3D model object of the physical prototype of the product

Abstract

This invention particularly relates to a computer-implemented method for a 3D modeling object used to design a physical prototype of a product. The 3D modeling object includes a wireframe based on at least one characteristic line. The method includes calculating the segmentation of a provided mesh, thus obtaining at least two regions and at least one boundary polyline between the at least two regions. The method then includes transforming each of the at least one boundary polyline into at least one characteristic line. The method further includes calculating a network of at least one characteristic line that forms the wireframe of the 3D modeling object.

Description

Technical Field

[0001] This invention relates to the field of computer programs and systems, and more specifically, to a method, system, and program for designing a 3D modeling object of a physical prototype of a product, the 3D modeling object comprising a wireframe based on at least one characteristic line. Background Technology

[0002] The market offers numerous systems and programs for the design, engineering, and manufacturing of objects. CAD is an acronym for Computer-Aided Design, which, for example, refers to software solutions for designing objects. CAE is an acronym for Computer-Aided Engineering, which, for example, refers to software solutions for simulating the physical behavior of future products. CAM is an acronym for Computer-Aided Manufacturing, which, for example, refers to software solutions for defining manufacturing processes and operations. In such computer-aided design systems, graphical user interfaces play a crucial role in technical efficiency. These technologies may be embedded in Product Lifecycle Management (PLM) systems. PLM refers to a business strategy that helps companies share product data, apply common processes, and leverage corporate knowledge to extend product development across the enterprise's concept from concept to the end of the product lifecycle. Dassault Systèmes (traded as CATIA, ENOVIA, and DELMIA) offers PLM solutions that provide an engineering center for organizing product engineering knowledge, a manufacturing center for managing manufacturing engineering knowledge, and an enterprise center that integrates and connects the engineering and manufacturing centers. The entire system provides an open object model that connects products, processes, and resources, enabling dynamic, knowledge-based product creation and decision support, thereby driving optimized product definition, manufacturing preparation, production, and service.

[0003] In this context, product design is crucial. Product design aims to create a virtual 3D model that is as close as possible to the physical prototype, which the designer considers the master model. This approach is commonly used in industries such as aerospace, automotive, marine, consumer goods, and consumer packaging for free-form prototypes (physical entity models). In some industries, companies focus on the quality of characteristic lines during reverse engineering. These companies are looking for ways to make their products quieter and more stable.

[0004] Figure 1 and Figure 2 A screenshot showing the simulation results on the prototype illustrates the impact of characteristic line quality on the car's aerodynamics. Figure 1 The first aerodynamic simulation results of the original design are shown. Figure 2 The second aerodynamic simulation results are shown after modifying the characteristic lines of the original design. The simulation results are different.

[0005] Similarly, Figure 3and Figure 4 A screenshot showing the simulation results on the prototype illustrates the impact of characteristic line quality on the noise generated by the vehicle body. Figure 4 The simulation results are relative to Figure 3 The simulation results were improved because the acoustic designer's engineers modified one or more characteristic lines.

[0006] The prototype needs to be digitally reconstructed. Once the prototype's results are satisfactory, the digital reconstruction process (also known as the "network method") typically involves the following manual steps:

[0007] - The prototype was created by designers while adhering to aerodynamic and noise standards;

[0008] - Use 3D scanning technology to measure the prototype to obtain the relevant point cloud;

[0009] - Clean up, repair, and optimize the variable-quality mesh derived from prototype digitization;

[0010] - Analyze the mesh to highlight its shape and characteristic lines. Characteristic lines are the primary paths that surface boundaries must follow. The purpose of this analysis is to replicate the surface of the product.

[0011] - Users create a set of curves based on prior analysis, following the characteristic lines. These curves represent the product's wireframe and define future surface boundaries; and

[0012] - Each cell of the network is filled by a surface that conforms to the maximum deviation from the mesh.

[0013] In these steps, the routine process of generating characteristic lines is performed manually by the user. This can be done based on their expertise and understanding of the object's shape, or as a visual guide using the results of the mesh analysis step (color plot). The user draws the character curves by manipulating the mesh to insert control points. The final curve must reproduce the shape of the mesh and adhere to the maximum deviation from it. Curvature analysis plots help the user distinguish the shape of the mesh. Deviation analysis tools are provided to help them verify or edit the curves to best fit the mesh.

