Three-dimensional rhinoplasty simulation method and system based on face model

Abstract

This application relates to a three-dimensional rhinoplasty simulation method and system based on a face model. The method includes: first, acquiring a three-dimensional face mesh model of the object to be simulated; performing three-dimensional facial key point detection on the model to obtain a key point set; calculating the face's symmetrical midline plane based on this set; then solving for the spatial intersection line between the midline plane and the mesh model; extracting the corresponding curve segment as the source curve of the nasal bridge contour; without changing the mesh structure, editing the source curve to obtain the target curve; establishing the parameter correspondence between the source curve and the target curve and generating driving rules; performing a one-time global deformation processing on the mesh model; and finally obtaining the rhinoplasty simulation result. This method, through the decoupling of curve editing and mesh deformation, ensures the accuracy and efficiency of rhinoplasty simulation, effectively solving problems such as insufficient accuracy, stuttering, and poor symmetry in traditional simulations, while also providing an intuitive reference for doctor-patient communication and preoperative planning.

Description

Technical Field

[0001] This invention belongs to the field of computer graphics processing, and in particular relates to a three-dimensional rhinoplasty simulation method and system based on a human face model. Background Technology

[0002] With the deep integration of medical aesthetics and digital technology, precise preoperative planning and effect simulation of rhinoplasty have become core elements for improving surgical safety, optimizing doctor-patient communication efficiency, and meeting personalized medical aesthetic needs. Three-dimensional rhinoplasty simulation technology based on facial models has been widely used in clinical scenarios such as preoperative medical aesthetic assessment and treatment plan design.

[0003] Currently, while existing 3D rhinoplasty simulation technology has achieved basic contour adjustment and effect presentation, it still faces many technical bottlenecks in practical applications: the complex structure of the human face and significant individual differences make it difficult for existing simulation schemes to accurately reproduce the 3D shape of the nasal bridge area; deviations in mid-axis plane positioning and symmetry constraints are prone to occur, resulting in a gap between the simulation effect and actual clinical needs; insufficient precision in mesh deformation control easily leads to problems such as local stretching distortion and destruction of symmetrical structures, making it difficult to balance the precise adjustment of the nasal bridge contour with the overall harmony of facial shape; deviations in mid-axis plane calculation and contour curve extraction are prone to occur, resulting in insufficient symmetry and poor curve smoothness, affecting the accuracy of the simulation;

[0004] Meanwhile, existing technologies struggle to effectively decouple editing operations from mesh deformation. Editing can easily disrupt the overall mesh shape, and the lack of clear standards for iterative optimization and convergence condition definition leads to unstable simulation results and low computational efficiency, failing to meet the actual clinical needs for refined preoperative planning, personalized simulation, and efficient doctor-patient communication in rhinoplasty. Furthermore, the lack of unified standards in key areas such as curve extraction, weight allocation, and convergence control further restricts simulation accuracy and clinical adaptability. Summary of the Invention

[0005] Therefore, it is necessary to provide a three-dimensional rhinoplasty simulation method and system based on a human face model that can effectively solve the problems of inaccurate contour adjustment, uncontrolled mesh deformation, and insufficient symmetry in traditional simulation methods, so that the rhinoplasty simulation effect can better meet the actual clinical needs and improve the accuracy and practicality of the simulation.

[0006] Firstly, this application provides a three-dimensional rhinoplasty simulation method based on a face model, including:

[0007] Obtain a 3D face mesh model of the object to be simulated, perform 3D facial key point detection on the 3D face mesh model, and obtain a set of 3D facial key points.

[0008] The facial symmetry midline is calculated based on a set of three-dimensional facial key points; the facial symmetry midline is a reference plane that passes through the midline of the bridge of the nose and is symmetrical with the left and right facial regions.

[0009] Solve for the spatial intersection line between the symmetrical midline plane of the face and the surface of the 3D face mesh model, and extract the curve segment from the center of the eyebrows to the upper lip position in the spatial intersection line to obtain the source curve of the nose bridge contour of the side face.

[0010] While keeping the geometry and vertex coordinates of the 3D face mesh model constant and without triggering any real-time mesh deformation, target shape editing processing is performed on the source curve of the side nose bridge contour to obtain the target curve of the side nose bridge contour.

[0011] Establish a parameterized spatial point pair correspondence between the source curve of the side profile nose bridge contour and the target curve of the side profile nose bridge contour, and generate mesh deformation driving rules.

[0012] A one-time global deformation process is performed on the 3D face mesh model based on the mesh deformation-driven rule to obtain the 3D rhinoplasty simulation result mesh.

[0013] In one embodiment, the calculation of the face symmetry midline plane based on a three-dimensional set of facial key points includes:

[0014] Core symmetry benchmark key points and nose bridge-specific anchor points are selected from a set of 3D facial key points.

[0015] An initial fitting plane is constructed based on the core symmetric benchmark key points using least squares fitting.

[0016] Initial Fitting Plane Satisfy the plane equation ;in, This represents the initial fitting plane normal vector. The three-dimensional centroid of the core symmetric reference key point. Represents any point in three-dimensional space. This represents the dot product of vectors.

[0017] The midline vector of the bridge of the nose formed by the specific anchor points of the bridge of the nose To constrain this, a rigid translation and rotation transformation is performed on the initial fitting plane so that the nasal bridge midline vector satisfies... The corrected plane is obtained ;in, This represents the corrected plane normal vector.

[0018] Based on minimizing the objective function of symmetric deviation For the corrected plane Symmetric robustness optimization is performed to obtain the final face symmetry midline surface. ;in, Indicates the first The signed distance from each core symmetric datum key point to the corrected plane.

[0019] In one embodiment, solving for the spatial intersection line between the face's symmetry midline plane and the surface of the 3D face mesh model includes:

[0020] Calculate any vertex in each triangle of a 3D face mesh model Relative to the central axis of human face symmetry Signed distance ;in, Representing a plane The unit normal vector, Representing a plane Internal fixation points.

[0021] For signed distance Dissimilar, with vertices located on the plane of facial symmetry. The intersection points of the triangular facets on both sides with the symmetrical central plane of the face are calculated by linear interpolation.

