Method for generating digital human body geometric model by tissue boundary segmentation, simplification and fusion

Abstract

The present invention relates to the technical field of digital human models and discloses a method for generating digital human body geometry models using tissue boundary segmentation, simplification, and fusion. First, the point cloud coordinates of each human tissue are obtained using a digital human body geometry model. Then, based on the point cloud coordinates of each tissue, the implicit field function of each tissue is determined. Based on the implicit field function, a fusion algorithm is used to segment, simplify, and fuse the tissue boundaries to obtain a surface model of the tissue. The fusion algorithm accurately identifies the tissue structure step by step, overcoming the challenge of shape recognition for complex morphological tissues and balancing morphological fidelity with model practicality.

Description

Technical Field

[0001] The present invention relates to the technical field of digital human body models, and in particular to a method for segmenting, simplifying and fusing tissue boundaries of digital human body geometric models. Background Art

[0002] Digital human finite element models (HBMs), an emerging area of finite element simulation, are attracting significant attention. HBMs are finite element simulation models constructed based on digital human body geometry. Existing HBM geometry is typically generated from CT images of human volunteers. Image data can intuitively reflect the composition of various human tissues, but in the field of finite element modeling, appropriate simplification of the body's tissue structure is necessary during the geometric model construction phase to ensure simulation stability and improve finite element model quality.

[0003] The surfaces of some tissues in the human body have complex shapes such as a large number of wrinkles and grooves, which poses a challenge to the shape recognition of human tissues. For example, the surface of the cerebral cortex is densely covered with deep grooves and winding folds, the intestinal mucosa is covered with tens of thousands of circular folds and fine protrusions, and the surface of articular cartilage presents a wavy micro-texture. These structures are essentially the concrete representation of physiological functions in anatomical morphology. First, completely retaining anatomical details will lead to an exponential increase in the number of model meshes, which not only greatly increases the demand for computing resources, but may also cause instability in the numerical solution due to unit distortion. However, if the surface features are oversimplified, the authenticity of biomechanical behavior may be destroyed.

[0004] Therefore, how to accurately identify human tissue and balance morphological fidelity and model practicality has become an urgent problem to be solved. Summary of the Invention

[0005] In order to solve the above technical problems, the present invention provides a method for segmenting, simplifying and fusing tissue boundaries of a digital human body geometric model, comprising the following steps:

[0006] Obtain the point cloud coordinates of each tissue of the human body through the digital human body geometric model;

[0007] Determining the implicit field function of each human tissue according to the point cloud coordinates of each human tissue; the implicit field function is used to characterize the surface relationship between the tissue and the digital human body geometric model;

[0008] Based on the implicit field function, the human tissue is segmented, simplified and fused by a fusion algorithm to obtain a surface model of the human tissue.

[0009] Furthermore, based on the implicit field function, the human tissue is segmented, simplified, and fused by a fusion algorithm, including:

[0010] Based on the implicit field function, according to the shape parameters and the offset parameters, the digital human body geometric model tissue boundary is segmented and simplified through a fusion algorithm to obtain an initial human body tissue surface model;

[0011] Based on the implicit field function and according to the fusion coefficient, the tissue boundary of the digital human body geometric model is fused through a fusion algorithm to obtain a human body tissue surface model.

[0012] Furthermore, based on the implicit field function, according to the shape parameters and the offset parameters, the digital human body geometric model tissue boundary is segmented and simplified through a fusion algorithm to obtain an initial human body tissue surface model, including:

[0013] Based on the implicit field function and according to the preset first shape parameter, the digital human body geometric model tissue boundary is segmented and simplified by a fusion algorithm to obtain a first human body tissue surface model;

[0014] Based on the implicit field function, according to the preset second shape parameter and offset parameter, the first human tissue surface model is segmented and simplified through a fusion algorithm to obtain an initial human tissue surface model.

