Method and system for generating three-dimensional support structures based on a simplified medial axis transform
By obtaining the center point through the central axis transformation technique, constructing the branch structure and selecting the central axis, and combining the special center point and slicing algorithm to generate the three-dimensional support structure, the problem of not being able to accurately extract the semantic information and geometric features of the model in the existing technology is solved, and a higher quality three-dimensional model simplification is achieved.
Patent Information
- Application Number
- CN202411334033.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-24
- Publication Date
- 2025-12-26
- Estimated Expiration
- 2044-09-24
AI Technical Summary
Existing 3D model slicing algorithms cannot accurately extract the semantic information and geometric features of the model, resulting in significant differences between the simplified support structure and the initial model.
The center point is obtained by using the central axis transformation technique, a branch structure is constructed, the branch structure with the largest average radius is selected as the central axis, and slices are made by combining the special center point and the preset total number of slices to generate a three-dimensional support structure. The final structure is then optimized by reconstructing errors.
It improves the ability to capture semantic information and geometric features during the simplification process of 3D models, generates more accurate and reasonable 3D support structures, and reduces reconstruction errors.
Smart Images
Figure CN119206133B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of three-dimensional model simplification, and particularly relates to a three-dimensional support structure generation method and system based on simplified center axis transformation. BACKGROUND
[0002] The generation of three-dimensional model support structure is mainly by using the uniform slicing method, but the conventional slicing method often only focuses on the selection of the main axis and the number of slices in processing, so that the operation is convenient for data processing and processing, and the demand of the support structure in the actual application scene is ignored.
[0003] The current common three-dimensional model slicing algorithm mainly includes the following two kinds: (1) generating a cross section by solving the intersection points of the grid patches and the cross section and connecting these intersection points; (2) seeking the best slicing direction and generating a layered file combined with the preset layer thickness. However, when determining the position and direction of the slices, the two methods do not consider the difference in semantic information contained in different parts of the model, so they cannot effectively capture and extract important information in the model. This limitation is particularly evident when dealing with models with complex geometric features and rich semantic information. Therefore, it is necessary to provide a three-dimensional support structure generation method and system based on simplified center axis transformation. SUMMARY
[0004] The present application provides a three-dimensional support structure generation method based on simplified center axis transformation. To solve the problem that in the prior art, when simplifying a three-dimensional model, the semantic information and geometric features contained in the model cannot be accurately extracted, resulting in a large difference between the simplified support structure and the initial three-dimensional model.
[0005] The three-dimensional support structure generation method based on simplified center axis transformation provided by the present application comprises: obtaining a three-dimensional model to be processed; performing center axis transformation and simplification processing on the three-dimensional model to obtain a plurality of center points, and constructing a plurality of branch structures according to the radius change amplitudes between the plurality of center points; selecting a branch structure with the largest average radius from the branch structures, setting it as a center axis, and forming different main axes according to different connection modes of the center axis and each of the remaining branch structures.
[0006] Processing each main shaft: identifying a special center point of the main shaft from each center point corresponding to the main shaft, and based on the special center point and a preset total amount of slices, slicing the three-dimensional model on the main shaft according to a corresponding slice algorithm to obtain a plurality of slices; wherein a slice is a cross section of the three-dimensional model through the main shaft; based on each slice, generating a three-dimensional support structure corresponding to the three-dimensional model, and comparing the generated structure with the three-dimensional model to obtain a reconstruction error; after all main shafts are processed, generating a final three-dimensional support structure of the three-dimensional model according to the main shaft with the smallest reconstruction error and all slices corresponding thereto.
[0007] In an embodiment of the present application, the center axis transformation and simplification processing of the three-dimensional model to obtain a plurality of center points, and constructing a plurality of branch structures according to the radius variation amplitude between the plurality of center points, comprises: performing center axis transformation and simplification processing on the three-dimensional model to obtain a plurality of center points; for each center point: calculating the radius variation amplitude between the center point and its adjacent center points, and connecting the center points with a radius variation amplitude less than a preset amplitude threshold to obtain the branch structure corresponding to the connected center points.
[0008] In an embodiment of the present application, the center axis transformation and simplification processing of the three-dimensional model to obtain a plurality of center points, and constructing a plurality of branch structures according to the radius variation amplitude between the plurality of center points, comprises: performing center axis transformation and simplification processing on the three-dimensional model to obtain a plurality of center points; for each center point: calculating the radius variation amplitude between the center point and its adjacent center points, and connecting the center points with a radius variation amplitude less than a preset amplitude threshold to obtain the branch structure corresponding to the connected center points.