[0022] If vertex , Cross-plane and The coordinates of the intersection point are: .

[0023] Connect all intersection points according to their spatial adjacency to form a spatial intersection line between the central plane of the face's symmetry and the surface of the three-dimensional face mesh model.

[0024] In one embodiment, the target shape editing process includes interactive editing based on control points, specifically:

[0025] According to the arc length parameter along the contour curve of the nose bridge of the side face Several control points were obtained through uniform sampling, and these control points were connected using a smooth spline curve; among them, It represents the total arc length of the contour curve of the nose bridge on the side of the face.

[0026] Respond to user drag operations on any control point, and constrain the dragged control point to the face symmetry midline plane. Inside, it satisfies the plane equation .

[0027] After each control point movement, the smoothed spline curve is updated to serve as the target curve for the current profile nose bridge contour.

[0028] The 3D face mesh model maintains its original vertex coordinates throughout the entire process of control point editing.

[0029] In one embodiment, the target morphology editing process includes automatic editing based on a pre-trained model, specifically:

[0030] The source curve of the nasal bridge contour of the side face and the facial aesthetic features of the simulated object are input into a pre-trained curve prediction model, and the target curve of the nasal bridge contour of the side face is output.

[0031] The visualization of the side profile nose bridge contour is replaced with the target curve of the side profile nose bridge contour in the visualization view.

[0032] In one embodiment, the target shape editing process includes editing based on curve template loading, specifically:

[0033] Select a target template curve from the preset nose shape curve template library, perform endpoint alignment and parameter domain linear mapping processing on the target template curve, match the endpoint position and parameter domain of the side nose bridge contour source curve, and obtain the side nose bridge contour target curve.

[0034] In one embodiment, a one-time global deformation process is performed on the 3D face mesh model based on mesh deformation driving rules, including:

[0035] The source curve of the side profile nose bridge contour and the target curve of the side profile nose bridge contour are matched with the same arc length parameter. One-to-one correspondence, generating a sequence of point pairs with matching parameters. ;in, This represents the total arc length of the contour curve of the nose bridge in profile. Represents the source curve parameter points of the profile nose bridge contour. This represents the parameter points of the target curve for the profile of the nose bridge.

[0036] Preset mesh vertex influence radius Construct mesh deformation driving rules:

[0037] Traverse all vertices of the 3D face mesh model and calculate each vertex. The closest distance to the contour curve of the nose bridge on the side of the face .

[0038] like Then, based on the displacement vector corresponding to the vertex and according to distance decay weight Apply displacement operations to the vertices.

[0039] like If so, then the vertex coordinates remain unchanged.

[0040] A one-time global deformation process is completed based on mesh deformation-driven rules to obtain the mesh of the three-dimensional rhinoplasty simulation result.

[0041] The one-time global deformation processing also includes:

[0042] Based on the current 3D face mesh model that has completed global deformation, the intersection line between the 3D face mesh model and the face symmetry midline plane is resolved, and the intersection line is used as the source curve of the updated side face nose bridge contour.

[0043] Using the profile of the nose bridge as the target curve, the 3D face mesh model is deformed again.

[0044] Until the corresponding parameter point deviation between the updated profile nose bridge contour source curve and the profile nose bridge contour target curve. The iteration stops when the shape of the 3D face mesh model tends to converge; among them, This indicates the preset convergence threshold.

[0045] Secondly, this application also provides a three-dimensional rhinoplasty simulation system based on a face model, the system comprising:

[0046] The facial key point detection module is used to obtain a 3D facial mesh model of the object to be simulated, perform 3D facial key point detection on the 3D facial mesh model, and obtain a set of 3D facial key points.

[0047] The Symmetry Mid-Axis Plane Calculation Module is used to calculate the symmetry mid-axis plane of a face based on a set of three-dimensional facial key points. The symmetry mid-axis plane of a face is a reference plane that passes through the midline of the bridge of the nose and is symmetrical with the left and right facial regions.

[0048] The source curve extraction module is used to solve the spatial intersection line between the symmetrical midline plane of the face and the surface of the 3D face mesh model, and to extract the curve segment from the center of the eyebrows to the upper lip position in the spatial intersection line to obtain the source curve of the nose bridge contour of the side face.

[0049] The target curve editing module is used to perform target shape editing processing on the source curve of the side nose bridge contour while keeping the geometry and vertex coordinates of the 3D face mesh model constant and without triggering any real-time mesh deformation, so as to obtain the target curve of the side nose bridge contour.

[0050] The mesh deformation driving module is used to establish the parameterized spatial point pair correspondence between the source curve of the side profile nose bridge contour and the target curve of the side profile nose bridge contour, and generate mesh deformation driving rules; it is also used to perform a one-time global deformation processing on the 3D face mesh model based on the mesh deformation driving rules to obtain the 3D rhinoplasty simulation result mesh.

[0051] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described above.

[0052] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned method.

[0053] The aforementioned 3D rhinoplasty simulation method, system, computer equipment, and storage medium based on a face model acquire a 3D face mesh model of the object to be simulated. 3D facial key point detection is performed on the 3D face mesh model to extract and obtain a set of 3D facial key points. Based on this set of 3D facial key points, a symmetrical midline plane of the face is calculated. This midline plane is a reference plane that passes through the midline of the nasal bridge and is symmetrical with the left and right facial regions. The spatial intersection line between this symmetrical midline plane and the surface of the 3D face mesh model is solved, and the curve segment from the center of the eyebrows to the upper lip region is extracted as the source curve of the nasal bridge contour. While maintaining the geometric shape and vertex coordinates of the 3D face mesh model and without triggering any real-time mesh deformation, target shape editing processing is performed on the source curve of the nasal bridge contour to obtain the target curve of the nasal bridge contour. A parameterized spatial point pair correspondence between the source curve and the target curve of the nasal bridge contour is established, generating a mesh deformation driving rule. Based on this driving rule, a one-time global deformation processing is performed on the 3D face mesh model, finally obtaining the 3D rhinoplasty simulation result mesh.

