weighting coefficients, the reconstruction accuracy of pore boundaries and fracture connectivity can be improved, the influence of binary digital core images on the reconstruction results can be minimized, and the difference between the reconstructed high-resolution image and the original image can be minimized. Then, based on the weighting coefficients and the residual images of the target size, a three-dimensional compensation image is calculated. This three-dimensional compensation image is then applied to the fused three-dimensional image for precise compensation, ultimately obtaining the reconstructed core image, as shown in the following formula:

[0061]

[0062] in, This refers to the VDSR image (i.e., the three-dimensional compensated image mentioned above). , and These represent the bicubic interpolation images (i.e., residual images of target size in different target directions) in the x, y, and z directions, respectively. , and This represents the correction coefficient (i.e., the weighting coefficient mentioned above).

[0063] The embodiments of this application optimize the quality of image reconstruction through the above-mentioned fusion strategy. By adaptively adjusting the influence of residual images in different directions, the reconstructed core images more accurately reflect the true structure of nanoscale shale matrix and micron-scale hydraulic fracturing fractures, further improving the accuracy of physical property parameter calculation.

[0064] The reconstructed core image can be further processed by threshold segmentation, and the specific steps are as follows.

[0065] In some embodiments of this application, threshold segmentation processing is performed on the reconstructed core image to obtain a target core image, including: determining the porosity of the target rock and determining a target grayscale threshold based on the porosity; traversing each pixel in the reconstructed core image and segmenting the pixels according to the target grayscale threshold to obtain the target core image, wherein, during the segmentation process, pixels with grayscale values ​​not less than the target grayscale threshold are marked as pores, and pixels with grayscale values ​​less than the target grayscale threshold are marked as rocks.

[0066] Specifically, the porosity of the target rock can be measured first, and a target grayscale threshold can be determined based on this. Then, each pixel in the reconstructed core image can be systematically traversed, and these pixels can be classified according to the set target grayscale threshold. For example, pixels with grayscale values ​​that reach or exceed the target grayscale threshold are regarded as porous parts, while pixels with grayscale values ​​that are lower than the threshold are marked as rock parts.

[0067] The segmentation strategy of this application can effectively distinguish between rocks and pores, ensuring accurate image resolution and providing accurate data support for subsequent physical property analysis, thereby improving computational accuracy and reliability. In this way, the reconstructed image not only presents a clear structural contrast visually, but also, at the quantitative analysis level, more accurately reflects the true physical properties of the rock.

[0068] Furthermore, embodiments of this application can also utilize the final obtained target core image for simulation, and visualize the flow velocity and pressure distribution in the simulation, as detailed below.

[0069] In some embodiments of this application, the method further includes: importing the target core image into a simulation environment and setting simulation parameters and conditions; performing fluid dynamics simulation in the simulation environment according to the simulation parameters and conditions to obtain fluid flow data in the pore network corresponding to the target core image during the simulation process, wherein the flow data includes at least one of the following: velocity data and pressure distribution data; and visualizing the flow data based on the target core image to obtain a target distribution image, wherein the distribution image includes at least one of the following: a three-dimensional pressure distribution image and a three-dimensional velocity distribution image.

[0070] Specifically, target core images can be imported into a professional simulation environment, and fluid dynamics simulation experiments can be conducted based on specific simulation parameters and conditions. During this process, the flow state of the fluid in the pore network corresponding to the target core image is recorded in detail, forming a complete set of velocity and pressure distribution data. Based on this data, intuitive three-dimensional pressure and velocity distribution images can be constructed using data visualization technology, providing strong support for in-depth analysis of the fluid dynamic characteristics inside the core.

[0071] This method, which combines super-resolution reconstructed images with fluid dynamics simulations, not only verifies the accuracy and reliability of image reconstruction but also reveals the impact of complex rock microstructures on fluid seepage behavior. This helps to optimize oil and gas reservoir development strategies and improve resource extraction efficiency.

[0072] To help those skilled in the art better understand the process steps in the embodiments of this application, the above process steps are further illustrated below with examples.

[0073] like Figure 3 As shown, the original binary images (original core images) of shale were obtained using focused ion beam scanning electron microscopy (FIB-SEM) and microscopic CT scanning. The two-dimensional binary images of shale in the x, y and z directions were analyzed using a trained target prediction model (e.g., VDSR model) to obtain two-dimensional residual images in these three directions.

