Method and system for extracting complex connected region of three-dimensional model, medium and equipment

By correcting the Allen-Cahn equation and iterative method, the problems of low efficiency and insufficient accuracy of the connection area extraction of complex three-dimensional models are solved, and efficient and accurate connection area extraction is achieved, which is suitable for the detection and analysis of complex industrial parts.

CN119068012APending Publication Date: 2024-12-03XI AN JIAOTONG UNIV
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411079430.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-07
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

When the prior art deals with three-dimensional models of complex geometric shapes and multi-branch structures, the connection area extraction efficiency is low and the accuracy is insufficient, making it difficult to meet the complex detection and analysis needs of modern industrial parts.

Method used

Based on phase field theory, the Allen-Cahn equation was corrected, and the Lagrange multiplier method and semi-implicit Crank-Nicolson format were used for discretization, and the phase field variables were solved using the Picard iteration method to extract the complex connected areas of the target model.

Benefits of technology

It realizes efficient and accurate connection area extraction of complex three-dimensional models, improves detection and analysis efficiency, reduces errors and rework in the production process, and improves product quality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119068012A_ABST
    Figure CN119068012A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of complex three-dimensional model analysis, in particular to a method, a system, a medium and equipment for extracting a complex connected region of a three-dimensional model, and the method comprises the following steps: representing a target three-dimensional model by a discrete function psi (x), and representing the target three-dimensional model by phi (x, y, z; t) represents a phase field variable, an Allen-Cahn equation is corrected based on a phase field theory, and a correction continuous system capable of realizing extraction of a connected region of a target model is established; step 2, a Lagrange multiplier method is adopted, a semi-implicit Crank-Nicolson format is applied to carry out discretization processing on the corrected continuous system, and a discrete system is obtained; 3, solving a phase field variable on a computational domain for the discrete system; in the solving process, the evolution process of the phase field variable is the extraction process of the connected region. The connected regions of the target model are distinguished by using the diffusion effect of the Allen-Cahn equation and the surface function of the three-dimensional model, so that the accuracy of extracting the connected regions of the target model is ensured, and the stability of the model is ensured by using the Lagrange multiplier method.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of complex three-dimensional model analysis, and in particular to a method, system, medium and equipment for extracting complex connected areas of a three-dimensional model. Background Art

[0002] Whether it is internal flow channel analysis in mechanical design, organ segmentation in biomedical imaging, or defect identification in nondestructive testing, the ability to accurately extract connected spatial areas inside or outside the target structure is a key link in solving related problems. Specifically, in medical imaging technologies such as CT and MRI, segmenting and reconstructing the precise geometric connectivity of organs or lesions is the basis for subsequent analysis. In the field of industrial nondestructive testing, using CT scan data to identify connected defect areas such as cracks and holes inside the inspected part is also an essential prerequisite.

[0003] In order to better address the above challenges, some commercial CAD software provides functional modules for extracting connected areas. These software usually adopt a mesh- or voxel-based algorithm framework to spatially divide the target solid model and identify internal cavities by searching for connected meshes or voxels that are not occupied by the solid model. However, these methods can only handle flow channel areas with simple topology well. When dealing with industrial parts with extremely high geometric complexity, such traditional algorithms often fail due to the exponential growth of meshing accuracy and data size. Zhu et al. used a semantic segmentation method based on deep learning to effectively handle the problem of extracting connected areas of extraocular muscles and optic nerves from orbital CT images and achieved high-accuracy volume measurement, but it may be affected by subjective factors and has insufficient processing of curves and other anatomical structures. Lee et al. used the level set method to solve the problem of liver segmentation. Through two-step seed region growing and 2.5D shape propagation, the connected areas of the liver can be accurately and quickly extracted. However, there are limitations caused by liver morphological changes and similar organs, as well as insufficient processing of abnormalities and small organs. Yang et al. used a method based on topological graph analysis to extract the skeleton of connected regions from binary images, and can well maintain the topological relationship between objects in the image. However, the limitation of the current algorithm is that it does not handle curve skeleton extraction. Bowden et al. used active contours and Poisson's equation to deal with the problem of image connected regions, improving the accuracy of boundary detection and noise suppression. However, this method requires users to select appropriate parameter values, and the computational complexity is high, especially when dealing with large-scale data sets. There are certain limitations. Therefore, these methods still have certain difficulties in dealing with complex geometries with high curvature and multiple branches. Since modern industrial parts often have complex geometric shapes and multi-branch structures, traditional algorithms are inefficient and inaccurate when dealing with these complex structures. Therefore, developing an efficient and reliable complex connected region discrimination algorithm can significantly improve the detection and analysis efficiency of complex parts, reduce errors and rework in the production process, and improve overall production efficiency and product quality. Summary of the invention