[0054] The above method effectively decouples the editing of the nasal bridge contour curve from the mesh deformation, ensuring precise adjustment of the nasal bridge contour in profile while avoiding the stuttering and errors caused by real-time mesh deformation. By locating the midline plane of facial symmetry and extracting the intersection line, the symmetry and accuracy of the nasal bridge contour are ensured, effectively solving problems such as inaccurate contour adjustment, uncontrolled mesh deformation, and insufficient symmetry in traditional simulation methods. At the same time, the one-time global deformation mode significantly improves simulation efficiency and reduces computational redundancy compared to real-time point-by-point deformation. Furthermore, through parameterized point pair correspondence and distance weight control, the rhinoplasty simulation effect is more in line with actual clinical needs, improving the accuracy and practicality of the simulation, reducing the difficulty of preoperative planning, and providing intuitive and accurate visual references for doctor-patient communication, further ensuring the safety and rationality of rhinoplasty surgery. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 A flowchart illustrating a three-dimensional rhinoplasty simulation method based on a human face model, provided in an embodiment of the present invention;

[0057] Figure 2 This is a structural block diagram of a three-dimensional rhinoplasty simulation system based on a human face model, provided in an embodiment of the present invention. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0059] In one embodiment, such as Figure 1 As shown, this application provides a three-dimensional rhinoplasty simulation method based on a face model, which may include the following steps:

[0060] Step S101: Obtain the three-dimensional face mesh model of the object to be simulated, perform three-dimensional facial key point detection on the three-dimensional face mesh model, and obtain a set of three-dimensional facial key points.

[0061] Specifically, the original 3D face mesh model of the object to be simulated is obtained, and the coordinates of all 3D vertices, the topological connection relationship of triangular facets, and the geometric data of the model surface are read. The 3D facial key point detection algorithm is called to perform global feature recognition and point localization on the imported 3D face mesh model. Various feature spatial points corresponding to the facial contour, facial structure, and bridge of the nose are extracted in sequence. All detected feature points are integrated and collected to finally form a complete set of 3D facial key points.

[0062] Step S102: Calculate the face symmetry midline based on the three-dimensional facial key point set; the face symmetry midline is a reference plane that passes through the midline of the bridge of the nose and is symmetrical with the left and right facial regions.

[0063] Based on the acquired set of 3D facial key points, facial symmetry reference feature points and nasal bridge midline feature points are selected. The face symmetry midline surface is obtained by solving spatial plane fitting and rigid constraint correction calculation. The face symmetry midline surface is a spatial reference plane that runs through the nasal bridge midline and can realize geometric symmetry constraints on the left and right areas of the face. This plane will serve as a unified spatial reference reference for the entire process of subsequent contour extraction, curve shape editing, and mesh deformation constraints.

[0064] Step S103: Solve for the spatial intersection line between the symmetrical midline plane of the face and the surface of the three-dimensional face mesh model, and extract the curve segment from the center of the eyebrows to the upper lip position in the spatial intersection line to obtain the source curve of the nose bridge contour of the side face.

[0065] Traverse all triangular faces of the 3D face mesh model, solve for the spatial coordinates of the intersection points between the face's symmetrical central axis and each triangular face, integrate all intersection points and connect them according to spatial adjacency topology to form a complete spatial intersection line; combine the anatomical position range of the human face to define the effective contour range, extract the effective curve segment extending from the center of the eyebrows to the upper lip on the spatial intersection line, remove irrelevant and redundant line segments in the intersection line, complete the curve selection and range limitation, and finally generate the original profile nose bridge contour source curve.

[0066] Step S104: While keeping the geometry and vertex coordinates of the 3D face mesh model constant and without triggering any real-time mesh deformation, perform target shape editing processing on the source curve of the side face nose bridge contour to obtain the target curve of the side face nose bridge contour.

[0067] Throughout the entire curve editing process, the overall geometry, mesh topology, and spatial coordinates of all vertices of the original 3D face mesh model are kept constant. Real-time mesh deformation calculations are not performed, and no modifications to the vertex coordinates of the original model are executed. Under the above constraints, the target shape editing process is carried out on the acquired profile nose bridge contour source curve to complete the reconstruction and adjustment of the curve shape, and finally generate the profile nose bridge contour target curve that meets the shape requirements.

[0068] Step S105: Establish the parameterized spatial point pair correspondence between the source curve of the side profile nose bridge contour and the target curve of the side profile nose bridge contour, and generate mesh deformation driving rules.

[0069] A unified arc length parameterized mapping method is adopted to normalize the parameters of the source curve and the target curve of the nasal bridge contour of the side face, respectively, and establish a one-to-one correspondence between parameterized spatial points between the two curves, clarifying the spatial mapping relationship of each parameter point. Combining the logic of determining the influence range of mesh vertices, calculating spatial distance, and allocating displacement weights, the point offset relationship is defined in a regular way, and integrated into a complete and executable mesh deformation driving rule.

[0070] Step S106: Perform a one-time global deformation process on the 3D face mesh model based on the mesh deformation driving rules to obtain the 3D rhinoplasty simulation result mesh.

[0071] The pre-built mesh deformation driving rules are invoked to perform global traversal calculations on the original 3D face mesh model. Based on the deformation rules, the displacement calculations and coordinate updates of all mesh vertices are completed. A unified one-time global mesh deformation processing is performed. The deformation adjustment is completed in an overall batch calculation mode throughout the process, without performing step-by-step real-time deformation iterations. After all deformation calculations are completed, the final output is the 3D rhinoplasty simulation result mesh model.

[0072] The aforementioned 3D rhinoplasty simulation method based on a face model first obtains a 3D facial mesh model of the subject to be simulated, and performs 3D facial key point detection on it to obtain a set of key points. Based on this set, the symmetrical midline plane of the face is calculated, and then the spatial intersection line between the midline plane and the mesh model is solved. The corresponding curve segment is extracted as the source curve of the nasal bridge contour of the side face. Without changing the mesh structure, the source curve is edited to obtain the target curve. The parameter correspondence between the source curve and the target curve is established and driving rules are generated. A one-time global deformation processing is performed on the mesh model to finally obtain the rhinoplasty simulation result. This method ensures the accuracy and efficiency of rhinoplasty simulation by decoupling curve editing and mesh deformation, effectively solving the problems of insufficient accuracy, stuttering, and poor symmetry in traditional simulations, while providing an intuitive reference for doctor-patient communication and preoperative planning.