[0074] Then, the sizes of the low-resolution image and the residual image can be adjusted in the x, y, and z directions. First, bicubic interpolation is applied to the 60×60 2D low-resolution image and the residual image to obtain a 300×300 image in the x, y, and z directions. Then, each layer of the original 60-layer image is copied five times, and the x, y, and z residual image is converted from 60 layers to 300 layers, resulting in a 300×300 bicubic interpolated image and residual image in the x, y, and z directions.

[0075] Subsequently, the concept of arithmetic mean can be applied to the bicubic interpolated images (the original core image of the target size) in the x, y, and z directions to reconstruct a 300×300×300 bicubic interpolated image (fused 3D image). During the VDSR reconstruction process, three α parameters (weighting coefficients) are introduced to ensure that the difference between the reconstructed high-resolution image and the original image is as small as possible, resulting in the reconstructed core image.

[0076] Finally, the porosity of the original binary image and the VDSR image (reconstructed core image) can be used as a threshold criterion to perform threshold segmentation on the image, ultimately obtaining the following result: Figure 4 The target core image is shown. Furthermore, the target core image can be used for simulation, and the flow velocity and pressure distribution in the simulation can be visualized to obtain a three-dimensional pressure distribution image of the shale (e.g., Figure 5 (as shown) and three-dimensional velocity distribution images (such as) Figure 6 (As shown).

[0077] This proposed solution, combining deep learning technology with residual image compensation, effectively improves the resolution and detail reproduction capability of digital core images, particularly excelling in processing highly heterogeneous shale with complex pore structures and rocks containing hydraulically fractured fractures. This solution not only enhances image clarity but also achieves significant progress in reconstructing pore boundaries and fracture connectivity, greatly reducing errors in calculating physical property parameters. Practical applications demonstrate that this solution can significantly improve the calculation accuracy of key physical properties such as permeability, providing more reliable technical support for the refined evaluation and development of oil and gas reservoirs.

[0078] According to an embodiment of this application, an embodiment of an image processing apparatus is also provided. Figure 7 This is a schematic diagram of the structure of an image processing apparatus according to an embodiment of this application. Figure 7 As shown, the device includes:

[0079] The residual determination module 70 is used to acquire the original core image of the target rock and determine the residual image corresponding to the original core image. The residual image is used to supplement the missing details in the original core image.

[0080] The size adjustment module 72 is used to perform magnification operations on the original core image and the residual image, wherein the magnification operation is used to adjust the size of the original core image and the residual image to the target size;

[0081] The image reconstruction module 74 is used to compensate and reconstruct the original core image of the target size based on the residual image of the target size, so as to obtain the reconstructed core image.

[0082] The threshold segmentation module 76 is used to perform threshold segmentation processing on the reconstructed core image to obtain the target core image. The threshold segmentation processing is used to distinguish the pores and rock parts of the image frame. The resolution of the target core image is higher than that of the original core image.

[0083] Optionally, obtaining the original core image of the target rock includes: scanning the target rock with a scanning device to obtain a three-dimensional scan image, wherein the scanning device includes at least one of the following: a focused ion beam scanning electron microscope and a micro CT scanner; extracting two-dimensional binary images in the target direction from the three-dimensional scan image, and determining the two-dimensional binary images as the original core image, wherein the target direction includes: the x direction, the y direction, and the z direction, the x direction, the y direction, and the z direction are three mutually orthogonal directions in three-dimensional space, and each target direction corresponds to a set of two-dimensional binary images.

[0084] Optionally, determining the residual image corresponding to the original core image includes: analyzing the original core image using a target prediction model to obtain the residual image, wherein a set of two-dimensional binary images for each target direction corresponds to a set of two-dimensional residual images; the training steps of the target prediction model include: obtaining a training dataset, wherein the training dataset includes: multiple first core image samples with a first resolution, and second core image samples with a second resolution corresponding to each first core image sample, the second resolution being higher than the first resolution; training the initial prediction model based on the training dataset to enable the model to learn how to recover the details of a high-resolution image from a low-resolution image, thereby obtaining the target prediction model.

[0085] Optionally, the magnification operation on the original core image and the residual image includes: using a bicubic interpolation algorithm to magnify the original core image in each target direction to obtain the original core image of the target size corresponding to each target direction; and using a bicubic interpolation algorithm to magnify the residual image in each target direction to obtain the residual image of the target size corresponding to each target direction.