[0015] Furthermore, when the first human tissue surface model is obtained, the expression of the fusion algorithm is:

[0016] ;

[0017] Among them, C 1,α represents the first human tissue surface model, α is the shape parameter, corresponding to the first shape parameter in this step, B(p i , α ) represents the point p i As the center, α is A spherical area of radius, Represents the point cloud coordinate set of human tissue, p i represents the i-th point cloud point in human tissue, It means to unite the spherical area cloud points formed by various points in human tissue.

[0018] Furthermore, when the initial human tissue surface model is obtained, the fusion algorithm expression is:

[0019] ;

[0020] Among them, C 2，α，d represents the initial human tissue surface model, α is the shape parameter, corresponding to the second shape parameter in this step, d is the offset parameter, B(p i ,α+d ) represents the point p i As the center, α is Radius, d is the spherical area of the offset, Represents the point cloud coordinate set of human tissue, p i represents the i-th point cloud point in human tissue, It means to unite the spherical area cloud points formed by various points in human tissue.

[0021] Furthermore, the second shape parameter is smaller than the first shape parameter.

[0022] Furthermore, when the human tissue surface model is obtained, the expression of the fusion algorithm is:

[0023] ;

[0024] Among them, C 3，α，d represents the surface model of human tissue, α is the shape parameter, corresponding to the second shape parameter in this step, d is the offset parameter, q is the fusion coefficient, B(p i , α+d ) represents the point p i As the center, α is Radius, d is the spherical area of the offset, Represents the point cloud coordinate set of human tissue, p i represents the i-th point cloud point in human tissue, It means to unite the spherical area cloud points formed by various points in human tissue.

[0025] Furthermore, the expression of the implicit field function is:

[0026] ;

[0027] Among them, SDF( x ) represents any node in the coordinate system x The implicit field function at ; α represents the shape parameter, d represents the offset parameter, surface Represents the surface composed of point cloud coordinates, distance ( x,surface ) represents a node x arrive surface The shortest distance on the surface.

[0028] Furthermore, based on the implicit field function, the human tissue is segmented, simplified, and fused by a fusion algorithm to obtain a surface model of the human tissue, and the shape of the surface model of the human tissue is optimized. The optimization expression is:

[0029] ;

[0030] Among them, E is the energy function, s is the arc length of the surface, is the boundary of the surface model of the human tissue to be optimized, k(s) is the boundary curvature, and ds represents the arc length integral of the boundary of the surface model.

[0031] The embodiments of the present invention have the following technical effects:

[0032] The present invention first uses a digital human body geometry model to obtain the point cloud coordinates of each tissue. Then, based on these point cloud coordinates, the implicit field functions of each tissue are determined. Finally, using this implicit field function, a fusion algorithm is used to segment, simplify, and fuse the tissue boundaries to obtain a surface model of the tissue. This fusion algorithm allows for precise, step-by-step identification of tissue structure, overcoming the challenges of shape recognition for complex morphologies and achieving a balance between morphological fidelity and model practicality. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0034] Figure 1 This is a flowchart of the steps of the method for segmenting, simplifying and fusing the tissue boundary of a digital human body geometric model provided by an embodiment of the present invention;

[0035] Figure 2 The embodiment of the present invention provides a shape parameter α Schematic diagram for controlling the shape of the package;

[0036] Figure 3 The embodiment of the present invention provides an offset parameter d Schematic diagram of controlling expansion and contraction;

[0037] Figure 4 This is a diagram showing a geometric model of the small intestine provided by an embodiment of the present invention;

[0038] Figure 5This is a surface model display diagram of the small intestine geometric model after segmentation, simplification and fusion processing provided by an embodiment of the present invention;

[0039] Figure 6 This is a comparison diagram of the large intestine before and after segmentation, simplification, and fusion provided by an embodiment of the present invention;

[0040] Figure 7 This is a comparison diagram of the brain before and after segmentation, simplification and fusion provided by an embodiment of the present invention;

[0041] Figure 8 This is a comparison diagram before and after muscle segmentation, simplification, and fusion provided by an embodiment of the present invention;

[0042] Figure 9 It is a flow chart of a method for segmenting, simplifying and fusing tissue boundaries of a digital human body geometric model provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0043] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention are described clearly and completely below. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are also within the scope of protection of the present invention.

