A method and system for reconstructing a three-dimensional particle system
By combining the Hough circle detection method and the circle-by-circle radius scanning method with filtering and segmentation techniques, the problem of poor quality of two-dimensional tomographic images in the reconstruction of three-dimensional particle systems was solved, and efficient and accurate reconstruction of three-dimensional particle systems was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- QINGDAO UNIV OF TECH
- Filing Date
- 2023-06-05
- Publication Date
- 2026-07-21
Smart Images
Figure CN116681832B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of microscopic technology of particulate systems, and in particular to a method and system for reconstructing three-dimensional particulate systems. Background Technology
[0002] The statements in this section merely refer to the background art related to this invention and do not necessarily constitute prior art.
[0003] Particulate systems are ubiquitous in people's daily lives. Essentially, they are nonlinear systems composed of a large number of discrete, irregularly ordered solid particles, existing somewhere between ideal fluid systems and ideal solid systems. Therefore, they cannot be studied using general solid or fluid theories.
[0004] Scholarly research on particulate matter can be divided into three aspects: theoretical research on interparticle contact at the microscale, research on force chains at the mesoscale, and research on the mechanical behavior of particulate systems at the macroscale. Many mechanical properties exhibited by three-dimensional particulate systems at the macroscale are determined by the complex mechanical responses of the force chain network at the mesoscale; therefore, the study of force chains at the mesoscale has become a key link between macroscopic and microscopic research. From a dimensional perspective, research on two-dimensional particulate systems is relatively complete, but in the study of three-dimensional particulate systems, there are no suitable experimental methods. Most scholars use numerical simulation methods for research; therefore, experimental research on three-dimensional particulate systems still has a long way to go.
[0005] Among existing research methods, CT scanning, magnetic resonance imaging, and confocal imaging are the only few non-destructive testing methods that can investigate the stress on three-dimensional materials. However, these methods require expensive equipment, are not widely available, and have significant limitations. Therefore, we chose to use refractive index matching tomography to conduct compression experiments on a three-dimensional particle system, obtain two-dimensional tomographic information of the particle system, and analyze and reconstruct the sequence of two-dimensional tomographic information to obtain the three-dimensional particle system.
[0006] In refractive index matching tomography experiments, if the laser intensity and thickness are optimally adjusted and there are no impurities such as bubbles within the specimen, theoretically, a clear two-dimensional tomographic image without thickness issues can be captured. However, in actual experiments or applications, conditions are not always ideal, and the quality of the obtained two-dimensional tomographic images is often unsatisfactory. The images frequently exhibit unclear outlines, missing outlines, excessively bright spots, and other noise problems, affecting the efficiency and accuracy of three-dimensional particle system reconstruction. Summary of the Invention
[0007] To address the shortcomings of existing technologies, this invention provides a method, system, electronic device, and computer-readable storage medium for reconstructing a three-dimensional particle system. It utilizes the Hough circle detection method to obtain the center coordinates of the circles in each cross section of a two-dimensional tomographic image, and proposes a circle-by-circle radius scanning method to obtain the particle edges. By obtaining the radius values of the circles in each cross section of the image, the particle system can be reconstructed in three dimensions.
[0008] In a first aspect, the present invention provides a method for reconstructing a three-dimensional particle system;
[0009] A method for reconstructing a three-dimensional particle system, comprising:
[0010] Two-dimensional tomographic images of a three-dimensional particle system are acquired, and the two-dimensional tomographic images are denoised; wherein, the two-dimensional tomographic images contain information about the circular cross-sections of the particles in the three-dimensional particle system;
[0011] The denoised two-dimensional tomographic image is scanned using the Hough transform circle detection method to determine the center coordinates of the particle cross-section circles in the two-dimensional tomographic image; using the center coordinates as a reference, the median radius of the particle cross-section circles in the two-dimensional tomographic image is obtained by the circle-by-circle radius scanning method; and a reference two-dimensional tomographic image is obtained based on the center coordinates and median radius of the particle cross-section circles.
[0012] Based on the baseline two-dimensional tomographic image, a three-dimensional view of the three-dimensional particle system is obtained.