[0073] In one embodiment, calculating the face symmetry midline plane based on a three-dimensional set of facial key points may include the following steps:

[0074] Step S201: Select core symmetry reference key points and nose bridge-specific anchor points from the three-dimensional facial key point set.

[0075] Step S202: Construct an initial fitting plane based on the core symmetric benchmark key points using least squares fitting.

[0076] Step S203, Initial Fitting Plane Satisfy the plane equation ;in, This represents the initial fitting plane normal vector. The three-dimensional centroid of the core symmetric reference key point. Represents any point in three-dimensional space. This represents the dot product of vectors.

[0077] Step S204, the nasal bridge midline vector formed by the nasal bridge-specific anchor points. To constrain this, a rigid translation and rotation transformation is performed on the initial fitting plane so that the nasal bridge midline vector satisfies... The corrected plane is obtained ;in, This represents the corrected plane normal vector.

[0078] Step S205, based on the objective function of minimizing symmetric deviation For the corrected plane Symmetric robustness optimization is performed to obtain the final face symmetry midline surface. ;in, Indicates the first The signed distance from each core symmetric datum key point to the corrected plane.

[0079] Specifically, the acquired set of 3D facial key points is first categorized and filtered to extract core symmetry reference key points for overall facial symmetry fitting, as well as nasal bridge-specific anchor points for locating the central structure of the nasal bridge. The filtered core symmetry reference key points are used as the basic computational data, and a least-squares fitting algorithm is used to perform spatial plane fitting operations on all points, generating an initial fitting plane that can initially fit the overall symmetry structure of the face. Then, based on the spatial distribution of the nasal bridge-specific anchor points, a corresponding nasal bridge midline spatial vector is constructed. This vector is used as a rigid transformation constraint condition, and the generated initial fitting plane is sequentially subjected to spatial translation and orientation transformations to adjust the plane's spatial position and orientation, ensuring that the plane's normal vector direction maintains a perpendicular constraint relationship with the nasal bridge midline structure. This completes the initial plane correction and yields the corrected plane. Finally, using the spatial distance deviation from all core symmetry reference key points to the plane as the optimization basis, a global robust iterative optimization is performed on the corrected plane using the objective optimization criterion of minimizing symmetry deviation. This reduces the fitting error caused by discrete key points, fine-tunes the plane's spatial pose, and ultimately calculates the facial symmetry midline plane that conforms to the true symmetry structure of the face.

[0080] This embodiment combines key point hierarchical screening, initial plane fitting, structural constraint correction, and symmetry deviation optimization into a multi-stage operation, which can effectively avoid the plane deviation problem caused by a single fitting method. Combined with the nasal bridge-specific structural constraints, it ensures that the mid-axis plane fits the real midline structure of the face. At the same time, the overall robustness of the plane fitting is improved by optimizing the objective function, reducing the calculation error caused by discrete feature points. The final obtained facial symmetry mid-axis plane benchmark has higher accuracy and stronger symmetry, which can provide a stable and reliable spatial reference for subsequent nasal bridge contour curve extraction, curve shape symmetry constraint, and mesh deformation benchmark limitation.

[0081] In one embodiment, solving for the spatial intersection line between the face's symmetry midline plane and the surface of the three-dimensional face mesh model may include the following steps:

[0082] Step S301: Calculate any vertex in each triangle of the 3D face mesh model. Relative to the central axis of human face symmetry Signed distance ;in, Representing a plane The unit normal vector, Representing a plane Internal fixation points.

[0083] Step S302, for signed distance Dissimilar, with vertices located on the plane of facial symmetry. The intersection points of the triangular facets on both sides with the symmetrical central plane of the face are calculated by linear interpolation.

[0084] Step S303, if vertex , Cross-plane and The coordinates of the intersection point are: .

[0085] Step S304: Connect all intersection points according to spatial adjacency to form a spatial intersection line between the face symmetry axis plane and the surface of the three-dimensional face mesh model.

[0086] Specifically, the process iterates through all triangular faces within the 3D face mesh model. For each vertex of a triangular face, the signed spatial distance of each vertex relative to the constructed face symmetry axis is calculated. The sign and magnitude of the distance values ​​determine the spatial orientation of each vertex relative to the reference plane. Then, all triangular faces are filtered for validity. Cross-plane triangular faces whose vertices are distributed on both sides of the face symmetry axis and whose corresponding distance values ​​differ are identified. Invalid triangular faces where all vertices are on the same side of the plane and there is no possibility of intersection are removed. For the filtered valid cross-plane triangular faces, spatial linear interpolation is used to calculate the intersection points. Based on the weighted proportions of the distances from the vertices on both sides of the triangular face to the reference plane, spatial coordinate interpolation is derived to accurately determine the spatial points where each valid triangular face intersects with the face symmetry axis. After completing the intersection calculation of all triangular facets, all spatial intersection data are summarized. Based on the inherent topological adjacency relationship, facet connection order, and spatial location correlation characteristics of the 3D face mesh model, all intersection points are sequentially connected according to the adjacent arrangement rules, ultimately forming a continuous and complete global spatial intersection line between the face symmetry axis plane and the surface of the 3D face mesh model that fits the curved shape of the model surface.

[0087] This embodiment uses this method to solve for spatial intersection lines, which can accurately calculate the intersection points across the entire domain based on the grid topology. It effectively improves the problems of discontinuous intersection lines and distorted contour edges caused by uneven distribution of grid vertices, ensuring that the obtained intersection line trajectory is complete and smooth with high geometric accuracy. This provides stable and accurate basic geometric data for subsequent contour cutting from the brow center to the upper lip and extraction of the source curve of the side profile nose bridge contour.