[0086] Optionally, based on the residual image of the target size, the original core image of the target size is compensated and reconstructed to obtain the reconstructed core image. This includes: fusing the original core images of the target size corresponding to different target directions through arithmetic averaging to obtain a fused three-dimensional image; determining the weight coefficients corresponding to the residual images of the target size in different target directions, wherein the weight coefficients are used to characterize the degree of influence of the residual images of different target directions on image reconstruction, and each target direction corresponds to a weight coefficient; determining a three-dimensional compensation image based on the weight coefficients of different target directions and the residual image of the target size, and using the three-dimensional compensation image to compensate the fused three-dimensional image to obtain the reconstructed core image.

[0087] Optionally, threshold segmentation processing is performed on the reconstructed core image to obtain the target core image, including: determining the porosity of the target rock and determining the target grayscale threshold based on the porosity; traversing each pixel in the reconstructed core image and segmenting the pixels according to the target grayscale threshold to obtain the target core image, wherein, during the segmentation process, pixels with grayscale values ​​not less than the target grayscale threshold are marked as pores, and pixels with grayscale values ​​less than the target grayscale threshold are marked as rock.

[0088] Optionally, the image processing device is further configured to: import the target core image into a simulation environment and set simulation parameters and conditions; perform fluid dynamics simulation in the simulation environment according to the simulation parameters and conditions to obtain fluid flow data in the pore network corresponding to the target core image during the simulation process, wherein the flow data includes at least one of the following: velocity data and pressure distribution data; and perform visualization processing on the flow data based on the target core image to obtain a target distribution image, wherein the distribution image includes at least one of the following: a three-dimensional pressure distribution image and a three-dimensional velocity distribution image.

[0089] It should be noted that each module in the above-mentioned image processing device can be a program module (for example, a set of program instructions to implement a certain function) or a hardware module. For the latter, it can be manifested in the following forms, but is not limited to them: each of the above-mentioned modules is manifested as a processor, or the functions of each of the above-mentioned modules are implemented by a processor.

[0090] It should be noted that the image processing apparatus provided in this embodiment can be used to perform... Figure 2 The image processing method shown above is also applicable to the embodiments of this application, and will not be repeated here.

[0091] This application embodiment also provides a non-volatile storage medium, which includes a stored computer program. The device containing the non-volatile storage medium executes the following image processing method by running the computer program: acquiring an original core image of the target rock and determining a residual image corresponding to the original core image, wherein the residual image is used to supplement missing details in the original core image; performing a magnification operation on the original core image and the residual image, wherein the magnification operation is used to adjust the size of the original core image and the residual image to a target size; based on the residual image of the target size, performing compensation reconstruction on the original core image of the target size to obtain a reconstructed core image; performing threshold segmentation processing on the reconstructed core image to obtain the target core image, wherein the threshold segmentation processing is used to distinguish between pores and rock parts in the image frame, and the resolution of the target core image is higher than the resolution of the original core image.

[0092] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the image processing method described in various embodiments of this application: acquiring an original core image of the target rock and determining a residual image corresponding to the original core image, wherein the residual image is used to supplement missing details in the original core image; performing a magnification operation on the original core image and the residual image, wherein the magnification operation is used to adjust the size of the original core image and the residual image to a target size; performing compensation reconstruction on the original core image of the target size based on the residual image of the target size to obtain a reconstructed core image; performing threshold segmentation processing on the reconstructed core image to obtain a target core image, wherein the threshold segmentation processing is used to distinguish between pores and rock parts in the image frame, and the resolution of the target core image is higher than the resolution of the original core image.

[0093] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0094] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0095] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0096] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0097] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0098] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0099] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. An image processing method, characterized in that, include: Obtain the original core image of the target rock and determine the residual image corresponding to the original core image, wherein the residual image is used to supplement the missing details in the original core image; The original core image and the residual image are magnified, wherein the magnification operation is used to adjust the size of the original core image and the residual image to a target size; Based on the residual image of the target size, the original core image of the target size is compensated and reconstructed to obtain the reconstructed core image; The reconstructed core image is subjected to threshold segmentation to obtain a target core image. The threshold segmentation is used to distinguish between pores and rock parts in the image frame. The resolution of the target core image is higher than that of the original core image.