[0044] In the field of digital human geometry model generation, simplifying or fusing complex tissues within extracted image data is a relatively novel approach. Compared to manually observing CT image grayscale values and manually demarcating boundaries to simplify tissue structures, using algorithms is more efficient and produces more accurate image processing. The resulting tissue, after division, fusion, and simplification, can maximize the preservation of the original tissue's geometric appearance and eliminate geometric factors that are detrimental to finite element modeling.

[0045] The present invention uses a fusion algorithm to perform operations on the boundaries of tissues and organs, fuses organs and other tissues in the provided HBM geometric model, removes boundaries, and segments the boundaries of different tissues, retaining the tissue morphology of the original tissues to the greatest extent possible. This ensures that each tissue in the model has a smooth appearance and complete structure, facilitates finite element meshing, shortens the HBM development cycle, and improves the HBM model quality and operational stability.

[0046] Figure 1 This is a flowchart of the steps of the method for segmenting, simplifying and fusing the tissue boundary of a digital human body geometric model provided by an embodiment of the present invention. Figure 91 is a flow chart of a method for segmenting, simplifying, and fusing digital human body geometry models provided by an embodiment of the present invention. The method for segmenting, simplifying, and fusing digital human body geometry models provided by the present invention comprises the following steps:

[0047] S1: Obtain the point cloud coordinates of each human tissue through the digital human body geometric model;

[0048] For example, a digital human body geometric model is obtained. Typically, tissue segmentation is performed on CT images followed by inverse reconstruction to obtain a three-dimensional digital model, which is then exported as an STL file to collect point cloud information of each human tissue.

[0049] Digital geometric models of human tissue are composed of triangular facets, each vertex of which can be interpreted as a coordinate point in space, forming a set of spatial coordinate points containing the geometric information of human tissue, namely, point cloud coordinates. The purpose of this invention is to morphologically divide, modify, and fuse organs and tissues formed by these point sets, thereby forming surface models of human tissues and organs with clear geometric features, relative independence and closure, and a relatively smooth surface without unnecessary wrinkles, which are conducive to finite element modeling.

[0050] S2: Determine the implicit field function of each tissue based on the point cloud coordinates of each tissue; the implicit field function is used to characterize the surface relationship between the tissue and the digital human body geometric model;

[0051] The expression of three-dimensional space is divided into explicit and implicit. The explicit form includes visualization methods such as point clouds and triangular patches. The implicit form is to express the three-dimensional space using functions. According to the point cloud coordinates of each piece of tissue, an implicit field function is constructed to characterize the surface relationship of the digital human body geometric model corresponding to the human tissue. This is to initialize the parameter values in the fusion algorithm. Subsequent boundary segmentation, simplification and fusion of human tissues need to be carried out on the basis of complying with the implicit field function.

[0052] In some embodiments, the implicit field function is expressed as:

[0053] ;(Formula 1)

[0054] Among them, SDF( x ) represents any node in the coordinate system x The implicit field function at ; α represents the shape parameter, d represents the offset parameter, surface Represents the surface composed of point cloud coordinates, distance ( x,surface ) represents a node x arrive surfaceShortest distance to the surface: A positive number indicates that the node x is outside the surface of the surface, and the distance is the shortest distance to the surface; a negative number indicates that the node x is inside the surface of the surface, and the absolute value represents the shortest distance to the surface; a zero value indicates that the node x is exactly on the surface.

[0055] S3: Based on the implicit field function, the human tissue is segmented, simplified and fused through the fusion algorithm to obtain the surface model of the human tissue.