[0009] In an embodiment of the present application, the center axis transformation and simplification processing of the three-dimensional model to obtain a plurality of center points, and constructing a plurality of branch structures according to the radius variation amplitude between the plurality of center points, comprises: performing center axis transformation and simplification processing on the three-dimensional model to obtain a plurality of center points; for each center point: calculating the radius variation amplitude between the center point and its adjacent center points, and connecting the center points with a radius variation amplitude less than a preset amplitude threshold to obtain the branch structure corresponding to the connected center points.
[0010] In an embodiment of the present application, the slicing the three-dimensional model on the main shaft according to a corresponding slicing algorithm based on the special center point and a preset total amount of slices comprises: slicing the three-dimensional model for each branch center point based on the main shaft to obtain a special slice corresponding to each branch center point; slicing the three-dimensional model for each radius maximum center point in each region based on the main shaft to obtain a special slice corresponding to each radius maximum center point; calculating the number of additional slices according to the total amount of slices and the number of special slices corresponding to the branch center points and the radius maximum center points; selecting one of the slicing algorithms from a preset slicing algorithm library in sequence, slicing the three-dimensional model on the main shaft according to the number of additional slices to obtain a plurality of additional slices, and taking the plurality of special slices and the plurality of additional slices as all slices corresponding to the main shaft.
[0011] In an embodiment of the present application, the generating a three-dimensional support structure corresponding to the three-dimensional model based on each slice and comparing the generated structure with the three-dimensional model to obtain a reconstruction error comprises: generating an initial three-dimensional support structure corresponding to the main shaft and all slices corresponding to the main shaft; and comparing the generated initial three-dimensional support structure with the three-dimensional model to obtain a reconstruction error corresponding to the main shaft.
[0012] In another aspect of the present application, a three-dimensional support structure generation system based on simplified central axis transformation is also provided. The system comprises: a sample acquisition module for acquiring a three-dimensional model to be processed; a central axis transformation module for performing central axis transformation and simplification processing on the three-dimensional model to obtain a plurality of center points and construct a plurality of branch structures according to the radius variation amplitudes between the center points; a main shaft screening module for screening a branch structure with the largest average radius from the branch structures, setting the branch structure as a central axis, and forming different main shafts according to different connection modes of the central axis and each of the remaining branch structures; a reconstruction module for processing each main shaft: identifying a special center point of the main shaft from the center points corresponding to the main shaft, and slicing the three-dimensional model on the main shaft according to a corresponding slicing algorithm based on the special center point and a preset total amount of slices to obtain a plurality of slices; wherein the slice is a cross section of the three-dimensional model passing through the main shaft; generating a three-dimensional support structure corresponding to the three-dimensional model based on each slice, and comparing the generated structure with the three-dimensional model to obtain a reconstruction error; and a support structure generation module for generating a final three-dimensional support structure of the three-dimensional model according to the main shaft with the smallest reconstruction error and all slices corresponding thereto after processing of all main shafts is completed.
[0013] The application provides a three-dimensional support structure generation method and system based on simplified central axis transformation. A set of key center points are extracted from an original three-dimensional model through the central axis transformation technology, and a plurality of branch structures are constructed according to the radius change amplitude between the center points. The constructed branch structures can effectively capture the basic geometric features of the three-dimensional model. The branch structure with the largest average radius in the branch structure is taken as a central axis, and is connected with different branch structures to form different main axes. From the center points corresponding to the main axes, special center points which are crucial for understanding the key areas of the three-dimensional model are identified. According to the preset slice number and the number of special center points, the three-dimensional model is sliced. The slices can not only provide detailed profile views of the three-dimensional model, but also facilitate more in-depth structural analysis. Then for each main axis: the reconstruction error of the three-dimensional support structure generated by the slice and the original three-dimensional model is calculated, the main axis with the smallest error and the corresponding slices are selected, and the final three-dimensional support structure of the original three-dimensional model is generated, so that the important features of the original three-dimensional model can be more accurately and reasonably expressed. BRIEF DESCRIPTION OF DRAWINGS
[0014] Figure 1 A flowchart of a three-dimensional support structure generation method based on simplified central axis transformation provided by an embodiment of the application is shown.
[0015] Figure 2 A result schematic diagram of each change process in an embodiment of the application is shown.
[0016] Figure 3 A flowchart of obtaining special center points and generating support structures in an embodiment of the application is shown.
[0017] Figure 4 A structure block diagram of a three-dimensional support structure generation system based on simplified central axis transformation provided by an embodiment of the application is shown. DETAILED DESCRIPTION
[0018] The embodiments of the application are described below by way of specific examples. Those skilled in the art can easily understand other advantages and effects of the application from the disclosure. The application can also be implemented or applied by different specific embodiments, and the details in the specification can be modified or changed based on different views and applications without departing from the spirit of the application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.