[0014] The quality of the curve directly affects the quality of the final surface. Moving the polyline extracted from the graph to a smooth curve that fits the mesh is a crucial step in this process. Currently, the decisions regarding the positioning of curve control points, the insertion of new curve control points, and the density until a good fit (while maintaining a smooth shape) is achieved depend on the user's expertise. This is a process that cannot be replicated.

[0015] Therefore, the mainstream process for generating curves is manual and time-consuming. Clearly, this is the most time-consuming step in the entire reconstruction workflow, accounting for approximately 70% of the total reconstruction time. Any auxiliary generation here will significantly improve productivity. The higher the quality of the curves (the car's outline) the user requires, the more complex the editing process becomes.

[0016] For shape-dominated products (e.g., car bodies), users must prioritize the quality of curves. Curves must be taut and not wobbly. Conversely, for mesh-dominated products (e.g., car interiors), curves must fit the mesh, and the number of control points can be more critical. The optimal trade-off between smooth curves and mesh-fitting curves depends on the user's expertise. This means that two users will not generate the same curve network or the same surface. Such high user variability is undesirable for a company. Manual processes are very tricky, and numerous user errors can affect the quality of the final surface. For example, multiple control points are often used to generate simple curves, and the view's position can mask certain mesh variations. Therefore, automating this process offers significant advantages in terms of productivity, quality, and repeatability.

[0017] Preliminary attempts have been made to automatically obtain characteristic lines for scanned products. However, the following drawbacks exist. First, curve generation is based on high curvature maps. This is a fast method, but not suitable for smooth characteristic lines or smooth surface variations. This method must be used in conjunction with an editing process to obtain smooth edges and manually supplement anything not provided by the system. High curvature is not the only criterion. Second, the generated curves are lines with high curvature in each region. Again, this is acceptable for high curvature areas, but unacceptable for smooth areas where it is preferable to obtain the boundaries of that area. This aspect has an even greater impact if surfaces are generated directly from curves. Third, the quality of the generated curves is poor. It cannot be used as surface boundaries, but only for dividing shapes and giving contour lines a preferred direction. Designers want their products to have a teaching character. Using this method, the generated lines begin to wobble. Each curve and surface must be manually edited to correspond to the fatigued design. Fourth, the surface generation method minimizes the deviation between the surface and the mesh, but does not give the user complete control over surface patching. In practice, each mesh unit is divided into smaller blocks, and the surface boundaries do not match the curve mesh. Fifth, the process is not a function of the kind of precision that users need.

[0018] Manually generating line curves (such as feature lines) is a repetitive and tedious task. The quality of the final curve is highly dependent on the user's expertise, as the user's perception depends on the view orientation on the grid derived from the prototype digitization.

[0019] Last but not least, noise introduced by the equipment used to compute the relevant point cloud from the prototype can also cause differences between the surface obtained as a digitization result and the surface of the prototype. The mesh obtained from the point cloud exhibits these differences relative to the prototype. Therefore, the surface obtained from the mesh must be as close to the mesh as possible so that the digital reconstruction adheres to tolerances relative to the prototype.

[0020] In this context, there is still a need for an improved approach to 3D modeling objects for designing physical prototypes of products, where the 3D modeling objects comprise wireframes based on at least one characteristic line. Summary of the Invention

[0021] Therefore, a computer-based method is provided for implementing a 3D modeling object for designing a physical prototype of a product, the 3D modeling object comprising a wireframe based on at least one characteristic line. The method includes: - providing a mesh for the 3D modeling object;

[0022] - Calculate the division of the provided mesh to obtain at least two regions and at least one boundary polyline between the at least two regions;

[0023] - Transform each of at least one boundary polyline into at least one characteristic line; and

[0024] - Calculate a network of at least one characteristic line, the network of at least one characteristic line forming the wireframe of a 3D modeling object.