[0056] In some embodiments, based on the implicit field function, a fusion algorithm is used to segment, simplify, and fuse the boundaries of human tissue, including:

[0057] S31: Based on the implicit field function, shape parameters and offset parameters, the digital human body geometry model tissue boundary is segmented and simplified through a fusion algorithm to obtain an initial human body tissue surface model;

[0058] In some embodiments, based on the implicit field function, according to the shape parameters and the offset parameters, a fusion algorithm is used to segment and simplify the tissue boundary of the digital human body geometric model to obtain an initial human tissue surface model, including:

[0059] S31a: Based on the implicit field function and according to the preset first shape parameter, the digital human body geometric model tissue boundary is segmented and simplified by a fusion algorithm to obtain a first human body tissue surface model;

[0060] In some embodiments, when the first human tissue surface model is obtained, the expression of the fusion algorithm is:

[0061] ;(Formula 2)

[0062] Among them, C 1,α represents the first human tissue surface model, α is the shape parameter, corresponding to the first shape parameter in this step, B(p i , α ) represents the point p i As the center, α is A spherical area of radius, Represents the point cloud coordinate set of human tissue, p i represents the i-th point cloud point in human tissue, It means to unite the spherical area cloud points formed by various points in the human body tissue, that is, to unite all these spheres to form an area that wraps the surface.

[0063] S31b: Based on the implicit field function, according to the preset second shape parameter and offset parameter, the first human tissue surface model is segmented and simplified through a fusion algorithm to obtain an initial human tissue surface model.

[0064] In some embodiments, when obtaining the initial human tissue surface model, the fusion algorithm expression is:

[0065] ;(Formula 3)

[0066] Among them, C 2，α，d represents the initial human tissue surface model, α is the shape parameter, corresponding to the second shape parameter in this step, d is the offset parameter, B(p i , α+d ) represents the point p i As the center, α is Radius, d is the spherical area of the offset, Represents the point cloud coordinate set of human tissue, p i represents the i-th point cloud point in human tissue, It means to unite the spherical area cloud points formed by various points in human tissue.

[0067] Taking the small intestine model as an example, a set of spatial coordinate points, namely point cloud coordinates, is constructed for the selected small intestine tissue, which is called a package. An offset is set for the package, and the shape of the package is iteratively optimized to obtain the ideal organ tissue shape. i , α+d ) is the point p i As the center, α is Radius, d The spherical area of the offset is adjusted by , you can control the shape details of the package, α The larger it is, the smoother and simpler the generated surface will be. α The smaller it is, the more detail it packs into the package. d Represents the offset parameter, which is used to expand or reduce the shape and size of the package. d >0 means expand outward, d <0 means inward contraction. This change is helpful in adjusting the concave and convex conditions of a certain area, making the wrapped structure closer to the original model.

[0068] Still taking the small intestine model as an example, the fusion algorithm is a wrapping deformation algorithm based on two parameters, including the shape description method of the two parameters α and d. α Represents shape parameters that control the size of cavities or holes that cannot be traversed during fusion, such as grooves and bends in the small intestine; d Represents the offset parameter, which is used to control the distance between the vertices of the output mesh and the vertices of the original mesh.

[0069] Specifically, in the process of obtaining the initial human tissue surface model in step S31b, the preset shape parameters are first α and preset offset parameters d , use formula (3) to perform fine wrapping, compare the result of fine wrapping with the expected shape, if it meets the expected shape, go to step S32; if it does not meet the expected shape, it is necessary to adjust the offset parameter d , to adjust the shape of the surface model after fine wrapping so that it can meet the expected shape requirements.

[0070] For example, the model can be placed in a whole human tissue model to see if there is any intersection or large gap between it and other tissues. If not, the expected shape requirement is met.

[0071] In some embodiments, the second shape parameter is smaller than the first shape parameter.