[0088] In one embodiment, the target shape editing process includes interactive editing based on control points, specifically:

[0089] Step S401, follow the contour curve of the side of the nose bridge according to the arc length parameter Several control points were obtained through uniform sampling, and these control points were connected using a smooth spline curve; among them, It represents the total arc length of the contour curve of the nose bridge on the side of the face.

[0090] Step S402: Respond to the user's drag operation on any control point, and constrain the dragged control point to the face symmetry midline plane. Inside, it satisfies the plane equation .

[0091] Step S403: After each control point movement, update the smoothed spline curve as the target curve for the current profile nose bridge contour.

[0092] Preferably, the original vertex coordinates of the 3D face mesh model remain unchanged throughout the entire process of control point editing.

[0093] Specifically, the system uniformly samples along the nasal bridge contour source curve according to the arc length parameter range to obtain several evenly distributed control points. A smooth spline curve is then used to continuously connect all control points, forming the initial contour spline curve. The system responds in real-time to user dragging and adjusting of any control point, strictly constraining the dragged control points within the face's symmetrical midline plane, ensuring that the control points remain within the spatial range of this plane. Each time a control point moves, the system automatically recalculates and updates the smooth spline curve, using the updated spline curve as the current nasal bridge contour target curve. Throughout the entire control point editing process, the original vertex coordinates of the 3D face mesh model remain fixed and unchanged.

[0094] This embodiment uses the above-described method to solve for spatial intersection lines, which can fully utilize the topological structure of the 3D face mesh to accurately capture the intersection points of the central plane and the model surface. This effectively avoids problems such as intersection line breakage and contour distortion caused by uneven vertex distribution and differences in surface structure, ensuring the continuity and accuracy of the intersection lines. At the same time, by filtering valid intersection points and eliminating invalid points, the accuracy of the intersection lines is further improved, providing a stable and accurate geometric basis for subsequent extraction of the contour from the brow center to the upper lip region and the extraction of a qualified nose bridge contour, avoiding the impact of intersection line errors on subsequent simulation results.

[0095] In one embodiment, the target shape editing process includes automatic editing based on a pre-trained model, specifically:

[0096] Step S501: Input the source curve of the side profile nose bridge contour and the facial aesthetic features of the simulated object into the pre-trained curve prediction model, and output the target curve of the side profile nose bridge contour.

[0097] Step S502: In the visualization view, the source curve of the side profile nose bridge contour is replaced with the target curve of the side profile nose bridge contour.

[0098] Specifically, the geometric shape data corresponding to the source curve of the profile nose bridge contour is extracted, and facial aesthetic feature data such as the facial structure proportions and the distribution of facial features of the subject to be simulated are collected. These two types of data are then input into a pre-trained curve prediction model. The model, combining the feature mapping rules learned during training, performs comprehensive calculations and analysis on the input contour data and facial feature data. Based on individual facial features, it performs adaptive inference of the curve shape and finally outputs a target profile nose bridge contour curve that fits the facial structure. After the target curve is generated, the system updates and replaces the original source curve in the visualization interface with the newly generated target profile nose bridge contour curve for visualization.

[0099] This embodiment achieves automated editing and generation of the nasal bridge contour through a pre-trained model. It can combine the facial structural features of different objects to complete personalized curve adaptation, reduce manual intervention, and improve the processing efficiency of contour adjustment. At the same time, it realizes real-time visual replacement and update of curve results, which facilitates intuitive comparison of contour shape differences. The generated target contour has good structural adaptability and can provide a stable and reliable target shape basis for subsequent mesh deformation calculation.

[0100] In one embodiment, the target shape editing process includes editing based on curve template loading, specifically:

[0101] Select a target template curve from the preset nose shape curve template library, perform endpoint alignment and parameter domain linear mapping processing on the target template curve, match the endpoint position and parameter domain of the side nose bridge contour source curve, and obtain the side nose bridge contour target curve.

[0102] Optionally, based on the requirements for adjusting the bridge of the nose, a suitable target template curve is selected from a pre-built and stored library of nose shape templates. Geometric calibration is then performed on the selected target template curves sequentially. First, precise alignment of the curve endpoints is achieved, unifying the spatial coordinates of the two curves. Then, a linear mapping transformation is performed on the overall parameter domain of the curves to complete scale normalization and interval adaptation adjustments. This ensures that the parameter range and spatial trend of the target template curve match the source curve of the side profile nose bridge contour, eliminating positional offsets and scale differences between the template curve and the original contour. Finally, the entire adaptation transformation is completed, generating the corresponding target curve for the side profile nose bridge contour.

[0103] In this embodiment, the method relies on a preset template library to generate the contour shape. Combined with endpoint alignment and parameter domain mapping correction, it can quickly complete the transformation of the nose bridge contour shape while ensuring a stable connection between the adjusted curve and the original contour structure. The processing flow has simple computational logic, strong shape standardization, and can efficiently generate target contours that meet the shape requirements.

[0104] In one embodiment, performing a one-time global deformation process on a 3D face mesh model based on mesh deformation driving rules may include the following steps:

[0105] Step S601: Match the source curve of the side profile nose bridge contour with the target curve of the side profile nose bridge contour using the same arc length parameter. One-to-one correspondence, generating a sequence of point pairs with matching parameters. ;in, This represents the total arc length of the contour curve of the nose bridge in profile. Represents the source curve parameter points of the profile nose bridge contour. This represents the parameter points of the target curve for the profile of the nose bridge.

[0106] Step S602, preset the radius of influence of mesh vertices Construct mesh deformation driving rules:

[0107] Step S603: Traverse all vertices of the 3D face mesh model and calculate the values ​​of each vertex. The closest distance to the contour curve of the nose bridge on the side of the face .

[0108] Step S604, if Then, based on the displacement vector corresponding to the vertex and according to distance decay weight Apply displacement operations to the vertices.

[0109] Step S605, if If so, then the vertex coordinates remain unchanged.

[0110] Step S606: Perform a one-time global deformation process based on the mesh deformation driving rules to obtain the mesh of the three-dimensional rhinoplasty simulation result.