[0013] Furthermore, the denoising process for the two-dimensional tomographic image includes:
[0014] Construct a discrete filter, and convolve the two-dimensional tomographic image with the discrete filter to obtain the discretely filtered two-dimensional tomographic image.
[0015] Threshold segmentation is performed on the discretely filtered two-dimensional tomographic image to obtain a denoised two-dimensional tomographic image.
[0016] Preferably, constructing the discrete filter includes:
[0017] Construct a zero-order function and sample it to obtain the circle center coordinate array;
[0018] Discrete filters are constructed using the discrete trigonometric function method based on the circle center coordinate array.
[0019] Furthermore, the step of obtaining the median radius of the particle cross-section circle in the two-dimensional tomographic image by using the circle center coordinates as a reference and employing a circle-by-circle radius scanning method includes:
[0020] Based on the noise-reduced two-dimensional tomographic image, using the center coordinates of the circle as a reference, and according to the gray value of the pixel, all pixels within a preset radius are traversed to determine the inner and outer diameters of the particle cross-section circle.
[0021] The midline radius of the particle cross-section circle is obtained based on the inner and outer diameters of the particle cross-section circle.
[0022] Furthermore, the specific steps for obtaining the reference two-dimensional tomographic image based on the center coordinates of the particle cross-section circle and the median radius are as follows:
[0023] The radius of the midline of the particle cross-section circle is used as the radius of the particle cross-section circle to determine the size information of the particle cross-section circle;
[0024] Based on the center coordinates and dimensions of the particle cross-section circles, the position information of the particle cross-section circles is determined; based on the position information of all particle cross-section circles in the two-dimensional tomographic image, a reference two-dimensional tomographic image is obtained.
[0025] Furthermore, the step of obtaining a three-dimensional view of the three-dimensional particle system based on the reference two-dimensional tomographic image specifically involves: performing interactive threshold segmentation on the reference two-dimensional tomographic image using three-dimensional visualization reconstruction software to obtain a three-dimensional view of the three-dimensional particle system.
[0026] Furthermore, it also includes:
[0027] Based on the three-dimensional view of the three-dimensional particle system, a three-dimensional coordinate system is constructed to obtain the center coordinates and radius of each particle in the three-dimensional particle system.
[0028] Secondly, the present invention provides a three-dimensional particle system reconstruction system;
[0029] A three-dimensional particle system reconstruction system, comprising:
[0030] The two-dimensional tomographic image acquisition module is configured to: acquire two-dimensional tomographic images of a three-dimensional particle system and perform noise reduction processing on the two-dimensional tomographic images; wherein, the two-dimensional tomographic images contain particle cross-sectional circular information of the three-dimensional particle system;
[0031] The particle cross-section circle information acquisition module is configured to: scan the denoised two-dimensional tomographic image using the Hough transform circle detection method to determine the center coordinates of the particle cross-section circle in the two-dimensional tomographic image; use the center coordinates as a reference to obtain the median radius of the particle cross-section circle in the two-dimensional tomographic image using the circle-by-circle radius scanning method; and obtain a reference two-dimensional tomographic image based on the center coordinates and median radius of the particle cross-section circle.
[0032] The 3D visualization reconstruction module is configured to: obtain a 3D view of a 3D particle system based on a baseline 2D tomographic image.
[0033] Thirdly, the present invention provides an electronic device;
[0034] An electronic device includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, wherein the computer instructions, when executed by the processor, complete the steps of the above-described three-dimensional particle system reconstruction method.
[0035] Fourthly, the present invention provides a computer-readable storage medium;
[0036] A computer-readable storage medium for storing computer instructions, which, when executed by a processor, complete the steps of the above-described three-dimensional particle system reconstruction method.
[0037] Compared with the prior art, the beneficial effects of the present invention are:
[0038] The technical solution provided by this invention utilizes the Hough circle detection method to obtain the center coordinates of each cross-sectional circle in a two-dimensional tomographic image. At the same time, it proposes a circle-by-circle radius scanning method to obtain the particle edge, eliminating the influence of the laser surface thickness in refractive index matching tomography and accurately obtaining the radius value of each cross-sectional circle in the image. It can accurately obtain the position and distribution information of the three-dimensional particle system and efficiently and accurately reconstruct the three-dimensional particle system. Attached Figure Description
[0039] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.
