A fast analysis method for the clearability of supporting structures in any direction

Abstract

The present invention discloses a method for rapid analysis of the clearability of a support structure in any direction. The steps of the present invention are as follows: Step 1. Voxelize a three-dimensional model for rapid judgment of the accessibility of separation points; Step 2. Rapidly generate a voxel path II approximating a straight line segment based on a three-dimensional Bresenham algorithm; Step 3. Parallel and efficiently detect the accessibility of all separation points of the three-dimensional model. The present invention solves the problem of limited and time-consuming clearing directions of support structures in 3D printed products, and supports rapid analysis of the clearability of support structures along any direction and optimized calculation of printing directions. The present invention extends the Bresenham algorithm to three-dimensional space, and can generate a voxel path approximating a straight line segment in any direction. The three-dimensional Bresenham algorithm is an integer traversal algorithm with a refined judgment formula. Not only is the amount of calculation small, but there is no cumulative error, which speeds up the search efficiency of the voxel path in step 1. Therefore, it is possible to quickly support the clearability analysis of support structures at any angle.

Description

Technical Field

[0001] The present invention relates to the field of 3D printing, and in particular to a method for quickly analyzing the removability of a support structure in any direction. Background Art

[0002] 3D printing has attracted widespread attention from industry and academia for its unique advantages such as supporting free design, personalized customization, and energy saving. When manufacturing complex geometric models, 3D printing often requires the addition of support structures to prevent local collapse, curling, and other phenomena in the printed results. At the same time, in order to ensure that the product can be manufactured effectively, the 3D printed results usually need to remove the support structure. Considering that the support structure is usually difficult to dissolve directly and needs to be removed manually, in order to avoid the failure of the printed results due to the inability to remove the support structure, thereby affecting the iterative design cycle of the product and wasting resources, it is usually necessary to perform a support structure removability analysis on the three-dimensional digital model before printing. However, the difficulty of manually judging whether the support structure can be removed after the three-dimensional digital model is manufactured increases with the increase of the geometric complexity of the three-dimensional model, and the effectiveness is difficult to guarantee.

[0003] At present, the typical automated methods for performing support structure cleanability analysis before 3D model printing include the following: (1) Octree neighbor search method, which uses the octree method to divide the 3D model bounding box and determines the accessibility of the support structure by judging whether the small cube containing the support structure can reach the boundary without obstacles in the axial direction; (2) Calculating all the cleanable directions of each support structure in C-Space, which requires a lot of preprocessing calculations, and solving the TSP problem on the search graph to obtain the lowest cost support structure removal order. Existing methods are usually very time-consuming and difficult to support cleanability analysis of support structures at any angle, and have a narrow scope of application.

[0004] In view of the above situation, the present invention proposes a method for quickly analyzing the cleanability of a support structure in any direction, which can quickly support the cleanability analysis of the support structure at any angle, and further effectively support the optimization calculation of the printing direction. Summary of the invention

[0005] The main purpose of the present invention is to provide a method for quickly analyzing the removability of a support structure in any direction, so as to solve the problem in the background art that it is difficult to quickly perform removability analysis of a support structure along any angle.

[0006] Related work analyzing the removability of support structures shows that: (1) the contact part between the support structure and the 3D model surface is called a separation point. By disconnecting the separation point, the support structure can be detached from the 3D model; (2) separation points that are accessible (visible) from the outside along the line of sight can be disconnected by default. In other words, the support structure associated with the separation point that can be accessed from the outside along a certain direction is considered to be removable.