[0004] In view of the problem in the prior art that the processing efficiency of connected regions of complex three-dimensional models is low and the accuracy is insufficient, the present invention provides a method, system, medium and device for extracting complex connected regions of three-dimensional models.

[0005] The present invention is achieved through the following technical solutions:

[0006] A method for extracting complex connected regions of a three-dimensional model comprises the following steps:

[0007] Step 1: The target three-dimensional model is represented by a discrete function ψ(x), the phase field variable is represented by φ(x, y, z; t), and the Allen-Cahn equation is modified based on the phase field theory to establish a modified continuous system that can realize the extraction of the connected area of ​​the target model;

[0008] Step 2, using the Lagrange multiplier method and applying the semi-implicit Crank-Nicolson format to discretize the modified continuous system in step 1 to obtain a discrete system;

[0009] Step 3, solving the phase field variables for the discrete system in the computational domain; during the solution process, the evolution of the phase field variables is the process of extracting the connected region.

[0010] Preferably, in step 1, the modified continuous system established based on phase field theory is as follows:

[0011]

[0012] Among them, ψ(x) is the surface discrete function of the target three-dimensional model, Refers to the area within the surface of a three-dimensional model;

[0013] M represents the diffusion rate;

[0014] φ is the phase field function;

[0015] ∈ is a parameter representing the interface transition layer;

[0016] F(φ)=φ 2 (φ-1) 2 / 4 is a double-well potential function.

[0017] Preferably, in step 2, first establish an equivalent system of the modified continuous system in step 1, then introduce the discrete operation operator and the second-order central difference method, and use a uniform grid to calculate the domain Discretize.

[0018] Preferably, in step 2, the expression of the equivalent system is as follows:

[0019]

[0020]

[0021] The initial condition of η(t) is set to η(0)=1. This equivalent system is equivalent to the modified continuous system in step 1.

[0022] Preferably, in step 3, the Picard iteration method is used to solve the phase field variables in the computational domain.

[0023] A complex connected region processing system for a three-dimensional model comprises a model correction module, a discretization processing module and a model processing module. The model correction module is used to correct the Allen-Cahn equation based on phase field theory to establish a continuous system that can realize the extraction of the connected region of the target model; the discretization processing module is used to discretize the continuous system by using the Lagrange multiplier method and the semi-implicit Crank-Nicolson format; and the model processing module is used to extract the complex connected region of the target three-dimensional model by using the Picard iteration method.

[0024] A computer-readable storage medium includes a computer program, which, when executed by a processor, implements the steps of the method for extracting complex connected areas of a three-dimensional model.

[0025] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the method for extracting complex connected areas of a three-dimensional model are implemented.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] The invention discloses a method for extracting complex connected regions of a three-dimensional model. The method applies the diffusion effect of the Allen-Cahn equation and the surface function of the three-dimensional model to distinguish the connected regions of the target model, thereby ensuring the accuracy when extracting the connected regions of the target model. The method also applies the Lagrange multiplier method to ensure the stability of the model. The method has high accuracy and reliability and can be widely used in the fields of three-dimensional image processing, computer vision, etc.

[0028] Furthermore, the Lagrange multiplier method and the semi-implicit Crank-Nicolson format are used to discretize the modified continuous system in order to transform the continuous mathematical model into a discrete system that can be solved on a computer.