[0040] Figure 1 This is a flowchart provided for an embodiment of the present invention;
[0041] Figure 2 This is a schematic diagram of a two-dimensional tomographic image provided in an embodiment of the present invention, wherein (a) is the original image; and (b) is the image after noise reduction processing.
[0042] Figure 3 The following is a schematic diagram of the cross-sectional circle of particles on a two-dimensional tomographic image provided in the embodiments of the present invention, wherein (a) is a schematic diagram of particles activated by a laser sheet, (b) is a schematic diagram of particles recorded by a camera, and (c) is an example of particles recorded by a camera in an experiment.
[0043] Figure 4 This is a schematic diagram of reducing the thickness of the laser surface to the laser midline provided in an embodiment of the present invention;
[0044] Figure 5 The diagram below illustrates the principle of the radius-by-radius scanning method provided in this embodiment of the invention, wherein (a) is a schematic diagram of the radius-by-radius scanning method and (b) is a schematic diagram of the results of the radius-by-radius scanning method.
[0045] Figure 6This is an example diagram illustrating the results of the radius-by-radius scanning method provided in an embodiment of the present invention;
[0046] Figure 7 The following is an example diagram of the two-dimensional tomographic image processing process provided in the embodiments of the present invention, wherein (a) is the original two-dimensional tomographic image recorded by the CCD camera, (b) is the two-dimensional tomographic image after noise reduction processing; (c) is a diagram of the position of each circle on the two-dimensional tomographic image identified by the Hough transform circle detection method, and (d) is a diagram of the position of each circle re-determined after using the circle-by-circle radius scanning method. The numbers marked on each circle in the diagram are the changes in radius after the circle-by-circle radius scanning method.
[0047] Figure 8 A three-dimensional view of the reconstructed three-dimensional particle system provided in an embodiment of the present invention. Detailed Implementation
[0048] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.
[0049] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments of the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. Furthermore, it should be understood that 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.
[0050] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.
[0051] Example 1
[0052] In existing technologies, using two-dimensional cross-sectional images to reconstruct three-dimensional particle systems has several drawbacks. First, due to the unavoidable presence of air bubbles in the liquid and on the particle surface, as well as noise caused by camera hardware limitations and indoor lighting conditions during image recording, the two-dimensional cross-sectional images inevitably contain noise. Second, in order for the laser to penetrate the entire specimen, a laser sheet is inevitably used, causing the cross-section of the particle to appear as a bright ring. Traditional detection methods cannot accurately obtain the dimensional information of the cross-section, thus affecting the three-dimensional reconstruction.
[0053] Combination Figures 1-8This invention provides a method for reconstructing a three-dimensional particle system, comprising the following steps:
[0054] S1. Obtain two-dimensional tomographic images of the three-dimensional particle system and perform noise reduction processing on the two-dimensional tomographic images. The two-dimensional tomographic images are cross-sectional images of the three-dimensional particle system on different two-dimensional faults. The most important information contained in them is how many particles are on the fault, as well as the cross-sectional shape and position coordinates of these particles on the fault.
[0055] In this embodiment, the selected particles are round particles, and the cross-sectional shape of the particles is round regardless of the angle from which the cross-section is taken.
[0056] like Figure 2 As shown in (a), due to the unavoidable presence of air bubbles in the liquid and on the surface of the particles, as well as noise in the two-dimensional tomographic images acquired by the camera due to camera hardware limitations and indoor lighting conditions when recording images, noise is inevitable.
[0057] Therefore, in this embodiment, the two-dimensional tomographic image is denoised. Specific steps include:
[0058] S101. Construct a discrete filter, convolve the two-dimensional tomographic image with the discrete filter, and obtain the two-dimensional tomographic image after discrete filtering.