[0111] The one-time global deformation processing also includes:

[0112] Step S607: Based on the current 3D face mesh model that has completed global deformation, re-solve the intersection line between the 3D face mesh model and the face symmetry midline plane, and use the intersection line as the updated side face nose bridge contour source curve.

[0113] Step S608: Using the profile nose bridge contour target curve as the deformation target, perform deformation operation on the 3D face mesh model again.

[0114] Step S609, until the corresponding parameter point deviation between the updated profile nose bridge contour source curve and the profile nose bridge contour target curve is found. The iteration stops when the shape of the 3D face mesh model tends to converge; among them, This indicates the preset convergence threshold.

[0115] Specifically, the source curve of the side profile nose bridge contour and the target curve of the side profile nose bridge contour are matched one-to-one with the same arc length parameter to generate a sequence of spatial point pairs with completely matched parameters. Each point pair consists of a parameter point on the source curve of the side profile nose bridge contour and a corresponding parameter point on the target curve of the side profile nose bridge contour. All point pairs are synchronized with parameters based on the total arc length of the source curve of the side profile nose bridge contour. A preset influence radius of the mesh vertex is used as the basis for constructing a complete mesh deformation driving rule: First, all vertices of the 3D face mesh model are traversed, and the nearest spatial distance from each vertex to the source curve of the side profile nose bridge contour is calculated one by one. If the distance is less than or equal to the preset influence radius, the vertex displacement weight is calculated based on the displacement vector of the corresponding parameter point pair and the distance decay weight formula. The displacement operation is applied to the vertex according to the weight to achieve precise adjustment of the vertex coordinates. If the distance is greater than the preset influence radius, the original coordinates of the vertex are kept unchanged, and no displacement operation is performed. Based on the aforementioned mesh deformation-driven rules, a one-time global deformation process is performed on the 3D face mesh model, while simultaneously initiating an iterative optimization process: After each deformation, based on the currently deformed 3D face mesh model, the spatial intersection line between the face's symmetry midline plane and the mesh model surface is resolved, and this intersection line is used as the updated profile nose bridge contour source curve; subsequently, using the initially determined profile nose bridge contour target curve as the deformation reference, the 3D face mesh model is deformed again. This process of deformation, intersection line update, and further deformation is repeated until the deviation between the corresponding parameter points of the updated profile nose bridge contour source curve and the profile nose bridge contour target curve is less than a preset convergence threshold. At this point, the shape of the 3D face mesh model is determined to be stable, the deformation has reached the preset accuracy, the iterative process is stopped, and the final 3D rhinoplasty simulation result mesh is obtained.

[0116] This embodiment achieves precise point-to-point matching between the source and target curves through synchronous arc length parameter synchronization, ensuring the accuracy of displacement-driven simulation. By utilizing preset influence radius and distance attenuation weights, gradient deformation of mesh vertices is achieved, avoiding abrupt local deformations and unnatural transitions, while reducing invalid deformation of irrelevant vertices and lowering computational redundancy. Through multiple rounds of iterative optimization and deviation judgment, deformation errors are gradually corrected, ensuring a high degree of fit between the nasal bridge contour and the target curve in the final simulation result, improving the accuracy and naturalness of rhinoplasty simulation. The entire process achieves standardization and precision in deformation-driven simulation, effectively solving problems such as insufficient accuracy, uncontrolled deformation, and poor convergence in traditional deformation methods. Furthermore, the combination of one-time global deformation and iterative optimization balances simulation efficiency and accuracy, providing precise, stable, and natural technical support for preoperative rhinoplasty simulation in clinical settings.

[0117] In one embodiment, this application also provides a method for clinical application of three-dimensional rhinoplasty simulation based on a face model, which may include:

[0118] For subjects undergoing pre-rhinoplasty planning, a 3D facial structure scan is performed to obtain full-domain 3D point cloud data of the subject's face. Through point cloud denoising, hole repair, and mesh reconstruction, a 3D facial mesh model with complete topological structure and qualified geometric accuracy is generated. This model is imported into a 3D rhinoplasty simulation system and coordinate normalization is performed. The system automatically calls the facial key point detection module to perform full-domain feature detection on the normalized 3D facial mesh model, extracting key feature points such as brow bone, bridge of nose, alar of nose, tip of nose, and upper lip to form a set of 3D facial key points unique to the subject.

[0119] The symmetry mid-plane calculation module, based on this key point set, completes the selection of core symmetry benchmark key points, initial plane fitting, nasal bridge vector constraint correction, and symmetry deviation optimization, accurately constructing a symmetry mid-plane that fits the subject's facial anatomy. The source curve extraction module traverses all triangular faces of the face mesh, calculates the spatial intersections between the symmetry mid-plane and the mesh surface, and connects them in an orderly manner. The curve segment from the center of the eyebrows to the upper lip is extracted to generate the subject's original profile nasal bridge contour source curve. On the clinical interactive terminal, the doctor, combining nasal aesthetic standards, the subject's facial proportions, and personalized plastic surgery needs, uses the target curve editing module to interactively drag and adjust the profile nasal bridge contour source curve, or selects a suitable clinical standard template from the nasal curve template library to complete the mapping and matching. Throughout the process, the original three-dimensional face mesh model remains unchanged, generating a profile nasal bridge contour target curve that conforms to the clinical surgical plan.

[0120] The mesh deformation driving module establishes a point-to-point correspondence between the source curve and the target curve based on the arc length parameterization rule. It presets the influence radius and distance attenuation weights of the mesh vertices, constructs exclusive deformation driving rules, performs a one-time global deformation on the 3D face mesh model, and it iteratively updates the intersection line and repeatedly corrects the deformation until the deviation of the nasal bridge contour curve reaches the preset convergence threshold, generating the final 3D rhinoplasty simulation result mesh.