[0007] The technical solution adopted by the present invention mainly comprises the following steps:

[0008] Step 1: voxelize the 3D model to quickly determine the accessibility of the separation points;

[0009] In order to avoid the time-consuming intersection calculation of straight lines and three-dimensional models, it is possible to quickly determine whether the separation point can be accessed along a given line of sight direction v. First, the three-dimensional model is voxelized; then, in the voxelized discrete space, a voxel path I approximating the v direction is found, and the voxels in the voxel path I are adjacent to each other; finally, by determining whether the voxel path I contains the internal voxels of the model, it is determined whether the separation point along the v direction can be accessed;

[0010] Step 2: quickly generate a voxel path approximating a straight line segment based on the three-dimensional Bresenham algorithm II;

[0011] The Bresenham algorithm is introduced into the voxel set of the three-dimensional model to find a three-dimensional voxel path II that approximates the v direction, avoiding floating point calculations and accelerating the efficiency of path finding, that is, accelerating the efficiency of analyzing the accessibility of separation points; the Bresenham algorithm is used to find a pixel set that approximates a straight line segment in a two-dimensional pixel space;

[0012] Step 3: Parallel and efficient detection of the accessibility of all separation points of the 3D model

[0013] Considering that the accessibility analysis of separation points is independent of each other, the accessibility analysis of all separation points associated with the support structures on the surface of the 3D model is performed in parallel, thereby further improving the efficiency of the cleanability analysis of the support structures of the 3D model. If a separation point is still inaccessible after traversing all the bounding box surface voxels, the support structure associated with the inaccessible separation point will be judged as being unable to be cleaned.

[0014] The beneficial effects of the present invention are as follows:

[0015] The present invention extends the Bresenham algorithm (originally an algorithm for approximating straight line segments through pixel points in a two-dimensional plane) to three-dimensional space, and can generate voxel paths that approximate straight line segments in any direction. The three-dimensional Bresenham algorithm is an integer traversal algorithm with a refined judgment formula. It not only has a small amount of calculation and no cumulative error, but also speeds up the search efficiency of the voxel path in step 1. Therefore, it can quickly support the clearability analysis of the support structure at any angle.

[0016] By performing accessible analysis on all separation points on the surface of the three-dimensional model, it is possible to effectively support the optimization calculation of the printing direction. The present invention solves the problem of limited and time-consuming support structure removal directions in 3D printed products, and supports rapid analysis of the removability of support structures in any direction and optimization of printing direction calculation. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flow chart of the present invention.

[0018] Figure 2 The input STL 3D model image.

[0019] Figure 3 This is the voxelization effect of the three-dimensional model.

[0020] Figure 4 This is the voxel classification effect of the three-dimensional model.

[0021] Figure 5 Display of separation points on the surface of a 3D model.

[0022] Figure 6 Display of accessible directions for separated point voxels.

[0023] Figure 7 Display of accessible directions for multiple discrete voxels. DETAILED DESCRIPTION

[0024] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0025] like Figure 1 As shown in the figure, the rapid analysis method of the clearability of the support structure in any direction mainly includes the following steps:

[0026] Step 1: Voxelize the 3D model to quickly determine the accessibility of the separation points

[0027] 1-1. Voxelize the 3D model in STL format, such as Figure 2 As shown, the AABB (i.e. axial) bounding box of the three-dimensional model relative to the world coordinate system is calculated, and then the bounding box is divided into a set of cube voxels of the same size according to the required resolution. Figure 3As shown, the voxels of the entire bounding box are obtained. The higher the resolution, the smaller the voxels obtained. The side length of the voxel is set to the diameter of the cleaning tool. The size of the bounding box of the 3D model in this example is 75*75*28 (unit: mm), and the side length of the voxel is set to 1mm. According to the positional relationship of the eight vertices of the voxel relative to the 3D model, all voxels in the voxel set can be classified into four categories:

[0028] (1.1) Voxels inside the 3D model: The eight vertices of the voxel are completely inside the 3D model, such as Figure 4 Medium and light grey voxels are represented.

[0029] (1.2) 3D model boundary voxel: Some of the eight vertices of the voxel are located inside the 3D model (no more than 2) and the rest are located outside the 3D model, such as Figure 4 Medium and dark grey voxels are indicated.

[0030] (1.3) Bounding box boundary voxel: Four of the eight vertices of the voxel coincide with the bounding box, such as Figure 4 The black voxels are indicated.

[0031] (1.4) Empty voxels: Voxels other than the voxels inside the 3D model, the voxels on the boundary of the 3D model, and the voxels on the boundary of the bounding box, such as Figure 4 White voxels are indicated in the middle.