[0059] Specifically, firstly, in order to construct the filter, a circle of diameter D (centered at (0,0)) is represented as a set of zero-order functions:
[0060] φ(x,y)=x 2 +y 2 +(D / 2) 2 (1)
[0061] The function is sampled on the grid (xm, yn) to obtain the array φ. m,n =φ(x m ,y n ).
[0062] Then, Smereka's discrete trigonometric function method is used to construct the discrete filter:
[0063] F(m,n)=δ (+x) (m,n)+δ (-x) (m,n)+δ (+y) (m,n)+δ (-y) (m,n) (2)
[0064] in:
[0065]
[0066]
[0067] In the above formula:
[0068]
[0069] Here, ε is a decimal introduced to avoid division by zero.
[0070] The two-dimensional slope image J(i,j) is then convolved with the discrete filter F(m,n) to obtain the transformed image:
[0071]
[0072] S2. The denoised two-dimensional tomographic image is scanned using the Hough transform circle detection method to determine the center coordinates of the particle cross-section circles in the two-dimensional tomographic image.
[0073] For a two-dimensional tomographic image, the most important thing is to clearly identify how many circles are in the two-dimensional tomographic image, and to know the coordinates of the center and radius of these circles.
[0074] Therefore, in this embodiment, the Hough transform circle detection method is selected to scan the denoised two-dimensional tomographic image. The principle of the Hough transform circle detection method is as follows:
[0075] (xa) 2 +(yb) 2 =r 2 (8)
[0076] Construct an ABR three-dimensional coordinate system. In the ABR coordinate system, given a point (a0, b0, r0), the equation of a circle (x-a0) in a Cartesian coordinate system can be uniquely determined. 2 +(y-b0) 2 =r0 2 All circles passing through a point (x0, y0) in a Cartesian coordinate system are represented as a three-dimensional curve (x0-a) in the abr three-dimensional coordinate system. 2 +(y0-b) 2 =r 2 If any two such curves intersect in the abr three-dimensional coordinate system, it is considered that the points (x1, y1) and (x2, y2) represented by these two curves are all located at the intersection point (a). i ,b i ,r i On the circle defined by ).
[0077] In this step, for a two-dimensional tomographic image after thresholding, any valid pixel with a grayscale value of 1 can be represented by a three-dimensional curve in the ABR three-dimensional coordinate system. Therefore, this point should have many intersecting curves in the ABR three-dimensional coordinate system, and the number of curves is the number of valid pixels. By judging the cumulative number of intersections in the ABR three-dimensional coordinate system, intersections that exceed a certain threshold are considered valid circles.
[0078] At the same time, the range of r also needs to be determined in advance. When the radius of the particle circle on the 2D tomographic image is less than 30 pixels, the particle will be very blurry and indistinguishable on the tomographic image. Also, the radius of the particle circle on the tomographic image cannot exceed 420 pixels, because this is the pixel value of the maximum particle radius on the 2D tomographic image.
[0079] The biggest advantage of the Hough transform circle detection method is its insensitivity to noise, because the number of pixels on the circle will always be greater than the noise. Therefore, the Hough transform circle detection method can accurately find the center of the circle of each particle cross section on a single two-dimensional tomographic image.
[0080] S3. Using the center coordinates of the circle as a reference, obtain the median radius of the particle cross-section circle in the two-dimensional fault image by scanning the radius circle by circle.
[0081] When capturing two-dimensional cross-sectional images, a laser is used to irradiate the specimen, creating a laser surface on the specimen. This laser surface has thickness in the z-direction. Ideally, the thickness of the laser surface should be only one pixel, but this is impossible to achieve because the smaller the thickness of the laser surface, the weaker the laser intensity, which would prevent the laser from penetrating the entire specimen. Therefore, an appropriate thickness of the laser surface must be maintained. However, this results in the cross-section of the particles appearing as a [missing information - likely a pixel or similar element] in the acquired two-dimensional tomographic image. Figure 3 (b) shows a bright ring. Applying the Hough transform circle detection method to such a two-dimensional tomographic image yields a circle radius that is randomly distributed between the outer and inner diameters of the ring. This distribution pattern depends entirely on which part of the ring is "brighter," meaning it is unpredictable and unstatistically predictable. Therefore, the radius values of the particle cross-section circles in the two-dimensional tomographic image obtained by the Hough transform circle detection method are inaccurate.