[0121] The system displays the original model and simulation results simultaneously from multiple perspectives on the clinical visualization interface, presenting quantitative indicators such as nasal bridge height, curvature, and symmetry in real time. Based on the simulation results, doctors can communicate with subjects before surgery, and after confirming the effect, they can repeatedly adjust the curve parameters and re-perform deformation calculations until a consensus is reached. Finally, the system automatically extracts key point data, symmetry midline plane parameters, contour curve parameters, deformation weights, and iteration convergence information from the simulation process, integrates them to generate a standardized rhinoplasty preoperative simulation report, and archives and stores the simulation result grid, report file, and all parameter data in a unified manner in the clinical database. At the same time, the simulation result data can be connected to the nasal surgery navigation system to provide digital reference for precise intraoperative operation.

[0122] Based on the same inventive concept, this application also provides a system for implementing the aforementioned three-dimensional rhinoplasty simulation method based on a face model. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in one or more embodiments of the three-dimensional rhinoplasty simulation system based on a face model provided below can be found in the limitations of the three-dimensional rhinoplasty simulation method based on a face model described above, and will not be repeated here.

[0123] In one embodiment, such as Figure 2 As shown, this application also provides a three-dimensional rhinoplasty simulation system based on a face model, the system may include:

[0124] The facial key point detection module 701 is used to acquire a three-dimensional face mesh model of the object to be simulated, perform three-dimensional facial key point detection on the three-dimensional face mesh model, and obtain a set of three-dimensional facial key points.

[0125] The symmetry mid-axis plane calculation module 702 is used to calculate the symmetry mid-axis plane of the face based on the three-dimensional facial key point set; the symmetry mid-axis plane of the face is a reference plane that passes through the midline of the bridge of the nose and is symmetrical with the left and right facial regions.

[0126] The source curve extraction module 703 is used to solve the spatial intersection line between the central plane of facial symmetry and the surface of the three-dimensional facial mesh model, and to extract the curve segment from the center of the eyebrows to the upper lip position in the spatial intersection line to obtain the source curve of the nose bridge contour of the side face.

[0127] The target curve editing module 704 is used to perform target shape editing processing on the source curve of the side nose bridge contour while keeping the geometry and vertex coordinates of the 3D face mesh model constant and without triggering any real-time mesh deformation, so as to obtain the target curve of the side nose bridge contour.

[0128] The mesh deformation driving module 705 is used to establish the parameterized spatial point pair correspondence between the source curve of the side nasal bridge contour and the target curve of the side nasal bridge contour, and generate mesh deformation driving rules; it is also used to perform a one-time global deformation processing on the three-dimensional face mesh model based on the mesh deformation driving rules to obtain the three-dimensional rhinoplasty simulation result mesh.

[0129] The aforementioned 3D rhinoplasty simulation system based on a face model completes the overall 3D rhinoplasty simulation process through the sequential collaboration of various functional modules, with data being transferred and seamlessly connected between modules. The facial key point detection module acquires the 3D face mesh model of the object to be simulated and performs 3D facial key point detection processing on the model, collecting all feature point data to form a complete 3D facial key point set. The symmetry mid-axis plane calculation module receives the key point set and constructs the face symmetry mid-axis plane based on the feature data. This plane is a spatial reference plane passing through the midline of the nasal bridge, constraining the left and right facial regions to maintain symmetry. The source curve extraction module solves for the spatial intersection line formed by the intersection of this symmetry mid-axis plane and the surface of the 3D face mesh model, and extracts the effective curve segment from the center of the eyebrows to the upper lip position on the intersection line to obtain the side nasal bridge contour source curve. The target curve editing module, while maintaining the geometric shape and all vertex coordinates of the 3D face mesh model unchanged throughout the process and without performing any real-time mesh deformation calculations, performs morphological editing processing on the side nasal bridge contour source curve to generate the corresponding side nasal bridge contour target curve. The mesh deformation driving module establishes a parameterized spatial point pair correspondence between the source curve and the target curve, thereby generating a complete mesh deformation driving rule. Based on this rule, a one-time global deformation processing is performed on the original 3D face mesh model, and finally the 3D rhinoplasty simulation result mesh is output.

[0130] The system's functional module architecture in this embodiment is clearly hierarchical, with well-defined division of labor and close integration among each stage. Precise feature detection and symmetry benchmark construction provide a reliable data foundation for contour curve extraction; curve editing and mesh deformation are independently separated, effectively avoiding interference to the original model structure during the editing process; relying on parametric mapping relationships to drive global mesh deformation enables standardized and controllable model morphology adjustments, improving the overall accuracy and structural rationality of rhinoplasty simulation. The system operates stably and reliably, and can be stably applied to pre-operative simulation scenarios for rhinoplasty.

[0131] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0132] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the three-dimensional rhinoplasty simulation method based on a face model as described above.

[0133] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0134] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0135] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. A three-dimensional rhinoplasty simulation method based on a face model, characterized in that, The method includes: Obtain a 3D face mesh model of the object to be simulated, and perform 3D facial key point detection on the 3D face mesh model to obtain a set of 3D facial key points; The facial symmetry midline plane is calculated based on the three-dimensional facial key point set; the facial symmetry midline plane is a reference plane that passes through the midline of the bridge of the nose and is symmetrical with the left and right facial regions; Solve for the spatial intersection line between the symmetrical central plane of the face and the surface of the three-dimensional face mesh model, and extract the curve segment from the center of the eyebrows to the upper lip position in the spatial intersection line to obtain the source curve of the nose bridge contour of the side face; While keeping the geometry and vertex coordinates of the three-dimensional face mesh model constant and without triggering any real-time mesh deformation, target shape editing processing is performed on the source curve of the side nose bridge contour to obtain the target curve of the side nose bridge contour. Establish a parameterized spatial point pair correspondence between the source curve of the side profile nose bridge contour and the target curve of the side profile nose bridge contour, and generate mesh deformation driving rules. Based on the mesh deformation driving rules, a one-time global deformation process is performed on the three-dimensional face mesh model to obtain the three-dimensional rhinoplasty simulation result mesh.