[0029] Furthermore, the Picard iteration method is used to solve the phase field variables. As the iteration process proceeds, the values ​​of the phase field variables will gradually stabilize and show the characteristics of connected regions. By observing the distribution of the phase field variables, the complex connected regions in the three-dimensional model can be extracted. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a flow chart of a method for extracting complex connected regions of a three-dimensional model according to the present invention;

[0031] Figure 2 This is a schematic diagram of extracting a connection path of a maze model in Example 1 of the present invention;

[0032] Figure 3 This is a schematic diagram of extraction of a Stanford rabbit model in Example 2 of the present invention:

[0033] Figure 4 This is a schematic diagram of extracting an invisible connected area inside a certain model in Example 3 of the present invention;

[0034] Figure 5 It is a schematic diagram of modules of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION

[0035] The present invention is further described in detail below in conjunction with specific embodiments, which are intended to explain the present invention rather than to limit it.

[0036] The invention discloses a method for extracting complex connected regions of a three-dimensional model, which aims to realize the discrimination of the connected regions of a complex model by mathematical and physical methods. The classical Allen-Cahn (AC) equation is modified based on phase field theory to adapt it to the requirements of complex connected region extraction. The modified AC equation uses the diffusion effect and the surface function of the three-dimensional model to distinguish the connected regions of the target model, so that the complex connected regions of the target part can be accurately filled and extracted. Table 1 shows the meanings of the main symbols.

[0037] The present invention discloses a method for extracting complex connected regions of a three-dimensional model. Figure 1 , including the following steps:

[0038] Step 1: The target three-dimensional model is represented by a discrete function ψ(x), the phase field variable is represented by φ(x, y, z; t), and the Allen-Cahn equation is modified based on the phase field theory to establish a modified continuous system that can extract the connected area of ​​the target model.

[0039] Among them, φ(x, y, z; t) represents the phase field variable, and ψ(x) is the surface discrete function of the target three-dimensional model.

[0040] Specifically, in the computational domain Define the point cloud data of any three-dimensional model, that is, x = (x, y, z) ∈ Ω; assume that the phase field function φ (x, t) moves under the influence of the mean curvature flow, and the normal velocity of the hypersurface of its movement is equal to the negative mean curvature. Set an initial value for the phase field on the surface of the three-dimensional model. The phase field function starts to diffuse from its initial state and stops when it encounters resistance on the model surface. The process of the phase field function diffusing from its initial state to stopping is the process of extracting the connected area, and finally the connected area on the surface of the target three-dimensional model is obtained.

[0041] Table 1 Meaning of main symbols

[0042]

[0043] The modified continuous system established based on phase field theory is as follows:

[0044]

[0045] Among them, ψ(x) is the surface discrete function of the target three-dimensional model, Refers to the area within the surface of a three-dimensional model;

[0046] M represents the diffusion rate;

[0047] φ is the phase field function;

[0048] ∈ is a parameter representing the interface transition layer;

[0049] F(φ)=φ 2 (φ-1) 2 / 4 is a double-well potential function.

[0050] A simple example in the two-dimensional domain Ω = [-1, 1] × [-1, 1] is used to illustrate the basic mechanism of the proposed model. Let ψ(x, y) be a discrete function consisting of three-dimensional surface data generated from two tangent circles, whose centers are located at (0.3, 0.5) and (-0.3, 0.5) respectively, and whose radius is 0.3. D represents the area within the boundaries of the two tangent circles. The initial phase field value φ(x, y, 0) is selected as a circle with (0.3, 0.5) as the center and 0.1 as the radius in the two tangent circles. The basic principle of the phase field model can be summarized as follows: Due to the diffusion effect of the Allen-Cahn equation, the area Ω within the tangent circle D When the diffusion of φ reaches the circular boundary, it encounters resistance due to the constraint imposed by ψ(x), and thus stops spreading, eventually filling the entire tangent circle area without overflowing the boundary.

[0051] The Allen-Cahn equation describes a law of geometric evolution, where the normal velocity of the evolving hypersurface is proportional to the negative value of its mean curvature:

[0052] V n =-Mκ.