[0082] Therefore, as Figure 4 As shown, the idea behind this step is to reduce the thickness l of the annulus to the size of the innermost ring. Figure 4 In order to make it easier for readers to observe, the thickness of the laser-etched surface was exaggerated during the labeling process. Figure 4The distance between BO and AO. In actual experiments, the same particle could be clearly observed in nearly one hundred tomographic images, meaning the distance between BO and AO is actually only one-hundredth of the particle's diameter, or about 0.3 millimeters. Therefore, arc AEB is actually a very short segment, allowing the entire laser surface to be approximately reduced to the very center pixel, i.e., the straight line ED. Since D is the midpoint of BO, and because arc AEB approaches infinitesimal, it can be approximated as a straight line. Therefore, point E can be approximated as the midpoint of AB, and point C can also be approximated as the midpoint of AO. In the image recorded by the camera, the thickness of the ring is the length of AO, and the position of point C is located on the midline of the ring.
[0083] Therefore, the method for finding the precise radius can be simplified to finding the radius of the midline of the annulus. The radius of the midline is simply the midpoint between the outer and inner diameters of the annulus. The specific steps include:
[0084] S301. Based on the two-dimensional tomographic image after noise reduction, using the center coordinates of the circle as a reference, and according to the gray value of the pixel, traverse all pixels within a preset radius to determine the inner and outer diameters of the particle cross-section circle.
[0085] S302. Obtain the midline radius of the particle cross-section circle based on its inner and outer diameters.
[0086] Taking scanning acrylic glass particles as an example, this step will be explained in detail, and the specific process is as follows:
[0087] Using the previous Hough transform circle detection method, the center coordinates of the ring on the 2D tomographic image can be obtained relatively accurately. Then, on the thresholded 2D tomographic image, using this pixel as a reference, all points within a 50-pixel radius are traversed. If the number of points with a grayscale value of 1 is greater than 35% of the total number of points within a 50-pixel radius, the ring is considered "bright" on the 50-pixel radius circle; conversely, if the number of points with a grayscale value of 1 is less than 35% of the total number of points within a 50-pixel radius circle, the ring is considered "dark". Next, using the center as a reference, all points within a 51-pixel radius are traversed. Again, based on the above criteria, it is determined whether the ring is "bright" or "dark" on the 51-pixel radius circle. Then, all points within distances of 52, 53... up to 450 pixels from the center are traversed... Figure 5 As shown in (a), the final result is the radius values of all the rings marked "bright" with that point as the center. The minimum value among these values is the inner diameter of the ring, and the maximum value is naturally the outer diameter of the ring, as shown in (a). Figure 5As shown in (b), the average of the two values is the radius of the median, which is the image of the particle on this two-dimensional tomographic image when we reduce the thickness l of the annulus to 1 pixel.
[0088] In this step, taking the scanning of plexiglass particles as an example, the scanning range is set to 50 to 450 pixels. The minimum value of 50 pixels is chosen because observation shows that when the radius of a ring on a 2D tomographic image is less than 50 pixels, the ring appears blurry. Therefore, these blurry rings are actively removed during this scan. Although this reduces some experimental data, it greatly facilitates the subsequent reconstruction work, and the missing parts can be easily filled in during the reconstruction. The maximum scanning range of 450 pixels is chosen because the plexiglass particles selected in this experiment are all 30 mm in diameter. Under the image recording environment of this experiment, the maximum particle radius will not exceed 450 pixels. A value greater than this may scan other particles on the 2D tomographic image, thus causing errors in the results. Finally, the judgment threshold is chosen to be 35% because if the threshold is too large, it is difficult to meet the conditions, thus failing to find the correct ring. If the threshold is too small, it is easy to meet the conditions, resulting in a lot of noise, which affects the judgment. After multiple experiments, it was found that the best final result was achieved when the threshold was set at 35%.