[0019] It is to be noted that the drawings provided in the following embodiments only schematically illustrate the basic concepts of the present application, and only the components related to the present application are shown in the drawings, rather than being drawn according to the number, shape and size of the components in actual implementation, and the type, number and proportion of each component in actual implementation can be arbitrarily changed, and the component layout type can also be more complex.
[0020] In the following description, a large number of details are discussed to provide a more thorough explanation of the embodiments of the present application, however, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details, and in other embodiments, the known structures and devices are shown in the form of block diagrams rather than in the form of details, to avoid making the embodiments of the present application difficult to understand.
[0021] The present application provides a three-dimensional support structure generation method based on simplified center axis transformation, based on the simplified center axis transformation expression, the change of each center point and radius generated after the center axis transformation is analyzed, according to the spatial relationship of each center point, the direction of the radius change between the center points is selected, which is less than the preset amplitude threshold, and the center points meeting the requirements are sequentially connected to obtain a branch structure. After all the branch structures are obtained, one of the branch structures with the largest average radius is selected as the center axis, and the remaining branch structures are selected as alternative branches. The center axis and all alternative branches are combined according to each possible connection mode to form different main axes. Each main axis is composed of a center axis and at least one alternative branch connected thereto. A series of special slices are obtained according to the special center points on the main axis, and the additional slices of the main axis are filled according to different filling strategies in the pre-stored algorithm library. According to the center point connection line of each slice, an initial three-dimensional support structure is formed, and the reconstruction error of the current initial three-dimensional support structure and the original three-dimensional model is calculated. The current reconstruction error is compared with the recorded reconstruction error, and when the current reconstruction error is smaller than the recorded reconstruction error, the recorded reconstruction error is updated using the current reconstruction error, and the slice algorithm corresponding to the current reconstruction error is recorded. Then, for the next main axis, the next slice algorithm in the algorithm library is used for filling, and the above process is repeated. Finally, after all the main axes are processed, the main axis with the smallest reconstruction error and the corresponding slice algorithm are used as the final scheme, and the final three-dimensional support structure of the three-dimensional model is obtained using the scheme. The present application can be applied to the simplification processing of three-dimensional models, and based on the simplified center axis transformation expression, a three-dimensional wireframe support structure with higher quality and more reasonable structure is generated.
[0022] As shown in Figure 1 , the three-dimensional support structure generation based on simplified center axis transformation includes the following steps:
[0023] S1, obtaining a three-dimensional model to be processed;
[0024] S2. Perform a central axis transformation and simplification on the three-dimensional model to obtain multiple center points, and construct multiple branch structures based on the radius variation between the multiple center points;
[0025] S3. Select the branch structure with the largest average radius from the multiple branch structures, set it as the central axis, and form different main axes according to the different connection methods between the central axis and each of the other branch structures;
[0026] S4. Processing for each main axis: Identify the special center point of the main axis from the center points corresponding to each main axis, and based on the special center point and the preset total number of slices, slice the 3D model on the main axis according to the corresponding slicing algorithm to obtain multiple slices; wherein, the slice is a cross-section of the 3D model through the main axis; based on each slice, generate a 3D support structure corresponding to the 3D model, and compare the generated structure with the 3D model to obtain the reconstruction error;
[0027] S5. After all the main axes have been processed, the final three-dimensional support structure of the three-dimensional model is generated based on the main axis with the smallest reconstruction error and all the corresponding slices.
[0028] The following provides a detailed explanation of each step:
[0029] S1. Obtain the 3D model to be processed.
[0030] like Figure 2 As shown in "a) Original Model", the 3D model M to be processed is a 3D model that needs to be simplified. By simplifying the 3D model using the method described in this invention, the corresponding 3D support structure can be obtained. It is understood that the 3D model can be obtained through 3D scanning or generated in 3D modeling software, and the specific method of obtaining it is not limited.
[0031] S2. Perform a central axis transformation and simplification on the three-dimensional model to obtain multiple center points, and construct multiple branch structures based on the radius variation between the multiple center points.
[0032] Medial Axis Transform (MAT) is a method for extracting the central axis or internal skeleton of a 3D model's geometry. After performing a medial axis transformation on the 3D model, a series of center points are obtained, where each center point is located inside the 3D model and has a corresponding radius value.