[0032] 1-2. According to the existing support structure generation method, the positions of the separation points associated with all the support structures of the three-dimensional model are obtained, and the voxels containing the separation points are called separation point voxels. For example, the cylindrical support generation method uses a three-dimensional model in STL format as input. If the angle between the normal vector of the triangular face of the three-dimensional model and the printing direction is greater than 135°, then the face requires a support structure. This standard is applied to all triangular faces of the three-dimensional model to identify all areas that require support structures. After the printing direction is given, the positions of the separation points associated with all the support structures of the three-dimensional model can be calculated before printing. Figure 5 As shown, assuming that the printing direction is the Z axis, the red point represents the separation point. In particular, when a surface voxel of the three-dimensional model contains a separation point, the voxel is called a separation point voxel.

[0033] 1-3. Starting from the center of the separation point voxel and connecting it with the center of any bounding box boundary voxel to form a straight line segment, as long as a straight line segment can be found so that the voxel path approaching the straight line segment does not contain any voxels inside the three-dimensional model, then the separation point voxel is considered to be accessible from the outside, that is, accessed along the straight line where the straight line segment is located.

[0034] Step 2: Quickly generate voxel paths that approximate straight line segments based on the 3D Bresenham algorithm

[0035] In order to more quickly determine whether the separation point voxel is accessed from the outside along the straight line segment connected to the bounding box boundary voxel, the present invention uses the three-dimensional Bresenham algorithm along the XYZ axis direction to perform axial accumulation from the separation point voxel to the bounding box boundary voxel (the side length of the voxel is the accumulation unit) to efficiently and quickly find a voxel path that is as close to the given straight line segment as possible. Specifically:

[0036] 2-1. Assume that the separation point voxel D to be detected vox The center coordinates of are (x1, y1, z1), that is, the distances from the center of the voxel to the origin of the world coordinate system are the side lengths of the voxels x1, y1, and z1, respectively. vox The center of the current bounding box is the voxel B vox The centers are connected, and the bounding box boundary voxel B vox The center coordinates of are (x2, y2, z2), let dx = |x2-x1|, dy = |y2-y1|, dz = |z2-z1|; when dx is greater than dy and dz at the same time, accumulate traversal along the X axis; when dy is greater than dx and dz at the same time, accumulate traversal along the Y axis; when dz is greater than dy and dx at the same time, accumulate traversal along the Z axis; separate point voxel D vox The cumulative traversal along the X, Y, and Z axes is done in the same way;

[0037] 2-2. Voxel accumulation traversal can be carried out along three axes (X axis, Y axis, Z axis). When dx is greater than dy and dz at the same time, the separation point voxel D vox When traversing along the X axis, the separation point voxel D vox Accumulate 1 unit of voxel side length along the X axis to obtain a new voxel; use formula (1) to determine the Y coordinate of the new voxel relative to the separation point voxel D vox Whether the Y coordinate changes; Substitute the result obtained from formula (1) into formula (2), and continue to recursively infer the separation point voxel D vox Whether the Y coordinates of the subsequent n+1 new voxels change relative to the Y coordinate of the nth new voxel; the recursive formula is as follows:

[0038] P xy (1) = y1-V l +(V l -x1)*k xy (1)

[0039]

[0040] If the current voxel increases by one unit voxel side length along the X-axis direction to obtain a new voxel, P xy Indicates whether the new voxel Y coordinate has changed relative to the current voxel Y coordinate; P xy ≥0 means the Y coordinate plus the length of a unit voxel side, Pxy <0 means the Y coordinate remains unchanged; n represents the number of voxels increasing along the X-axis; V l The length of the voxel increment, the default is 1;

[0041] Similarly, judging whether the Z coordinate of the current voxel has changed is as follows:

[0042] P xz (1) = z1-V l +(V l -x1)*k xz (3)

[0043]

[0044] If the current voxel increases by one unit voxel side length along the X-axis direction to obtain a new voxel, P xz Indicates whether the new voxel Z coordinate has changed relative to the current voxel Z coordinate; P xz ≥0 means the Z coordinate plus the length of one unit voxel side, P xz <0 means the Z coordinate remains unchanged; n represents the number of voxels increasing along the X-axis.