[0089] S4. Based on the center coordinates of the particle cross-section circle and the midline radius, obtain the benchmark two-dimensional fault image.
[0090] Specifically, using the center coordinates of the circles identified by the Hough transform circle detection method and the radii identified by the circle-by-circle radius scanning method, the cross-sections of each circle in the original image are drawn to obtain the reference two-dimensional tomographic image, such as... Figure 7 As shown in (d).
[0091] All images used for reconstruction were redrawn to obtain all the baseline two-dimensional tomographic images used for reconstruction. In addition, in this embodiment, the images were pre-thresholded and binarized during the redrawing process.
[0092] S5. Based on the baseline two-dimensional tomographic image, obtain a three-dimensional view of the three-dimensional particle system.
[0093] Specifically, the 3D particle system is reconstructed using AVIZO software. Interactive thresholding of the particles within AVIZO yields a 3D view of the entire particle system, such as... Figure 8 As shown.
[0094] Furthermore, it also includes: constructing a three-dimensional coordinate system based on the three-dimensional view of the three-dimensional particle system, and obtaining the center coordinates and radius of each particle in the three-dimensional particle system for use in subsequent analysis processes.
[0095] Example 2
[0096] This embodiment discloses a three-dimensional particle system reconstruction system, including:
[0097] The two-dimensional tomographic image acquisition module is configured to: acquire two-dimensional tomographic images of a three-dimensional particle system and perform noise reduction processing on the two-dimensional tomographic images; wherein, the two-dimensional tomographic images contain particle cross-sectional circular information of the three-dimensional particle system;
[0098] The particle cross-section circle information acquisition module is configured to: scan the denoised two-dimensional tomographic image using the Hough transform circle detection method to determine the center coordinates of the particle cross-section circle in the two-dimensional tomographic image; use the center coordinates as a reference to obtain the median radius of the particle cross-section circle in the two-dimensional tomographic image using the circle-by-circle radius scanning method; and obtain a reference two-dimensional tomographic image based on the center coordinates and median radius of the particle cross-section circle.
[0099] The 3D visualization reconstruction module is configured to: obtain a 3D view of a 3D particle system based on a baseline 2D tomographic image.
[0100] It should be noted that the two-dimensional tomographic image acquisition module, the particle cross-sectional circle information acquisition module, and the three-dimensional visualization reconstruction module mentioned above correspond to the steps in Embodiment 1. The examples and application scenarios implemented by these modules and their corresponding steps are the same, but they are not limited to the content disclosed in Embodiment 1. It should be noted that these modules, as part of the system, can be executed in a computer system such as a set of computer-executable instructions.
[0101] Example 3
[0102] Embodiment 3 of the present invention provides an electronic device, including a memory and a processor, as well as computer instructions stored in the memory and running on the processor. When the computer instructions are executed by the processor, they complete the steps of the above-described three-dimensional particle system reconstruction method.
[0103] Example 4
[0104] Embodiment 4 of the present invention provides a computer-readable storage medium for storing computer instructions, which, when executed by a processor, complete the steps of the above-described three-dimensional particle system reconstruction method.
[0105] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0106] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0107] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment, whereby a series of operational steps are performed to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0108] The descriptions of each embodiment in the above embodiments have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.
[0109] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. A method for reconstructing a three-dimensional particle system, characterized in that, include: Two-dimensional tomographic images of a three-dimensional particle system are acquired, and the two-dimensional tomographic images are denoised; wherein, the two-dimensional tomographic images contain information about the circular cross-sections of the particles in the three-dimensional particle system; The denoised two-dimensional tomographic image is scanned using the Hough transform circle detection method to determine the center coordinates of the particle cross-section circles in the two-dimensional tomographic image; using the center coordinates as a reference, the median radius of the particle cross-section circles in the two-dimensional tomographic image is obtained by the circle-by-circle radius scanning method; and a reference two-dimensional tomographic image is obtained based on the center coordinates and median radius of the particle cross-section circles. The method of obtaining the median radius of the particle cross-section circle in a two-dimensional tomographic image by using the circle center coordinates as a reference and scanning the radius circle by circle includes: Based on the noise-reduced two-dimensional tomographic image, using the center coordinates of the circle as a reference, and according to the gray value of the pixel, all pixels within a preset radius are traversed to determine the inner and outer diameters of the particle cross-section circle. Based on the inner and outer diameters of the particle cross-section circle, obtain the radius of the midline of the particle cross-section circle; Based on the baseline two-dimensional tomographic image, a three-dimensional view of the three-dimensional particle system is obtained.