[0033] Specifically, in an embodiment of the present application, the three-dimensional model is subjected to a center-axis transformation and a simplification process to obtain a plurality of center points, and a plurality of branch skeleton structures are constructed according to the radius variation amplitudes between the plurality of center points, including:
[0034] The three-dimensional model is subjected to a center-axis transformation and a simplification process to obtain a plurality of center points;
[0035] For each center point, the radius variation amplitude between the center point and its adjacent center points is calculated, and the center points with a radius variation amplitude less than a preset amplitude threshold are connected to obtain the branch structure corresponding to the connected center points. After the three-dimensional model M is subjected to a center-axis transformation, a plurality of points equidistant from the surface of the three-dimensional model are obtained, which are used as initial center points, and each initial center point has a radius radius. The radius radius is the spherical radius distance from the corresponding initial center point to the surface of the three-dimensional model. These initial center points and corresponding radius radii constitute an initial center-axis transformation expression MAT t (as shown in Figure 2 b) initial center-axis transformation expression). The existing method generally realizes the expression of the three-dimensional model through the initial center points and corresponding radius radii. Although the center-axis transformation can greatly reduce the complexity of the model expression, the inventors have found that due to the uneven surface of the three-dimensional model, a large amount of redundant data will appear in the initial center-axis transformation expression MAT t , and therefore a simplification algorithm is needed to simplify the preliminary center-axis transformation expression. Specifically, for each initial center point, the distance between it and its adjacent initial center points can be calculated, and the points that are far away from other initial center points are selected as isolated center points from the initial center points according to the distance. The spatial density of each initial center point in the three-dimensional model is calculated, and the complexity e of the initial center-axis transformation is calculated according to the spatial density and the number of isolated center points, wherein the complexity e reflects the geometric complexity of the three-dimensional model and the degree of detail of the center-axis transformation.
[0036] According to the calculated complexity e, the simplification coefficient θ of the initial center-axis transformation expression MAT t can be determined, and a plurality of grid simplification algorithms are used to simplify each initial center point in the initial center-axis transformation expression MAT t under the constraint of the simplification coefficient θ, to simplify it to below a preset threshold, to obtain a plurality of center points, and form a simplified center-axis transformation expression MAT s (as shown in Figure 2S1, simplifying the initial center point set to obtain a simplified center point set, wherein the simplification includes, but is not limited to, reducing the number of initial center points, merging adjacent initial center points, or removing unimportant initial center points, and the simplified center point set is more concise and representative than the initial center point set. s S2, selecting a center point randomly from the simplified center point set, calculating the variation range of the radius between the selected center point and the center points around the selected center point, and connecting all the center points meeting the condition along the direction in which the variation range is less than a preset range threshold. Figure 2 S3, selecting a center point randomly from the simplified center point set, calculating the variation range of the radius between the selected center point and the center points around the selected center point, and connecting all the center points meeting the condition along the direction in which the variation range is less than a preset range threshold.
[0037] S3, selecting a center point randomly from the simplified center point set, calculating the variation range of the radius between the selected center point and the center points around the selected center point, and connecting all the center points meeting the condition along the direction in which the variation range is less than a preset range threshold.
[0038] Specifically, in an embodiment of the present application, the step of selecting a center point randomly from the simplified center point set, calculating the variation range of the radius between the selected center point and the center points around the selected center point, and connecting all the center points meeting the condition along the direction in which the variation range is less than a preset range threshold, includes:
[0039] S3, selecting a center point randomly from the simplified center point set, calculating the variation range of the radius between the selected center point and the center points around the selected center point, and connecting all the center points meeting the condition along the direction in which the variation range is less than a preset range threshold.
[0040] S3, selecting a center point randomly from the simplified center point set, calculating the variation range of the radius between the selected center point and the center points around the selected center point, and connecting all the center points meeting the condition along the direction in which the variation range is less than a preset range threshold.
[0041] S3, selecting a center point randomly from the simplified center point set, calculating the variation range of the radius between the selected center point and the center points around the selected center point, and connecting all the center points meeting the condition along the direction in which the variation range is less than a preset range threshold.
[0042] S3, selecting a center point randomly from the simplified center point set, calculating the variation range of the radius between the selected center point and the center points around the selected center point, and connecting all the center points meeting the condition along the direction in which the variation range is less than a preset range threshold. L= {L1, L2,..., L m}, wherein L i represents one of the branch structures, and m represents the total number of the branch structures. Since each center point has a corresponding radius value, all the branch structures are traversed, for each branch structure, all the center points contained therein are traversed, and according to the radius value of each center point and the total number of the center points, the average radius of the center points of the branch structure is obtained. After the average radius of the center points of all the branch structures is calculated, all the average radii of the center points are compared, and the branch structure with the largest average radius is selected as the center axis L m . The center axis is connected with each of the remaining branch structures to form different main axes, and a main axis set G A = {A1, A2,..., A l} is obtained, wherein A i represents one of the possible connection schemes, and the structure after the connection is referred to as a main axis, and the connection scheme is referred to as a main axis scheme. It can be understood that, since the combinations of the center axis and the branch structures are different, the main axes formed are also different. For example, the main axis A includes the center axis and the branch structure 1 and the branch structure 2, and the main axis B includes the center axis and the branch structure 2.