[0045] The cumulative traversal method process for the three axes (X axis, Y axis, and Z axis) is similar, and here we take the cumulative traversal along the X axis as an example.

[0046] From the above recursion, we can see that if the current voxel is V(Xi, Yi, Zi), the next voxel coordinate may be one of the four coordinates: V1(Xi+1, Yi+1, Zi), V2(Xi+1, Yi, Zi), V3(Xi+1, Yi+1, Zi+1), and V4(Xi+1, Yi, Zi+1).

[0047] (3) Figure 6 As shown in the figure, the plane diagram is a layered diagram of the XY surface of the three-dimensional model. The gray straight line segment represents D vox Center coordinates (x1, y1, z1) and B vox The straight line connecting the center coordinates (x2, y2, z2) and the medium-gray voxels close to the gray straight line segment represent the voxels from D vox Start accumulating and traversing along the X axis. The voxels obtained by recursion according to formulas (2) and (4) are not voxels inside the 3D model and are used to represent voxel D vox The accessible direction.

[0048] Step 3: Parallel and efficient detection of the accessibility of all separation points of the 3D model

[0049] like Figure 7As shown, in a multi-threaded manner, based on the method in step 2, steps 1-3 are performed simultaneously on all separated point voxels on the surface of the 3D model. Specifically:

[0050] 3-1. Taking the current separation point voxel as the center, traverse each bounding box boundary voxel in sequence, and test whether the line segment formed between the current separation point voxel and each bounding box boundary voxel and the corresponding voxel path meet the separation point accessibility requirement;

[0051] 3-2. If a separation point voxel is determined to be accessible when traversing the bounding box boundary voxels, the separation point voxel is immediately excluded from the subsequent retesting;

[0052] 3-3. If after traversing all bounding box boundary voxels, there is still a separation point that cannot be accessed, the support structure associated with the inaccessible separation point is determined to be unable to be cleaned.

Claims

1. A rapid analysis method for the removability of a support structure in any direction, characterized by The following steps are included: Step 1. Voxelize the 3D model to quickly determine the accessibility of the separation point; First, the three-dimensional model is voxelized; Then, in the voxelized discrete space, a voxel path I approaching the v direction is searched, and the voxels in the voxel path I are adjacent to each other; Finally, whether the separation point along the v direction can be accessed is determined by judging whether the voxel path I contains the internal voxels of the model; Step 2. Rapidly generate voxel paths approximating straight line segments based on the three-dimensional Bresenham algorithm II; The Bresenham algorithm is introduced into the voxel set of the 3D model to find a 3D voxel path approximating the v direction II, avoiding floating point calculations and accelerating the efficiency of path finding, that is, accelerating the analysis efficiency of the accessibility of the separation point; The Bresenham algorithm is used to find a pixel set that approximates a straight line segment in a two-dimensional pixel space; Step 3. Efficiently check the accessibility of all separated points of the 3D model in parallel The accessibility of separation points associated with all supporting structures on the surface of the 3D model is analyzed simultaneously in parallel. If a separation point is still inaccessible after traversing all voxels on the bounding box surface, the supporting structure associated with the inaccessible separation point will be judged as unable to be cleaned.

2. The method for rapid analysis of the removability of a support structure in any direction according to claim 1, characterized in that In step 1, a voxel path I close to the v direction is found to avoid time-consuming intersection calculation between the straight line and the three-dimensional model. The specific steps are as follows: 1-1. Voxelize the 3D model in STL format and calculate the AABB bounding box of the 3D model relative to the world coordinate system; then divide the bounding box into a set of cubic voxels of the same size according to the required spatial resolution, and divide the voxels in the voxel set into four categories: bounding box boundary voxels, 3D model boundary voxels, 3D model internal voxels, and empty voxels; 1-2. According to the existing support structure generation method, the positions of the separation points associated with all the support structures of the three-dimensional model are obtained, and the voxels containing the separation points are called separation point voxels; 1-3. Starting from the center of the separation point voxel and connecting it with the center of any bounding box boundary voxel to form a straight line segment, as long as a straight line segment can be found so that the voxel path approaching the straight line segment does not contain any voxels inside the three-dimensional model, then the separation point voxel is considered to be accessible from the outside, that is, accessed along the straight line where the straight line segment is located.