2. The method according to claim 1, characterized in that, The calculation of the facial symmetry midline plane based on the three-dimensional facial key point set includes: Select core symmetry benchmark key points and nose bridge-specific anchor points from the set of three-dimensional facial key points; Based on the key points of the core symmetry benchmark, an initial fitting plane is constructed using least squares fitting. The initial fitting plane Satisfy the plane equation ;in, This represents the initial fitting plane normal vector. The three-dimensional centroid of the core symmetric reference key point. Represents any point in three-dimensional space. Represents the dot product of vectors; The nasal bridge midline vector formed by the aforementioned nasal bridge-specific anchor points As a constraint, a rigid translation and rotation transformation is performed on the initial fitting plane to ensure that the nasal bridge midline vector satisfies... The corrected plane is obtained ;in, This represents the corrected plane normal vector; Based on minimizing the objective function of symmetric deviation For the corrected plane Symmetric robustness optimization is performed to obtain the final face symmetry midline surface. ;in, Indicates the first The signed distance from each of the core symmetric reference key points to the corrected plane.

3. The method according to claim 1, characterized in that, The step of solving for the spatial intersection line between the symmetrical midline plane of the face and the surface of the three-dimensional face mesh model includes: Calculate any vertex in each triangle of the 3D face mesh model. Relative to the central axis of the face Signed distance ;in, Representing a plane The unit normal vector, Representing a plane Internal fixation points; For the signed distance Dissimilar, the vertices are located on the symmetrical midline plane of the face. The intersection points of the triangular facets on both sides with the symmetrical central plane of the face are calculated by linear interpolation. If the vertex , Cross-plane and The coordinates of the intersection point are: ; Connect all the intersection points according to their spatial adjacency to form the spatial intersection line between the central plane of the face symmetry and the surface of the three-dimensional face mesh model.

4. The method according to claim 1, characterized in that, The target shape editing process includes interactive editing based on control points, specifically: According to the arc length parameter along the contour curve of the nose bridge of the face Several control points are obtained through uniform sampling, and these control points are connected using a smooth spline curve; wherein, This represents the total arc length of the source curve of the nasal bridge contour of the side profile; Responding to the user's drag operation on any of the control points, and constraining the dragged control points to the central axis of the face symmetry plane. Inside, it satisfies the plane equation ; Each time the control point moves, the smooth spline curve is updated as the target curve for the current profile nose bridge contour; The three-dimensional face mesh model maintains its original vertex coordinates throughout the entire process of control point editing.

5. The method according to claim 1, characterized in that, The target morphology editing process includes automatic editing based on a pre-trained model, specifically: The source curve of the side profile nose bridge contour and the facial aesthetic features of the simulated object are input into a pre-trained curve prediction model, and the target curve of the side profile nose bridge contour is output. The visualization of the side profile nose bridge contour source curve is replaced by the target curve of the side profile nose bridge contour in the visualization view.

6. The method according to claim 1, characterized in that, The target shape editing process includes editing based on curve template loading, specifically: Select a target template curve from the preset nose shape curve template library, perform endpoint alignment and parameter domain linear mapping processing on the target template curve, match the endpoint position and parameter domain of the side nose bridge contour source curve, and obtain the side nose bridge contour target curve.

7. The method according to claim 1, characterized in that, The step of performing a one-time global deformation process on the 3D face mesh model based on the mesh deformation driving rule includes: The source curve of the side profile nose bridge contour and the target curve of the side profile nose bridge contour are aligned with the same arc length parameter. One-to-one correspondence, generating a sequence of point pairs with matching parameters. ;in, This represents the total arc length of the source curve of the nasal bridge contour of the profile. The parameter points of the source curve of the side profile nose bridge contour are represented. This represents the target curve parameter points of the side profile nose bridge contour; Preset mesh vertex influence radius Construct mesh deformation-driven rules: Traverse all vertices of the 3D face mesh model and calculate the values ​​of each vertex. The closest distance to the source curve of the nasal bridge contour of the side face ; like Then, based on the displacement vector corresponding to the vertex and according to distance decay weight A displacement operation is applied to the vertex; like If so, then the vertex coordinates remain unchanged; A one-time global deformation process is completed based on the mesh deformation driving rules to obtain the three-dimensional rhinoplasty simulation result mesh; The one-time global deformation processing further includes: Based on the three-dimensional face mesh model that has completed global deformation, the intersection line between the three-dimensional face mesh model and the face symmetry midline plane is resolved, and the intersection line is used as the updated side face nose bridge contour source curve. Using the target curve of the nose bridge contour of the side face as the deformation target, the three-dimensional face mesh model is deformed again; Until the corresponding parameter point deviation between the updated source curve of the side profile nose bridge contour and the target curve of the side profile nose bridge contour. The iteration stops when the shape of the 3D face mesh model tends to converge; wherein, This indicates the preset convergence threshold.

8. A three-dimensional rhinoplasty simulation system based on a human face model, characterized in that, The system includes: The facial key point detection module is used to acquire a three-dimensional face mesh model of the object to be simulated, and to perform three-dimensional facial key point detection on the three-dimensional face mesh model to obtain a set of three-dimensional facial key points. The symmetry mid-axis plane calculation module is used to calculate the symmetry mid-axis plane of the face based on the three-dimensional facial key point set; the symmetry mid-axis plane of the face is a reference plane that passes through the midline of the bridge of the nose and is symmetrical with the left and right facial regions; The source curve extraction module is used to solve the spatial intersection line between the symmetrical central plane of the face and the surface of the three-dimensional face mesh model, and to extract the curve segment from the center of the eyebrows to the upper lip position in the spatial intersection line to obtain the source curve of the side nose bridge contour. The target curve editing module is used to perform target shape editing processing on the source curve of the side nose bridge contour while keeping the geometry and vertex coordinates of the three-dimensional face mesh model constant and without triggering any real-time mesh deformation, so as to obtain the target curve of the side nose bridge contour. The mesh deformation driving module is used to establish the parameterized spatial point pair correspondence between the source curve of the side nasal bridge contour and the target curve of the side nasal bridge contour, and generate mesh deformation driving rules; it is also used to perform a one-time global deformation processing on the three-dimensional face mesh model based on the mesh deformation driving rules to obtain the three-dimensional rhinoplasty simulation result mesh.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images