[0043] S4, processing is performed on each main axis:
[0044] S41, special center points of the main axis are identified from the center points corresponding to the main axis, and the three-dimensional model is sliced on the main axis according to a corresponding slicing algorithm based on the special center points and a preset total amount of slices, to obtain a plurality of slices; wherein the slice is a cross section of the three-dimensional model passing through the main axis;
[0045] S42, a three-dimensional support structure corresponding to the three-dimensional model is generated based on the slices, and the generated structure is compared with the three-dimensional model to obtain a reconstruction error.
[0046] For each main axis: from all the center points on the main axis, the properties of the center points are evaluated according to a preset evaluation principle, such as the number of branches connected to the center point or the radius of the center point, so as to select a plurality of points with special properties from the center points of the main axis as special center points. According to the special center points and a preset total amount of slices, a slicing algorithm is sequentially selected from an algorithm library, and the three-dimensional model is sliced according to the direction of the main axis, to obtain a series of slices. According to the obtained slices, a three-dimensional support structure corresponding to the main axis is constructed, and the three-dimensional support structure is compared with the original three-dimensional model to calculate the reconstruction error therebetween.
[0047] Specifically, in an embodiment of the present application, the special center points of the main axis are identified from the center points corresponding to the main axis, including:
[0048] For the center point located on the main axis, select the center points that are connected to the center points located outside the main axis and use them as branch center points;
[0049] The selected branch center points are deleted from the center points on the main axis. Based on the pre-set regions of the 3D model, the center point with the largest radius is selected from the center points that are still located on the main axis after deletion as the center point with the largest radius in the region; wherein, the radius is the distance from the corresponding center point to the 3D model.
[0050] The center point of each branch and the center point with the largest radius in each region are taken as the special center point of the main axis.
[0051] like Figure 3 As shown, all center points on the main axis form the set of main axis center points. For each center point in the set of main axis center points: determine whether the center point is connected to N or more other center points simultaneously. If so, the center point is designated as a branch center point and removed from the set of main axis center points. Other center points are those located on non-main axes. It is understood that the specific value of N can be adapted to the needs of the actual model simplification by those skilled in the art; the specific value is not limited here. For example, N is 3. Since the branch center points on the main axis retain rich semantic information, they need to be extracted and recorded. After deleting all branch center points from the set of main axis center points, based on the position of each center point on the main axis in the deleted set of main axis center points, and based on the pre-set regions of the 3D model, traverse each center point in a forward-to-back order. In each region, select the center point with the largest radius as the center point with the largest radius in that region. The center point with the largest radius in a region retains the most obvious features within that region. Construct a set G of special center points by designating all branch center points and all center points with the largest radius in each region as special center points. v .
[0052] When simplifying a 3D model, the user needs to pre-set a total number of slices. Based on the number of special center points and the user-preset total number of slices, the number of additional slices can be obtained. Based on the additional slices and special center points, different slicing algorithms are used to slice the 3D model to obtain multiple slices.
[0053] Specifically, in one embodiment of the present invention, the step of slicing the three-dimensional model along the main axis according to a corresponding slicing algorithm based on the special center point and a preset total number of slices to obtain multiple slices includes:
[0054] Based on the main axis, the three-dimensional model is sliced for each branch center point to obtain a special slice corresponding to each branch center point.
[0055] Based on the main axis, the three-dimensional model is sliced according to the center point of the largest radius in each region to obtain special slices corresponding to each center point of the largest radius.
[0056] The number of additional slices is calculated based on the total number of slices, the number of special slices corresponding to the branch center point and the center point with the largest radius;
[0057] One slicing algorithm is selected sequentially from a preset slicing algorithm library. Based on the number of additional slices, the 3D model is sliced on the main axis to obtain multiple additional slices. The multiple special slices and the multiple additional slices are used together as all slices corresponding to the main axis.
[0058] Specifically, for the set of special central points G v In one embodiment of the present invention, the step of slicing the three-dimensional model based on the main axis for each branch center point to obtain a special slice corresponding to each branch center point includes:
[0059] Find the branch with the longest distance from the center point of each branch, excluding the central axis.
[0060] On the central axis, along the line connecting the center point of each branch to the endpoint of the branch with the longest distance from the branch other than the central axis, the three-dimensional model is sliced to obtain special slices corresponding to the center points of each branch.