[0025] The method may include one or more of the following:

[0026] - The computational segmentation includes: computing a first segmentation to obtain: at least two first regions in the provided mesh; and at least one first boundary polyline between the at least two distinct regions; - computing a second segmentation performed at a higher optimization subdivision level than the first segmentation to obtain: at least two second regions in the provided mesh, which belong to at least one of the at least two first regions; and at least one second boundary polyline between the at least two second regions;

[0027] The computational network includes selecting at least one first boundary polyline; connecting at least one second boundary polyline to the selected at least one first boundary polyline by performing a snapping operation; and facilitating at least one first boundary polyline emanating from the first segment;

[0028] - Calculating the segmentation of the provided mesh includes: detecting at least one principal boundary polyline of the provided mesh, which is a polyline with higher curvature compared to other polylines of the provided mesh;

[0029] - The detection of the at least one principal boundary polyline includes: calculating a contrast map of the grid by applying image filtering to the grid; - calculating curvature evolution and identifying the extreme values ​​of the curvature evolution, and calculating at least one principal boundary polyline by linking line segments generated by previous calculations;

[0030] The calculation of the segmentation of the provided mesh also includes fine segmentation by calculating the normal clustering on each region of the mesh.

[0031] - After calculating the segmentation of the provided mesh: select one of at least two regions and split the selected region by using the curvature map of the provided mesh, and / or select at least two regions and merge them into one region;

[0032] - The transformation includes calculating a smooth curve for each of the at least one boundary polyline;

[0033] - Place control points for the smooth curve, where placement is performed to minimize: the deviation relative to the smooth curve and the provided grid, and / or the number of control points, and / or the degree of smoothness of the curve;

[0034] - Calculate surface components based on continuity constraints of wireframe and 3D modeled objects.

[0035] A computer program is also provided, which includes instructions for performing the method.

[0036] A computer-readable storage medium on which a computer program is recorded is also provided.

[0037] A system is also provided that includes a processor coupled to a memory and a graphical user interface, wherein a computer program is recorded on the memory. Attached Figure Description

[0038] Embodiments of the invention will now be described by way of non-limiting examples and with reference to the accompanying drawings, wherein:

[0039] - Figure 1 and Figure 2 An example is shown illustrating the effect of characteristic lines on the aerodynamics of a car;

[0040] - Figure 3 and Figure 4 An example is shown showing the effect of characteristic lines on noise caused by the car body;

[0041] - Figure 5 A flowchart illustrating an example of this method is shown;

[0042] - Figure 6 A flowchart illustrating an example of step S20 of the method is shown;

[0043] - Figure 7 An example of a clean mesh obtained from a physical prototype is shown;

[0044] - Figure 8 An example of the quality of a mesh obtained from a physical prototype is shown;

[0045] - Figure 9 An example of voxel-based normal clustering is shown;

[0046] - Figures 10 to 12 Examples of clustering based on low, medium, and high clustering levels are shown;

[0047] - Figures 13 to 17 An example of feature lines extracted from the prototype's mesh is shown;

[0048] - Figure 18 This is an example of initializing a segmentation graph;

[0049] - Figure 19 This is an example of the final segmentation image;

[0050] - Figure 20 An example of automatic curve generation is shown;

[0051] - Figure 21 This is an example of an automatically generated curve needle;

[0052] - Figure 22 An example of a wireframe without a master curve is shown;

[0053] - Figure 23 It shows the principal curve Figure 22 Example of a wireframe;

[0054] - Figure 24 An example of the system's graphical user interface is shown; and

[0055] - Figure 25 An example of the system is shown. Detailed Implementation

[0056] refer to Figure 5The flowchart presents a computer-implemented method for designing a 3D modeling object for a physical prototype of a product, the 3D modeling object being designed to include a wireframe based on at least one characteristic line. The method includes providing a mesh for the 3D modeling object. The mesh can be obtained from the physical prototype, for example, by acquiring a relevant point cloud of the physical prototype using 3D scanning technology. The method also includes calculating a segmentation of the provided mesh. The segmentation generates at least two distinct regions from the provided mesh. The segmentation of the 3D mesh of the prototype also includes at least one boundary polyline located between at least two distinct regions; at least a portion of the boundary between the at least two regions provides the at least one boundary polyline. The method then includes transforming each of the at least one boundary polyline into at least one characteristic line. The characteristic line is one of the lines characterizing the physical prototype, and therefore the 3D modeling object to be designed. Next, the method includes calculating a network of at least one characteristic line. The network of at least one characteristic line forms the wireframe of the 3D modeling object. Thus, the obtained wireframe is composed of the main lines (characteristic lines) of the product to be designed.