[0072] For example, in constructing a small intestine model wrapped α It needs to be divided into two steps. First, the original tissue model (such as Figure 2 As shown in (a), the small intestine model is roughly engraved to form the first human tissue surface model, whose outline is a rough range, as shown in Figure 2 Then, a smaller second shape parameter is used to refine the details and form a smooth initial human tissue surface model without gaps, as shown in (b). Figure 2 As shown in (d), the outer edge contour is further segmented and simplified to outline a more accurate tissue contour. Compared with directly using the second shape parameter to identify and segment the boundary of the original tissue model, it is beneficial to save computing time and ensure the accuracy of the model. The outer contour of the first human tissue surface model is compared with the outer contour of the initial human tissue surface model, as shown in Figure 2 As shown in (c), the dark color at the outline represents the outer contour of the first human tissue surface model, and the outline of the light-colored part close to the center is the outer contour of the initial human tissue surface model.

[0073] The first shape parameter, second shape parameter, and offset parameter are determined based on the overall size of the model, the required recognition accuracy, and computational efficiency. Specifically, the second shape parameter must be smaller than the first shape parameter. The larger the overall size of the model, the larger the shape parameter and offset parameter can be, and vice versa. High recognition accuracy requires smaller shape parameters and offset parameters, and vice versa. High computational efficiency requires larger shape parameters and offset parameters, and vice versa.

[0074] S32: Based on the implicit field function and the fusion coefficient, the digital human body geometric model tissue boundary is fused through a fusion algorithm to obtain a human body tissue surface model.

[0075] In some embodiments, based on the implicit field function and according to the fusion coefficient, the digital human body geometric model tissue boundary is fused through a fusion algorithm to obtain a human body tissue surface model. The expression of the fusion algorithm used is:

[0076] ;(Formula 4)

[0077] Among them, C 3，α，d represents the surface model of human tissue, α is the shape parameter, corresponding to the second shape parameter in this step, d is the offset parameter, q is the fusion coefficient, B(p i , α+d ) represents the point p i As the center, α is Radius, d is the spherical area of the offset, Represents the point cloud coordinate set of human tissue, p i represents the i-th point cloud point in human tissue, This means combining the spherical area cloud points formed by various points in the human body. The fusion coefficient q is greater than or equal to 1. The larger the q, the larger the volume of fat that needs to be fused.

[0078] The human intestines are surrounded by abdominal fat. In a normal human body, this fat layer is fluid and has no fixed shape. Modeling the intestines and periintestinal fat separately, creating one intestinal model and one fat mass model, does not conform to the true human anatomy. However, during CT image segmentation, fat and internal organs have different refractive indices for radiation, making it impossible to clearly segment fat tissue in the image and generate a 3D fat model. However, fat does exist in human tissue, wrapping around organs and occupying a certain volume, especially in organs like the intestine.

[0079] To address this issue, the present invention fuses tissue boundaries in digital human body models by adding a fusion coefficient, q, to simulate the intertwining of internal organs with surrounding tissues such as fat. Therefore, adding a fusion coefficient to the tissue-organ fusion process incorporates the volume of fat into the organs, increasing the model's bio-fidelity.

[0080] Taking the small intestine as an example, the intestinal anatomical structure is surrounded by fat, which intersects with the small intestine. Therefore, when fusing only the small intestine, the volume of fat should also be considered, fusing the intestinal model with the periintestinal fat model. Setting q to 1.3 is more consistent with the actual human anatomy while also accounting for the volume of periintestinal fat in the human body.

[0081] In some embodiments, based on the implicit field function, the human body tissue is segmented, simplified and fused by a fusion algorithm to obtain a surface model of the human body tissue, which also includes

[0082] S4: Shape optimization of the surface model of human tissue is performed, and the optimization expression is:

[0083] ;(Formula 5)

[0084] Among them, E is the energy function, s is the arc length of the surface, is the boundary of the surface model of the human tissue to be optimized, k(s) is the boundary curvature, and ds represents the arc length integral of the surface model boundary, that is, the differential arc length element along the surface boundary.