[0053] Among them, V n represents the normal velocity of the geometry, and κ represents the mean curvature of the surface. When the mean curvature of the volume reaches zero, the flow stops, thus distinguishing the connected areas of the complex model in the computational domain. 2 The AC equation is obtained by using the gradient flow:

[0054]

[0055] By using L 2 The AC equation obtained by calculating the variational derivative of the above energy functional can be expressed as

[0056] Step 2: Use the Lagrange multiplier method and the semi-implicit Crank-Nicolson format to discretize the modified continuous system in step 1 to obtain a discrete system.

[0057] Specifically, the modified continuous system in step 1 is first rewritten as an equivalent system, and then the discrete operation operator and the second-order central difference method are introduced to discretize the computational domain using a uniform grid;

[0058] The expression of the equivalent system is as follows:

[0059]

[0060]

[0061] By setting the initial condition of η(t) to η(0)=1.

[0062] Then introduce the discrete operation operator and the formula of the second-order central difference method, considering the three-dimensional region Use uniform mesh for the computational domain Discretize, the grid size is calculated as h = L x / N x =L y / N y =L z / N z , where N x 、N y and N z are positive even numbers in the x-, y-, and z-directions, respectively. Representing i, j, and k as direction indices, we can define is φ(x i ,y j ,z j ,nΔt), where x i =(i-0.5)h,y j =(j-0.5)h,z j =(k-0.5)h, Δt=T / N, T is the final time, N is the total time step.

[0063] The discrete inner product is defined as:

[0064]

[0065] The following numerical scheme is developed based on the Crank–Nicolson time discretization:

[0066]

[0067]

[0068] in Here

[0069]

[0070] Step 3: For discrete systems, use the Picard iteration method to solve the phase field variables in the computational domain and update and φ n+1 As shown below:

[0071]

[0072]

[0073] in m is the number of Picard iterations. The initial condition is chosen to be η 0 =1,μ n+1,0 =2μ n -μ n-1 When the calculation is performed until the error is less than the given tolerance, that is,

[0074]

[0075] At this time, solving for φ n+1 =φ n+1,m+1 .exist Find an approximate solution near 0 = 1 as the initial condition, the calculation cost is low and simple. During the solution process, the evolution process of the phase field variables is the process of extracting the connected region.

[0076] The complex connected region extraction method of a three-dimensional model proposed in the present invention modifies the classic Allen-Cahn equation, applies the diffusion effect of the AC equation, and distinguishes the connected regions of the target model based on the surface function of the three-dimensional target model, so as to accurately fill and extract the complex connected regions of the target part.

[0077] Example 1

[0078] Reference Figure 2 Taking a maze as an example, the green area represents the maze model, and the yellow area shows the phase field function μ in Ω d The diffusion in the maze model is used to determine the evolution of the connected paths.

[0079] Example 2

[0080] Reference Figure 3 Taking a Stanford rabbit model as an example, the phase field function φ starts from an initial sphere and diffuses outward, showing the evolution process of the three-dimensional Stanford rabbit model.

[0081] Example 3

[0082] Take a model whose internal connected area cannot be seen through as an example, where 4(a) is a schematic diagram of the model surface, 4(b) is an internal perspective view of the model, 4(c) is the surface connected area extracted from the experiment, and 4(d) is a matching diagram between the extracted surface connected area and the internal connection channel of the model: the yellow area represents the internal connected area extracted from the experiment, and the green area represents the model.

[0083] The present invention also discloses a complex connected region processing system for a three-dimensional model, comprising a model correction module, a discretization processing module and a model processing module. The model correction module is used to correct the Allen-Cahn equation based on phase field theory to establish a continuous system that can realize the extraction of the connected region of the target model; the discretization processing module is used to discretize the continuous system using the Lagrange multiplier method and the semi-implicit Crank-Nicolson format; the model processing module is used to extract the complex connected region of the target three-dimensional model using the Picard iteration method.

[0084] The present invention also discloses a computer-readable storage medium, including a computer program, which implements the steps of the method for extracting complex connected areas of a three-dimensional model when the computer program is executed by a processor. The computer program includes computer program code, which can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory and a read-only memory (ROM).

[0085] The present invention also discloses an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of a method for extracting complex connected regions of a three-dimensional model when executing the computer program. The electronic device can be integrated in various computing devices to improve its ability to process three-dimensional models.