2. The three-dimensional particle system reconstruction method as described in claim 1, characterized in that, The denoising process for the two-dimensional tomographic image includes: Construct a discrete filter, and convolve the two-dimensional tomographic image with the discrete filter to obtain the discretely filtered two-dimensional tomographic image. Threshold segmentation is performed on the discretely filtered two-dimensional tomographic image to obtain a denoised two-dimensional tomographic image.
3. The method for reconstructing a three-dimensional particle system as described in claim 2, characterized in that, The construction of the discrete filter includes: Construct a zero-order function and sample it to obtain the circle center coordinate array; Discrete filters are constructed using the discrete trigonometric function method based on the circle center coordinate array.
4. The method for reconstructing a three-dimensional particle system as described in claim 1, characterized in that, The specific steps for obtaining the reference two-dimensional tomographic image based on the center coordinates and median radius of the particle cross-section circle are as follows: The radius of the midline of the particle cross-section circle is used as the radius of the particle cross-section circle to determine the size information of the particle cross-section circle; Based on the center coordinates and dimensions of the particle cross-section circles, the position information of the particle cross-section circles is determined; based on the position information of all particle cross-section circles in the two-dimensional tomographic image, a reference two-dimensional tomographic image is obtained.
5. The method for reconstructing a three-dimensional particle system as described in claim 1, characterized in that, The process of obtaining a three-dimensional view of a three-dimensional particle system based on a reference two-dimensional tomographic image specifically involves: using three-dimensional visualization reconstruction software to perform interactive threshold segmentation on the reference two-dimensional tomographic image to obtain a three-dimensional view of the three-dimensional particle system.
6. The method for reconstructing a three-dimensional particle system as described in claim 1, characterized in that, Also includes: Based on the three-dimensional view of the three-dimensional particle system, a three-dimensional coordinate system is constructed to obtain the center coordinates and radius of each particle in the three-dimensional particle system.
7. A three-dimensional particle system reconstruction system, characterized in that, include: The two-dimensional tomographic image acquisition module is configured to: acquire two-dimensional tomographic images of a three-dimensional particle system and perform noise reduction processing on the two-dimensional tomographic images; wherein, the two-dimensional tomographic images contain particle cross-sectional circular information of the three-dimensional particle system; The particle cross-section circle information acquisition module is configured to: scan the denoised two-dimensional tomographic image using the Hough transform circle detection method to determine the center coordinates of the particle cross-section circle in the two-dimensional tomographic image; use the center coordinates as a reference to obtain the median radius of the particle cross-section circle in the two-dimensional tomographic image using the circle-by-circle radius scanning method; and obtain a reference two-dimensional tomographic image based on the center coordinates and median radius of the particle cross-section circle. The method of obtaining the median radius of the particle cross-section circle in a two-dimensional tomographic image by using the circle center coordinates as a reference and scanning the radius circle by circle includes: Based on the noise-reduced two-dimensional tomographic image, using the center coordinates of the circle as a reference, and according to the gray value of the pixel, all pixels within a preset radius are traversed to determine the inner and outer diameters of the particle cross-section circle. Based on the inner and outer diameters of the particle cross-section circle, obtain the radius of the midline of the particle cross-section circle; The 3D visualization reconstruction module is configured to: obtain a 3D view of a 3D particle system based on a baseline 2D tomographic image.
8. An electronic device, characterized in that, It includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, which, when executed by the processor, perform the steps described in any one of claims 1-6.
9. A computer-readable storage medium, characterized in that, Used to store computer instructions, which, when executed by a processor, perform the steps described in any one of claims 1-6.