[0057] This method improves the design of 3D modeling objects for physical prototypes of products, where the 3D modeling objects include wireframes based on at least one characteristic line. The method can automatically generate curves that form the wireframes based on mesh analysis.

[0058] Notably, good curve quality can be achieved without manipulating the view or requiring user interaction during curve creation. Wireframe quality is improved, leading to increased productivity. Furthermore, the process of designing 3D modeled objects is repeatable. This method creates object wireframes based on characteristic lines by creating and initializing segmentation maps. A combination of image filtering and / or normal clustering and / or curvature analysis can be used.

[0059] The resulting diagram of this invention corresponds to the main line (wireframe) of the object. Users can use it to obtain the main curves and generate surfaces placed on this network. Additionally, if the cloud quality is not too poor, this method generates most curves. Users can complete the process with a few edits, and most lines have been calculated using this method. Furthermore, the process can be specialized and can meet accuracy requirements; for example, the surfaces obtained from the wireframe can be of class A, B, or C.

[0060] Interestingly, the quality of the resulting curve does not depend on the user's viewing direction. The projection onto the mesh depends on the normal direction of the cluster, and undesirable variations are constrained by system rules: undesirable variations can be computed to achieve a best fit to the skeleton using as few control points as possible. Reducing the number of control points improves the quality of the resulting surface, thereby limiting noise phenomena caused by complex curves.

[0061] Furthermore, this process is optimally repeatable for users. Two designers working on the same project will achieve similar results.

[0062] Other advantages of the invention will be discussed below.

[0063] The method is computer-implemented. This means that the steps (or essentially all steps) of the method are executed by at least one computer or any similar system. Therefore, the execution of the method's steps by a computer may be fully automatic or semi-automatic. In the example, at least some steps of the method may be triggered through user-computer interaction. The required level of user-computer interaction may depend on the anticipated level of automation and be balanced with the need to fulfill the user's intentions. In the example, this level may be user-defined and / or predefined.

[0064] For example, as already discussed, users can complete or correct the generated curves, or even create curves manually.

[0065] A typical example of a computer implementation of the method is to execute it using a system suitable for this purpose. This system may include a processor coupled to memory and a graphical user interface (GUI) on which a computer program is recorded, containing instructions for performing the method. The memory may also store a database. The memory is any hardware suitable for such storage and may comprise several physically distinct sections (e.g., one for the program and possibly another for the database).

[0066] This method typically manipulates modeling objects. A modeling object is any object defined by data, for example, stored in a database. By extension, the expression "modeling object" refers to the data itself. Depending on the type of system, modeling objects can be defined by different kinds of data. This system can actually be any combination of CAD, CAE, CAM, PDM, and / or PLM systems. In those different systems, modeling objects are defined by the corresponding data. Therefore, one can refer to CAD objects, PLM objects, PDM objects, CAE objects, CAM objects, CAD data, PLM data, PDM data, CAM data, and CAE data. However, these systems are not mutually exclusive, as modeling objects can be defined by data corresponding to any combination of these systems. Therefore, it will be apparent from the definition of such a system provided below that a system can very well be both a CAD system and a PLM system.

[0067] The term "CAD system" also implies any system suitable for designing modeling objects based on their graphical representations, such as CATIA. In this context, the data defining the modeling object includes the data that allows the modeling object to be represented. A CAD system can provide a representation of a CAD modeling object, for example, using edges or lines (and in some cases, faces or surfaces). Lines, edges, or surfaces can be represented in various ways, such as non-uniform rational B-splines (NURBS). Specifically, a CAD file contains specifications from which geometry can be generated, thus allowing the generation of representations. The specifications of the modeling object can be stored in a single CAD file or multiple CAD files. The typical size of a file representing a modeling object in a CAD system is in the range of 1 MB per part. A modeling object can typically be an assembly of thousands of parts.