[0061] like Figure 3 As shown, in the branching structure, the branch with the longest distance (excluding the central axis) corresponding to each branch center point is determined, and this branch is taken as the target branch for the corresponding branch center point. A line is drawn connecting the end of this branch center point and the corresponding target branch, and the 3D model is sliced along the central axis along the direction of this line to obtain a special slice corresponding to each branch center point. This slicing algorithm aims to capture the structural features from the branch center point to the corresponding target branch. This method of slicing highlights significant linear features in the 3D model structure.
[0062] For a special set of central points G v In this algorithm, for each region, the 3D model is sliced along a direction perpendicular to the central axis to obtain a special slice containing the largest radius center point within that region. This special slice's plane passes through the corresponding region's largest radius center point and perpendicularly through the central axis. This slicing algorithm can better highlight specific regions of the 3D model that have the largest volume or the most significant protrusions. This is achieved through the set of special center points G. v All the special slices obtained will form a special slice set. Where, N sThe special slice quantity is the sum of the special slice corresponding to the special center point and the special slice corresponding to the maximum radius center point in the region. s The user's preset total slice quantity N and the special slice quantity N s Subtracting, the number of additional slices needed to fill N is obtained r According to the preset slice algorithm in the algorithm library, the obtained additional slice quantity is evenly distributed on the main shaft, the three-dimensional model is sliced, and a plurality of additional slices are obtained to form an additional slice set The additional slice set And the special slice set Merging, a plurality of slices of the three-dimensional model are obtained to form a group of slice sets G P It should be noted that if the total slice quantity N is less than the special slice quantity N s The user needs to reset the total slice quantity N until the total slice quantity N is greater than or equal to the special slice quantity N s .
[0063] Further, in an embodiment of the present application, the preset slice algorithm is multiple, and the three-dimensional model is sliced on the main shaft according to the number of additional slices based on the preset slice algorithm, a plurality of additional slices are obtained, and the plurality of special slices and the plurality of additional slices are collectively used as the slices, comprising:
[0064] For each preset slice algorithm:
[0065] According to the number of additional slices, the three-dimensional model is sliced on the main shaft to obtain a group of different additional slices; wherein a plurality of special slices and a group of additional slices are used as a group of slices.
[0066] Because different slice algorithms correspond to different filling strategies, for each preset slice algorithm in the algorithm library, the main shaft is filled according to the corresponding filling strategy according to the number of additional slices, the slicing of the three-dimensional model is realized, and the obtained additional slices are also different. Therefore, for each slice algorithm, the obtained additional slices and a plurality of special slices will constitute a set of slice sets of the three-dimensional model. In the present application, the concept of special center point of the center axis transformation is defined, and the generation position of the slice is defined. In the selection of the slice position, the present application proposes the concept of the special center point, that is, there are certain center points with certain characteristics on the main shaft of the center axis transformation skeleton, which contain the most abundant semantic information in a certain interval and can represent the general characteristics of the interval. At the same time, because the total number of slices needs to be set artificially, when the number of slices is greater than the number of special center points, there will be remaining slices, and the positions of these slices will be determined by trying different methods in the preset algorithm library. It can be understood that the slice algorithm is the corresponding slice filling strategy obtained by geometric constraint, and each time the main shaft is filled with additional slices, a slice algorithm is selected to fill and process the main shaft.
[0067] S5、All main shafts are processed, and the final three-dimensional support structure of the three-dimensional model is generated according to the main shaft with the minimum reconstruction error and all the slices corresponding to the main shaft.
[0068] According to the front and rear order of the center points corresponding to each slice on the main shaft L, the three-dimensional support structure F is connected. Specifically, in an embodiment of the present application, based on each slice, the three-dimensional support structure corresponding to the three-dimensional model is generated, including:
[0069] According to each group of slices, a plurality of initial three-dimensional support structures are generated;
[0070] The plurality of generated initial three-dimensional support structures are compared with the three-dimensional model to obtain a plurality of errors;
[0071] The initial three-dimensional support structure with the minimum error is taken as the three-dimensional support structure.
[0072] For each group of slices, an initial three-dimensional support structure F is generated, and for each initial three-dimensional support structure F: the error between the initial three-dimensional support structure F and the three-dimensional model is calculated, and the error is recorded as E F , the error E F is compared with the minimum error E min , and the smaller value is used to update the minimum error E min . When the error calculation of all initial three-dimensional support structures is completed, the final minimum error E min corresponding initial three-dimensional support structure is selected as the final three-dimensional support structure of the three-dimensional model. The initial value of the minimum error E min is infinity.