[0086] Reference Figure 5 Taking an electronic device as an example, the electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on the at least one processor 102, and at least one communication bus 104.

[0087] The memory 101 can be used to store a computer program 103, and the processor 102 implements the steps of the complex connected region extraction method of the three-dimensional model of embodiments 1, 2, and 3 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101. The memory 101 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data (such as audio data) created according to the use of the electronic device 100, etc. In addition, the memory 101 can include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices.

[0088] At least one processor 102 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 102 may be a microprocessor or any conventional processor, etc. The processor 102 is the control center of the electronic device 100, and uses various interfaces and lines to connect various parts of the entire electronic device 100.

[0089] The memory 101 in the electronic device 100 stores a plurality of instructions to implement a method for extracting complex connected regions of a three-dimensional model, and the processor 102 can execute the plurality of instructions to implement the method.

[0090] If the module / unit integrated in the terminal device is implemented in the form of 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 present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc.

[0091] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0092] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0093] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0094] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0095] The above description is only a preferred embodiment of the present invention and is not intended to impose any limitation on the technical solution of the present invention. Those skilled in the art should understand that, without departing from the spirit and principles of the present invention, the technical solution can also be subjected to several simple modifications and substitutions, and these modifications and substitutions are also within the scope of protection covered by the claims.

Claims

1. A method for extracting complex connected regions of a three-dimensional model, characterized in that: The following steps are involved: Step 1: The target three-dimensional model is represented by a discrete function ψ(x), the phase field variable is represented by φ(x, y, z; t), and the Allen-Cahn equation is modified based on the phase field theory to establish a modified continuous system that can realize the extraction of the connected area of ​​the target model; Step 2, using the Lagrange multiplier method and applying the semi-implicit Crank-Nicolson format to discretize the modified continuous system in step 1 to obtain a discrete system; Step 3, solving the phase field variables for the discrete system in the computational domain; during the solution process, the evolution of the phase field variables is the process of extracting the connected region.

2. The method for extracting complex connected regions of a three-dimensional model according to claim 1, characterized in that: In step 1, the modified continuous system based on phase field theory is as follows: Among them, ψ(x) is the surface discrete function of the target three-dimensional model, Ω D Refers to the area within the surface of a three-dimensional model; M represents the diffusion rate; φ is the phase field function; ∈ is a parameter representing the interface transition layer; F(φ)=φ 2 (φ-1) 2 / 4 is a double-well potential function.

3. The method for extracting complex connected regions of a three-dimensional model according to claim 1, characterized in that: In step 2, first establish the equivalent system of the modified continuous system in step 1, then introduce the discrete operation operator and the second-order central difference method, and use a uniform grid to calculate the domain Discretize.

4. The method for extracting complex connected regions of a three-dimensional model according to claim 3, characterized in that: In step 2, the expression of the equivalent system is as follows: The initial condition of η(t) is set to η(0)=1. This equivalent system is equivalent to the modified continuous system in step 1.

5. The method for extracting complex connected regions of a three-dimensional model according to claim 1, characterized in that: In step 3, the Picard iteration method is used to solve the phase field variables in the computational domain.

6. A complex connected region processing system for a three-dimensional model, characterized in that: It includes a model correction module, a discretization processing module and a model processing module. The model correction module is used to correct the Allen-Cahn equation based on the phase field theory and establish a continuous system that can realize the extraction of the connected area of ​​the target model; The discretization processing module is used to discretize the continuous system using the Lagrange multiplier method and the semi-implicit Crank-Nicolson format, and the model processing module is used to extract the complex connected areas of the target three-dimensional model using the Picard iteration method.

7. A computer-readable storage medium, characterized in that: The invention comprises a computer program, which, when executed by a processor, implements the steps of the method for extracting complex connected areas of a three-dimensional model as claimed in any one of claims 1 to 5.

8. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the method for extracting complex connected areas of a three-dimensional model as described in any one of claims 1 to 5 are implemented.

Citation Information

Cited By

  • Brain tumor diagnosis system based on nuclear magnetic resonance image and mixed numerical segmentation algorithm

    CN121639667A