[0068] In the context of CAD, modeling objects can typically be 3D modeling objects, such as representations of products, like parts or assemblies of parts, or possibly components of a product. A "3D modeling object" refers to any object modeled from data that allows for its 3D representation. 3D representation allows parts to be viewed from various angles. For example, when represented in 3D, 3D modeling objects can be manipulated and rotated around any of their axes or any axis on the screen displaying the representation. Specifically, this does not include 2D icons that are not 3D modeled. The display of 3D representations aids in design (i.e., increases the statistical speed at which designers complete tasks). Since product design is part of the manufacturing process, it can accelerate the manufacturing process in the industry.

[0069] 3D modeling objects can represent the geometry of a product to be manufactured in the real world after its virtual design is completed using, for example, CAD software solutions or CAD systems. Examples include (e.g., mechanical) parts or component assemblies (or equivalent components of parts, since from a methodological perspective, a component of a part can be viewed as the part itself, or the method can be applied independently to each part of the assembly) or more generally, any rigid body assembly (e.g., a moving mechanism). CAD software solutions allow for the design of products in a wide range of virtually unlimited industrial sectors, including: aerospace, architecture, construction, consumer goods, high-tech equipment, industrial equipment, transportation, marine and / or offshore oil and gas production or transportation. Therefore, the 3D modeling object designed using this method can represent an industrial product, which can be any mechanical part, such as a part of a land vehicle (including automobile and light truck equipment, racing cars, motorcycles, truck and motor vehicle equipment, trucks and buses, trains), a part of an aircraft (including fuselage equipment, aerospace equipment, propulsion equipment, defense products, aviation equipment, space equipment), a part of a naval vehicle (including naval equipment, commercial ships, marine equipment, yachts and workboats, marine equipment), a part of a general mechanical part (including industrial manufacturing machinery, heavy mobile machinery or equipment, installation equipment, industrial equipment products, metal products, tire products), electromechanical or electronic parts (including consumer electronics products, safety and / or control and / or instrumentation products, computing and communication equipment, semiconductors, medical devices and equipment), consumer goods (including furniture, home and garden products, leisure products, fashion products, products of hard goods retailers, products of soft goods retailers), and packaging (including food and beverage and tobacco, beauty and personal care, and household product packaging).

[0070] A PLM system also refers to any system suitable for managing modeled objects representing physically manufactured products (or products to be manufactured). Therefore, in a PLM system, modeled objects are defined by data suitable for manufacturing physical objects. These can typically be dimensional values ​​and / or tolerance values. Having such values ​​is indeed better for correctly manufacturing objects.

[0071] CAM solutions also refer to any solution, hardware, or software applicable to managing product manufacturing data. Manufacturing data typically includes information related to the product to be manufactured, the manufacturing process, and the resources required. CAM solutions are used to plan and optimize the entire manufacturing process of a product. For example, it can provide CAM users with information about feasibility, the duration of the manufacturing process, or the amount of resources (such as specific robots) that can be used at a particular step in the manufacturing process; thus, decisions can be made regarding management or required investment. CAM is a follow-up process to CAD processes and potential CAE processes. Such CAM solutions are produced by Dassault Systèmes under the trademark DELMIA. supply.

[0072] CAE solutions also refer to any solution, hardware or software, applicable to analyzing the physical behavior of the modeled object. The Finite Element Method (FEM) is a well-known and widely used CAE technique that typically involves decomposing the modeled object into elements whose physical behavior can be calculated and simulated using equations. Such CAE solutions are developed by Dassault Systèmes under the trademark SIMULIA. Provided. Another evolving CAE technology involves modeling and analyzing complex systems composed of multiple components from different physical realms, without CAD geometric data. CAE solutions can perform simulations, allowing for the optimization, improvement, and validation of products to be manufactured. Such CAE solutions are offered by Dassault Systèmes under the trademark DYMOLA. supply.

[0073] PDM stands for Product Data Management. A PDM solution refers to any solution, both hardware and software, suitable for managing all types of data related to a specific product. All stakeholders in the product lifecycle can use a PDM solution: primarily engineers, but also project managers, finance personnel, sales staff, and purchasing agents. PDM solutions are typically based on a product-oriented database. It allows stakeholders to share consistent data across their products, thus preventing stakeholders from using conflicting data. This type of PDM solution is produced by Dassault Systèmes under the trademark ENOVIA. supply.