2. The image processing method according to claim 1, characterized in that, Obtaining raw core images of the target rock includes: The target rock is scanned using a scanning device to obtain a three-dimensional scan image, wherein the scanning device includes at least one of the following: a focused ion beam scanning electron microscope and a microscopic CT scanner; Two-dimensional binary images are extracted from the target direction in the three-dimensional scan image, and the two-dimensional binary images are determined as the original core image. The target direction includes the x direction, y direction, and z direction. The x direction, y direction, and z direction are three mutually orthogonal directions in three-dimensional space. Each target direction corresponds to a set of two-dimensional binary images.

3. The image processing method according to claim 2, characterized in that, Determining the residual image corresponding to the original core image includes: The original core image is analyzed using a target prediction model to obtain the residual image, wherein a set of two-dimensional binary images for each target direction corresponds to a set of two-dimensional residual images; The training steps of the target prediction model include: obtaining a training dataset, wherein the training dataset includes: multiple first core image samples with a first resolution, and a second core image sample with a second resolution corresponding to each first core image sample, wherein the second resolution is higher than the first resolution; and training an initial prediction model based on the training dataset so that the model learns how to recover details of a high-resolution image from a low-resolution image, thereby obtaining the target prediction model.

4. The image processing method according to claim 3, characterized in that, The magnification operation on the original core image and the residual image includes: A bicubic interpolation algorithm is used to magnify the original core image for each target direction to obtain the original core image of the target size corresponding to each target direction; The bicubic interpolation algorithm is used to magnify the residual image for each target direction to obtain the residual image of the target size corresponding to each target direction.

5. The image processing method according to claim 4, characterized in that, Based on the residual image of the target size, the original core image of the target size is compensated and reconstructed to obtain the reconstructed core image, including: By using arithmetic averaging, the original core images of the target size corresponding to different target directions are fused to obtain a fused three-dimensional image; Determine the weight coefficients corresponding to the residual images of the target size for different target directions, wherein the weight coefficients are used to characterize the degree of influence of the residual images of different target directions on image reconstruction, and each target direction corresponds to one weight coefficient; Based on the weighting coefficients of the different target directions and the residual images of the target size, a three-dimensional compensation image is determined, and the three-dimensional compensation image is used to compensate the fused three-dimensional image to obtain the reconstructed core image.

6. The image processing method according to claim 1, characterized in that, The reconstructed core image is subjected to threshold segmentation to obtain the target core image, which includes: Determine the porosity of the target rock, and based on the porosity, determine the target grayscale threshold; Each pixel in the reconstructed core image is traversed, and the pixels are segmented according to the target grayscale threshold to obtain the target core image. During the segmentation process, pixels with grayscale values ​​not less than the target grayscale threshold are marked as pores, and pixels with grayscale values ​​less than the target grayscale threshold are marked as rocks.

7. The image processing method according to claim 1, characterized in that, The method further includes: Import the target core image into the simulation environment and set the simulation parameters and conditions; Based on the simulation parameters and conditions, a fluid dynamics simulation is performed in the simulation environment to obtain the flow data of the fluid in the pore network corresponding to the target core image during the simulation process. The flow data includes at least one of the following: flow velocity data and pressure distribution data. Based on the target core image, the flow data is visualized to obtain a target distribution image, wherein the distribution image includes at least one of the following: a three-dimensional pressure distribution image and a three-dimensional velocity distribution image.

8. An image processing apparatus, characterized in that, include: The residual determination module is used to acquire the original core image of the target rock and determine the residual image corresponding to the original core image, wherein the residual image is used to supplement the missing details in the original core image; The size adjustment module is used to magnify the original core image and the residual image, wherein the magnification operation is used to adjust the size of the original core image and the residual image to a target size; The image reconstruction module is used to compensate and reconstruct the original core image of the target size based on the residual image of the target size, so as to obtain the reconstructed core image; A threshold segmentation module is used to perform threshold segmentation processing on the reconstructed core image to obtain a target core image. The threshold segmentation processing is used to distinguish between pores and rock parts in the image frame. The resolution of the target core image is higher than that of the original core image.

9. An electronic device, characterized in that, include: A memory and a processor, the processor being configured to run a program stored in the memory, wherein the program, when running, performs the image processing method according to any one of claims 1 to 7.

10. A non-volatile storage medium, characterized in that, The non-volatile storage medium includes a stored computer program, wherein the device containing the non-volatile storage medium executes the image processing method according to any one of claims 1 to 7 by running the computer program.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the image processing method according to any one of claims 1 to 7.

Images