3. The method for rapid analysis of the removability of a support structure in any direction according to claim 2, characterized in that Whether a support structure is removable is determined by determining the accessibility of all associated separation points of the support structure.

4. The method for rapid analysis of the removability of a support structure in any direction according to claim 1, 2 or 3, characterized in that In step 2, a three-dimensional Bresenham algorithm is used to quickly generate a voxel path II that approximates a straight line segment. The specific steps are as follows: 2-1. Assume that the separation point voxel D vox The center coordinates of the bounding box are (x1, y1, z1), and the bounding box boundary voxel B vox The center coordinates of are (x2, y2, z2), let dx = |x2-x1|, dy = |y2-y1|, dz = |z2-z1|; when dx is greater than dy and dz at the same time, accumulate traversal along the X axis; when dy is greater than dx and dz at the same time, accumulate traversal along the Y axis; when dz is greater than dy and dx at the same time, accumulate traversal along the Z axis; separate point voxel D vox The cumulative traversal along the X, Y, and Z axes is done in the same way; 2-2. When dx is greater than dy and dz at the same time, the separation point voxel D vox When traversing along the X axis, the separation point voxel D vox Accumulate 1 unit of voxel side length along the X axis to obtain a new voxel; use formula (1) to determine the Y coordinate of the new voxel relative to the separation point voxel D vox Whether the Y coordinate changes; Substitute the result obtained from formula (1) into formula (2), and continue to recursively infer the separation point voxel D vox Whether the Y coordinates of the subsequent n+1 new voxels change relative to the Y coordinate of the nth new voxel; the recursive formula is as follows: P xy (1)=y1-V l +(V l -x1)*k xy (1) If the current voxel increases by one unit voxel side length along the X-axis direction to obtain a new voxel, P xy Indicates whether the new voxel Y coordinate has changed relative to the current voxel Y coordinate; P xy ≥0 means the Y coordinate plus the length of a unit voxel, P xy <0 means the Y coordinate remains unchanged; n represents the number of voxels increasing along the X-axis; V l The length of the voxel increment, the default is 1; Similarly, judging whether the Z coordinate of the current voxel has changed is as follows: P xz (1)=z1-V l +(V l -x1)*k xz (3) If the current voxel increases by one unit voxel side length along the X-axis direction to obtain a new voxel, P xz Indicates whether the new voxel Z coordinate has changed relative to the current voxel Z coordinate; P xz ≥0 means the Z coordinate plus the length of one unit voxel side, P xz <0 means the Z coordinate remains unchanged; n represents the number of voxels increasing along the X-axis.

5. The method for rapid analysis of the removability of a support structure in any direction according to claim 4, characterized in that In a multi-threaded manner, all separated point voxels on the surface of the 3D model are processed simultaneously, and steps 1-3 are implemented as follows: 1-3-1. Taking the current separation point voxel as the center, traverse each bounding box boundary voxel in sequence, and test whether the line segment formed between the current separation point voxel and each bounding box boundary voxel and the corresponding voxel path meets the separation point accessibility requirement; 1-3-2. If a separation point voxel is determined to be accessible when traversing the bounding box boundary voxels, the separation point voxel is immediately excluded from the ranks that need to be tested later; 1-3-3. If after traversing all bounding box boundary voxels, there is still a separation point that cannot be accessed, the support structure associated with the inaccessible separation point is determined to be unable to be cleaned.

6. The method for rapid analysis of the removability of a support structure in any direction according to claim 5, characterized in that By efficiently and in parallel testing the accessibility of all separation points of the 3D model, rapid analysis of the removability of support structures in any printing direction can be achieved.

Images