[0074] Figure 24 An example of the system's GUI is shown, where the system is a CAD system.

[0075] The GUI 2100 may be a typical CAD-like interface with standard menu bars 2110, 2120 and bottom and side toolbars 2140, 2150. These menu bars and toolbars contain a set of user-selectable icons, each associated with one or more operations or functions as known in the art. Some of these icons are associated with software tools suitable for editing and / or working on the 3D modeled object 2000 displayed in the GUI 2100. The software tools may be grouped into workbenches. Each workbench contains a subset of the software tools. In particular, one of the workbenches is an editing workbench suitable for editing the geometry of the modeled product 2000. In operation, the designer may, for example, pre-select a portion of the object 2000 and then initiate an operation (e.g., change dimensions, color, etc.) or edit geometric constraints by selecting the appropriate icon. For example, a typical CAD operation is modeling the drilling or folding of a 3D modeled object displayed on the screen. The GUI may, for example, display data 2500 related to the displayed product 2000. In this example diagram, the data 2500, shown as a "feature tree," and its 3D representation 2000 relate to a braking assembly including brake calipers and discs. The GUI can further illustrate various types of graphical tools 2130, 2070, 2080, such as those used to facilitate 3D orientation of objects, to trigger operations on editing the product, or to simulate or render various properties of the displayed product 2000. A cursor 2060 can be controlled by a haptic device to allow the user to interact with the graphical tools.

[0076] Figure 25 An example of a system is shown, where the system is a client computer system, such as a user's workstation.

[0077] The client computer in this example includes a central processing unit (CPU) 1010 connected to an internal communication bus 1000, and random access memory (RAM) 1070 also connected to the bus. The client computer is also provided with a graphics processing unit (GPU) 1110 associated with video random access memory 1100 connected to the bus. The video RAM 1100 is also referred to in the art as a frame buffer. A mass storage device controller 1020 manages access to a mass storage device (e.g., a hard disk drive 1030). Mass storage devices suitable for tangibly representing computer program instructions and data include all forms of non-volatile memory, including, for example, semiconductor memory devices such as EPROM, EEPROM, and flash memory devices; disks, such as internal hard disks and removable disks; magneto-optical disks; and CD-ROM disks 1040. Any of the above can be supplemented or incorporated by a specially designed ASIC (Application-Specific Integrated Circuit). A network adapter 1050 manages access to a network 1060. The client computer may also include a haptic device 1090, such as a cursor control device, a keyboard, etc. A cursor control device is used on the client computer to allow the user to selectively position the cursor at any desired location on the monitor 1080. Furthermore, the cursor control device allows the user to select various commands and input control signals. The cursor control device includes multiple signal generating devices for inputting control signals to the system. Typically, the cursor control device can be a mouse, with buttons used to generate signals. Alternatively or additionally, the client computer system may include a sensitive pad and / or a sensitive screen.

[0078] The computer program may include computer-executable instructions, which include means for causing the system to perform the method. The program may be recorded on any data storage medium, including the system's memory. The program may be implemented, for example, as digital electronic circuitry or computer hardware, firmware, software, or a combination thereof. The program may be implemented as a means tangibly embodied in a machine-readable storage device for execution by a programmable processor, such as a product. The method steps can be performed by a programmable processor executing the instruction program to perform the function of the method by manipulating input data and generating output. Therefore, the processor may be programmable and coupled to receive data and instructions from the data storage system, at least one input device, and at least one output device, and to send data and instructions to the data storage system, at least one input device, and at least one output device. If desired, the application program may be implemented in a high-level procedural or object-oriented programming language, or assembly or machine language. In any case, the language may be a compiled language or an interpreted language. The program may be a complete installation or update program. In any case, the application of the program to the system will result in instructions for performing the method.

[0079] "Designing a 3D modeling object" specifies any action or series of actions, at least as part of the process of describing the 3D modeling object. Therefore, the method could include creating a mesh for the 3D modeling object, for example, a mesh obtained by using 3D scanning technology to acquire a relevant point cloud of a physical prototype.