[0073] Further, in an embodiment of the present application, according to each group of the slices, a corresponding initial three-dimensional support structure is generated, including:
[0074] Selecting a plurality of edge points on the edge of each slice in the group at a plurality of different preset angles;
[0075] Connecting the edge points of the same angle on each slice in sequence to generate a connection line;
[0076] Combining each connection line to generate the initial three-dimensional support structure corresponding to the group of slices.
[0077] In each group of slice sets G P , for the center point of each slice P i , a plurality of rays with different preset angles are emitted from the center point to the outside, and the intersection points of the rays and the edges of the slices are taken as edge points A ij , where i represents the serial number of the corresponding slice, and j represents the serial number of the corresponding angle. The angle between each edge point and the main shaft is the angle between the corresponding ray and the main shaft. Wherein, the angle refers to the angle formed by the ray and the main shaft. According to the front and back order of the edge points on the main shaft, the edge points of the same angle on each slice are connected (i.e. all edge points A ij with the same angle serial number j are connected in order), forming a connection line. Combining each connection line forms a continuous frame structure, which is the initial three-dimensional support structure.
[0078] For example, Figure 4As shown, the three-dimensional support structure generation system 100 based on simplified central axis transformation includes a sample acquisition module 110, a central axis transformation module 120, a main axis screening module 130, a reconstruction module 140, a slice construction module 150, and a support structure generation module 160. The sample acquisition module 110 is configured to acquire a three-dimensional model to be processed. The central axis transformation module 120 is configured to perform central axis transformation and simplification on the three-dimensional model, obtain a plurality of center points, and construct a plurality of branch structures according to the radius variation amplitudes between the center points. The main axis screening module 130 is configured to screen a branch structure with the largest average radius from the plurality of branch structures, set the branch structure as a central axis, and form different main axes according to different connection modes of the central axis and each of the remaining branch structures. The reconstruction module 140 is configured to process each main axis: identify a special center point of the main axis from the center points corresponding to the main axis, and slice the three-dimensional model according to a corresponding slice algorithm on the main axis based on the special center point and a preset total amount of slices to obtain a plurality of slices; wherein a slice is a cross section of the three-dimensional model passing through the main axis; generate a three-dimensional support structure corresponding to the three-dimensional model based on each slice, and compare the generated structure with the three-dimensional model to obtain a reconstruction error. The support structure generation module 150 is configured to generate a final three-dimensional support structure of the three-dimensional model according to the main axis with the smallest reconstruction error and all slices corresponding to the main axis after processing of all main axes is completed.
[0079] The specific limitations of the three-dimensional support structure generation system based on simplified central axis transformation can be referred to the limitations of the three-dimensional support structure generation method based on simplified central axis transformation in the foregoing, which will not be repeated here. Each module in the three-dimensional support structure generation system based on simplified central axis transformation can be realized by software, hardware, or a combination thereof, in whole or in part. Each module can be embedded in or independent of a processor in a computer device in hardware format, or stored in a memory in a computer device in software format, so that the processor can call the operations of each module.
[0080] It should be noted that, in order to highlight the innovative part of the present application, modules not closely related to solving the technical problems proposed in the present application are not introduced in this embodiment, but this does not mean that there are no other modules in this embodiment.
[0081] In summary, the three-dimensional support structure generation method and system based on simplified central axis transformation disclosed in the present application is based on the central axis transformation expression of a three-dimensional model, automatically performs different degrees of simplification, and generates a corresponding three-dimensional wireframe support structure by finding a specific position and adding a slice on the basis of the simplification result, thereby realizing the simplification of the model and the expression mode satisfying the physical law, and being beneficial to subsequent three-dimensional printing and other practical applications of the model. Therefore, the present application effectively overcomes various shortcomings in the prior art and has a high industrial utilization value.
[0082] The above embodiments only exemplarily illustrate the principles and effects of the present application, and are not used to limit the present application. Any person skilled in the art can modify or change the above embodiments without departing from the spirit and scope of the present application. Therefore, all equivalent modifications or changes completed by those skilled in the art without departing from the spirit and technical thought disclosed in the present application should be covered by the claims of the present application.