[0085] To optimize the wrapping shape, an energy function, E, can be defined to achieve a smoother shape. The energy function is defined as the system reaching a stable state when the energy value is minimized. In this model, this is when the energy value is minimized and the model reaches its optimal smoothness. During the calculation process, minimizing E yields a smooth offset wrapping surface, removes self-overlaps between patches, and repairs holes formed by the offset.

[0086] To further illustrate the effects of the method for segmenting, simplifying, and fusing digital human body geometry model tissue boundaries in the present invention, the following example is used for demonstration:

[0087] Figure 4 This is a diagram showing the geometric model of the small intestine provided by an embodiment of the present invention. The small intestine has a complex structure and a large number of groove structures on the surface, which is very unfavorable for finite element modeling and model simulation. The small intestine model is wrapped in a bounding box using the shrink wrap method. Starting from the bounding box, it is continuously refined and gradually "carved" out a mesh close to the input geometry. The volume of fat is taken into account and merged with the small intestine. The first shape parameter is set to 0.002m, the second shape parameter is 0.001m, the offset parameter is 0.001m, and q is 1.3. After segmenting, simplifying and fusing the small intestine geometric model, the surface model of the fused and smoothed small intestine model is obtained. The result is as follows: Figure 5 shown.

[0088] The present invention first obtains the point cloud coordinates of each human tissue using a digital human body geometry model. Then, based on the point cloud coordinates of each tissue, the implicit field function of each tissue is determined. Based on the implicit field function, a fusion algorithm is used to segment, simplify, and fuse the tissue boundary to obtain a surface model of the tissue. Using the fusion algorithm, a relatively rough first tissue surface model is first outlined using shape parameters. The shape parameters are then optimized and a preset offset parameter is added to obtain a more detailed initial tissue surface model. The offset parameter can be further adjusted to further adjust the shape of the initial tissue surface model to meet the desired shape requirements, accurately identifying tissue and overcoming the shape recognition challenges of complex tissues with numerous wrinkles and grooves. A fusion coefficient is then added to account for the volume of fat within the organ, increasing the model's biofidelity. Finally, the energy function E is used to optimize the wrapping shape for a smoother shape. Based on these steps, the human tissue structure is accurately and stepwise identified, overcoming the challenges of shape recognition of complex tissues and balancing morphological fidelity with model practicality.

[0089] This method can be applied to small intestine models, brain models, muscle models and other soft tissues. Through this method, they can be quickly integrated and simplified into the expected geometric shape that is conducive to finite element modeling. The appearance of the original model is protected, and the grooves are eliminated, making the surface of the organ or tissue smooth and non-interfering with each other, shortening the finite element modeling development cycle.

[0090] For example, Figure 6 This is a comparison diagram of the large intestine before and after segmentation, simplification and fusion provided by an embodiment of the present invention. Figure 6 The middle left picture shows the large intestine before fusion. Figure 6 The middle right picture shows the fused image.

[0091] For example, Figure 7 This is a comparison diagram of the brain before and after segmentation, simplification and fusion provided by an embodiment of the present invention. Figure 7 The middle left picture shows the brain before fusion. Figure 7 The middle right picture shows the fused image.

[0092] For example, Figure 8 This is a comparison diagram before and after muscle segmentation, simplification and fusion provided by an embodiment of the present invention. Figure 8 The middle left picture shows the muscle before fusion. Figure 8 The middle right picture shows the fused image.

[0093] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A method for segmenting, simplifying and fusing tissue boundaries of a digital human body geometric model, characterized in that: The steps include: Obtain the point cloud coordinates of each tissue of the human body through the digital human body geometric model; Determining the implicit field function of each human tissue according to the point cloud coordinates of each human tissue; the implicit field function is used to characterize the surface relationship between the tissue and the digital human body geometric model; Based on the implicit field function, the human tissue is segmented, simplified, and fused by a fusion algorithm to obtain a surface model of the human tissue, including: Based on the implicit field function, according to the shape parameters and the offset parameters, the digital human body geometric model tissue boundary is segmented and simplified through a fusion algorithm to obtain an initial human body tissue surface model; Based on the implicit field function and according to the fusion coefficient, the initial human tissue surface model is fused through a fusion algorithm to obtain a human tissue surface model.