[0080] This method can be incorporated into a manufacturing process that, after execution, produces a physical product corresponding to the modeled object. In any case, the modeled object designed using this method can represent a manufactured object. The manufactured object can be a product, such as a part or part assembly. Because this method improves the design of the modeled object, it also improves the manufacturing of the product, thereby increasing the productivity of the manufacturing process.

[0081] Refer again Figure 5 The flowchart describes a step S10 where a mesh of a 3D modeling object is provided. The mesh includes elements as vertices, edges connecting the vertices, and faces formed by at least three vertices. Faces are polygons, such as triangles. The mesh represents the shape of a prototype. A prototype is typically a physical model of a product to be manufactured. This product belongs to a wide variety of industrial sectors. Examples of industrial sectors have already been given above. In the following discussion, the product to be manufactured is a car body; it should be understood that the invention can be applied to other industries.

[0082] The mesh can be obtained from a laser scan of the prototype. This yields a point cloud, which is then meshed. Any technique capable of representing a physical object as a point cloud can be used. Similarly, any technique capable of meshing a point cloud can be used. The provided mesh represents the shape of the product to be designed.

[0083] The designed 3D modeled object includes a wireframe (or wire-frame) based on at least one characteristic line. This means that the physical prototype from which the provided mesh is derived also contains these characteristic lines. The wireframe includes the most important lines on which the product surface is placed. A wireframe represents a three-dimensional (3D) physical object used in 3D computer graphics. This representation can be a visual representation. A wireframe includes the edges of a physical object where two surfaces intersect. A wireframe is typically a 3D wireframe computer model, i.e., a 3D modeled object. Wireframes can be used to construct and manipulate solids and solid surfaces.

[0084] A wireframe is based on at least one characteristic line. A characteristic line is a line that describes the features of a product. A product is a physical or multi-physics system, meaning that the product behaves with or in its environment, and the characteristic lines contribute to the product's behavior in the real world. The behavior of a product can be modeled using at least one (i.e., one or more) physical models from at least one physical domain (e.g., one of the examples of physical domains mentioned above). This means that the characteristic lines influence the physical model, and the results of the physical model depend (partially or entirely) on one or more characteristic lines of the product. A real-world system or physical entity can be an electronic product, an electrical product, a mechanical product, an electromechanical product, a particle system, or an electromagnetic product. A physical model can be an electronic model, an electrical model, a mechanical model, a statistical model, a particle model, a hydraulic model, a quantum model, a geological model, an astronomical model, a chemical model, an electromagnetic model, or a fluid model. A physical model can be a multi-physics model. Both refer to real-world systems or any physical entity with subsystems that are interconnected through physical or logical relationships (e.g., mechanical relationships (e.g., connections corresponding to the transmission of force or motion), electrical relationships (e.g., electrical connections, such as in a circuit), hydraulic relationships (e.g., behavior corresponding to the transmission of flux), logical relationships (e.g., corresponding to information flow), fluid relationships (e.g., corresponding to fluid flow), chemical relationships, and / or electromagnetic relationships). "Multi-physics" means that the physical or logical relationships of a multi-physics system can belong to multiple physical domains (although not necessarily). After its virtual design is completed through this invention, the physical or multi-physics system can correspond to a (industrial) product to be manufactured in the real world.

[0085] Therefore, one or more characteristic lines of a wireframe can contribute to one or more physical fields, such as electronics, electricity, mechanics, electromechanics, fluid mechanics, gravity, statistical mechanics, wave physics, statistical physics, particle systems, hydraulic systems, quantum physics, geophysics, astrophysics, chemistry, aerospace, geomagnetism, electromagnetism, and plasma physics.

[0086] As mentioned earlier, the 3D modeling object being designed includes a wireframe based on at least one characteristic line. The wireframe can therefore be used in the product simulation process. Figures 1 to 4 This is a simulation example of the physical behavior of a car body, where the wireframe of the car body contains feature lines of the physical or multi-physics system (i.e., the car body) that affect the simulation.

[0087] Therefore, the provided mesh includes information about the wireframe and one or more characteristic lines, but this information must be extracted from the mesh.