Claims
1. A method for generating a three-dimensional support structure based on a simplified medial axis transform, characterized by, The method comprises: acquiring a three-dimensional model to be processed; performing a center axis transformation and simplification processing on the three-dimensional model to obtain a plurality of center points, and constructing a plurality of branch structures according to the radius variation amplitudes between the plurality of center points; selecting a branch structure with the largest average radius from the plurality of branch structures, setting the branch structure as a center axis, and forming different main axes according to different connection modes of the center axis and each of the remaining branch structures; processing each main axis: identifying special center points of the main axis from the center points corresponding to the main axis, and slicing the three-dimensional model on the main axis according to a corresponding slicing algorithm based on the special center points and a preset total amount of slices to obtain a plurality of slices; wherein a slice is a cross section of the three-dimensional model passing through the main axis; generating a three-dimensional support structure corresponding to the three-dimensional model based on each slice, and comparing the generated structure with the three-dimensional model to obtain a reconstruction error; after all main axes are processed, generating a final three-dimensional support structure of the three-dimensional model according to the main axis with the smallest reconstruction error and all slices corresponding thereto.
2. The method of claim 1, wherein, The center axis transformation and simplification processing on the three-dimensional model to obtain a plurality of center points, and constructing a plurality of branch structures according to the radius variation amplitudes between the plurality of center points comprises: performing a center axis transformation and simplification processing on the three-dimensional model to obtain a plurality of center points; for each center point: calculating the variation amplitude of the corresponding radius between the center point and its adjacent center points, and connecting the center points with a radius variation amplitude less than a preset amplitude threshold to obtain the branch structure corresponding to the connected center points.
3. The method of claim 1, wherein, The selection of a branch structure with the largest average radius from the plurality of branch structures, the setting of the branch structure as a center axis, and the formation of different main axes according to different connection modes of the center axis and each of the remaining branch structures comprise: for each branch structure: calculating the average radius of the center points of the branch structure according to the radii of all the center points included in the branch structure; selecting a branch structure with the largest average radius of the center points from all the branch structures as a center axis; connecting the center axis with each of the remaining branch structures to form a main axis composed of the center axis and different branch structures.
4. The method of claim 1, wherein, The identification of special center points of the main axis from the center points corresponding to the main axis comprises: selecting, from the center points on the main axis, center points having a connection relationship with center points located outside the main axis as branch center points; deleting each branch center point from the center points on the main axis, and selecting, from the center points still located on the main axis after the deletion, a center point with the largest radius as a largest-radius center point in each region according to a preset region of the three-dimensional model; wherein the radius is the distance from the corresponding center point to the three-dimensional model; taking each branch center point and each largest-radius center point in a region as a special center point of the main axis.
5. The method of claim 4, wherein, The slicing of the three-dimensional model on the main axis according to a corresponding slicing algorithm based on the special center points and a preset total amount of slices to obtain a plurality of slices comprises: Based on the main shaft, slice the three-dimensional model for each branch center point to obtain a special slice corresponding to each branch center point; Based on the main shaft, slice the three-dimensional model for each region maximum radius center point to obtain a special slice corresponding to each region maximum radius center point; According to the total number of slices, the number of special slices corresponding to the branch center points and the region maximum radius center points, calculate the number of additional slices; From a preset slice algorithm library, sequentially select one of the slice algorithms, slice the three-dimensional model on the main shaft according to the number of additional slices to obtain a plurality of additional slices, and use the plurality of special slices and the plurality of additional slices as all slices corresponding to the main shaft.
6. The method of claim 1, wherein, The three-dimensional support structure corresponding to the three-dimensional model is generated based on each slice, and the generated structure is compared with the three-dimensional model to obtain a reconstruction error, including: According to the main shaft and all slices corresponding to the main shaft, an initial three-dimensional support structure is generated; The generated initial three-dimensional support structure is compared with the three-dimensional model to obtain a reconstruction error corresponding to the main shaft.
7. A three-dimensional support structure generation system based on a simplified medial axis transform, characterized by, The system includes: A sample acquisition module for acquiring a three-dimensional model to be processed; A central axis transformation module for performing central axis transformation and simplification processing on the three-dimensional model to obtain a plurality of center points, and constructing a plurality of branch structures according to the radius change amplitudes between the plurality of center points; A main shaft screening module for screening a branch structure with the largest average radius from the branch structures, setting it as a central axis, and forming different main shafts according to different connection modes of the central axis and each of the remaining branch structures; A reconstruction module for processing each main shaft: Identify a special center point of the main shaft from the center points corresponding to the main shaft, and slice the three-dimensional model on the main shaft according to a corresponding slice algorithm based on the special center point and a preset total number of slices to obtain a plurality of slices; wherein a slice is a cross section of the three-dimensional model through the main shaft; Based on each slice, a three-dimensional support structure corresponding to the three-dimensional model is generated, and the generated structure is compared with the three-dimensional model to obtain a reconstruction error; A support structure generation module for generating a final three-dimensional support structure of the three-dimensional model according to the main shaft with the smallest reconstruction error and all slices corresponding thereto after all main shafts are processed.