2. The method for segmenting, simplifying and fusing digital human body geometric model tissue boundaries according to claim 1, characterized in that: Based on the implicit field function, according to the shape parameters and the offset parameters, the digital human body geometric model tissue boundary is segmented and simplified by a fusion algorithm to obtain an initial human body tissue surface model, including: Based on the implicit field function and according to the preset first shape parameter, the digital human body geometric model tissue boundary is segmented and simplified by a fusion algorithm to obtain a first human body tissue surface model; Based on the implicit field function, according to the preset second shape parameter and offset parameter, the first human tissue surface model is segmented and simplified through a fusion algorithm to obtain an initial human tissue surface model.

3. The method for segmenting, simplifying and fusing tissue boundaries of a digital human body geometric model according to claim 2, characterized in that: When the first human tissue surface model is obtained, the expression of the fusion algorithm is: ； Among them, C 1,α represents the first human tissue surface model, α is the shape parameter, corresponding to the first shape parameter in this step, B(p i , α ) represents the point p i As the center, α is A spherical area of radius, Represents the point cloud coordinate set of human tissue, p i represents the i-th point cloud point in human tissue, It means to unite the spherical area cloud points formed by various points in human tissue.

4. The method for segmenting, simplifying and fusing tissue boundaries of a digital human body geometric model according to claim 2, wherein: When the initial human tissue surface model is obtained, the fusion algorithm expression is: ； Among them, C 2，α，d represents the initial human tissue surface model, α is the shape parameter, corresponding to the second shape parameter in this step, d is the offset parameter, B(p i , α+d ) represents the point p i As the center, α is Radius, d is the spherical area of the offset, Represents the point cloud coordinate set of human tissue, p i represents the i-th point cloud point in human tissue, It means to unite the spherical area cloud points formed by various points in human tissue.

5. The method for segmenting, simplifying and fusing tissue boundaries of a digital human body geometric model according to claim 2, characterized in that: The second shape parameter is smaller than the first shape parameter.

6. The method for segmenting, simplifying and fusing tissue boundaries of a digital human body geometric model according to claim 2, characterized in that: When the human tissue surface model is obtained, the expression of the fusion algorithm is: ； Among them, C 3，α，d represents the surface model of human tissue, α is the shape parameter, corresponding to the second shape parameter in this step, d is the offset parameter, q is the fusion coefficient, B(p i , α+d ) represents the point p i As the center, α is Radius, d is the spherical area of the offset, Represents the point cloud coordinate set of human tissue, p i represents the i-th point cloud point in human tissue, It means to unite the spherical area cloud points formed by various points in human tissue.

7. The method for segmenting, simplifying and fusing tissue boundaries of a digital human body geometric model according to claim 1, characterized in that: The expression of the implicit field function is: ； Among them, SDF( x ) represents any node in the coordinate system x The implicit field function at ; α represents the shape parameter, d represents the offset parameter, surface Represents the surface composed of point cloud coordinates, distance ( x,surface ) represents a node x arrive surface The shortest distance on the surface.

8. The method for segmenting, simplifying and fusing tissue boundaries of a digital human body geometric model according to claim 1, characterized in that: Based on the implicit field function, the human tissue is segmented, simplified and fused by a fusion algorithm to obtain a surface model of the human tissue, and the shape of the surface model of the human tissue is optimized. The optimization expression is: ； Among them, E is the energy function, s is the arc length of the surface, is the boundary of the surface model of the human tissue to be optimized, k(s) is the boundary curvature, and ds represents the arc length integral of the boundary of the surface model.

Images