[0088] The provided mesh can be a clean mesh. A clean mesh provides a closed envelope (e.g., no holes in the mesh) and / or a non-overlapping manifold envelope (where no vertices overlap). If the mesh is not clean, it can be automatically repaired based on user actions or using any known mesh repair methods.

[0089] Figure 7 This is an illustration of a clean mesh. In this example, the mesh faces have been rendered with shading applied for illustrative purposes only. The mesh is clean because there are no holes on its surface.

[0090] Figure 8 yes Figure 7 The representation of the mesh, utilizing image filtering techniques, can help assess mesh quality. Noisy meshes (with low positional precision at each vertex) or coarse meshes (with large triangles) display a mixed grayscale image rather than a pure black and white image. For example, region 80, represented in grayscale, exhibits noise compared to other areas of the mesh.

[0091] Next, in S20, the segmentation of the provided mesh is calculated, thus obtaining a segmented mesh. Mesh segmentation is the process of decomposing a mesh into smaller, meaningful sub-mesh. Sub-mesh is called a region; therefore, a region is part of the provided mesh. As a result of the segmentation, at least two regions are obtained. These regions are distinct regions, meaning that no face of the mesh belongs to two distinct regions. It should be understood that two regions can share common vertices and edges, with the common edges connecting the common vertices. The shared edges between two regions form a boundary polyline. Therefore, as a result of the segmentation, at least one boundary polyline is obtained.

[0092] Different segmentation techniques can be applied. The purpose of segmentation is to discover the characteristic lines of the provided mesh, and the selected segmentation technique should preferably be able to extract them from the mesh's outer boundaries and / or the tightest and longest lines. The shortest and longest lines can form the mesh's streamlines.

[0093] In the example, the calculation of the provided mesh segmentation includes the detection of isocurvature lines and the curvature evolution of the provided mesh. Isocurvature lines are contour lines that provide a good understanding of mesh curvature. Contour lines are lines where the eigenvalues ​​of the mesh surface are constant. Isocurvature lines are lines where the curvature values ​​of the mesh surface are constant; therefore, isocurvature lines can effectively show and visualize changes in curvature. For high curvature values, isocurvature lines represent the tightest lines representing the characteristic features of the mesh.

[0094] In the example, one technique for segmenting the mesh is done by calculating the normal clusters of the provided mesh. Figure 9This is a diagram illustrating the normal clustering process. Each face of the mesh is associated with a normal. This can be performed as is known in the art. The point cloud is associated with a voxel mesh structure. The voxel mesh structure divides the region into several regions along the same normal direction. Vertices of faces whose normal direction is close to one of the voxels are associated with that voxel. The number of regions created depends on the clustering level. The mesh with the most voxels has the highest clustering level.

[0095] Figures 10 to 12 This demonstrates clustering using computed normals for... Figure 7 An example of clean mesh computation and segmentation. In Figure 10 In this case, the clustering level is low, resulting in a low number of regions. Figure 11 In this process, clustering levels were added, and more regions were identified. Figure 12 This represents the cluster with the largest recognizable region on the provided grid. Figure 10 The area obtained above has been Figure 11 Subdivided into one or more sub-regions, and Figure 11 and Figure 12 Similar to each other. The more grids there are, the higher the number of boundary polylines between regions.

[0096] Users can define the optimal clustering level, as the graph may be highly segmented or fuzzily segmented. In this step of the method, a lower clustering level has the advantage of reducing the impact of mesh quality. If the mesh is noisy, the segmentation can be kept at a lower level so that mesh defects do not lead to the detection and creation of regions without interesting information, thus compromising the identification of boundary polygons that may represent future feature lines. Regular variations in normal direction have an impact on the segmentation process. For example, a smooth sphere is segmented with this level of normal because the normal vector covers the entire space. Therefore, sharp and smooth edges will not be reproduced with a low clustering level. A good clustering level is one that reproduces feature lines. Users can edit the clustering level to adjust the results to their goals. In practice, several levels of clustering can be considered depending on the goals of 3D modeling:

[0097] • Sharp modeling, where the model is reconstructed as a sharp point without rounded corners, and then rounded corners can be added;

[0098] • Smooth modeling, where the segmentation includes rounded corner regions, and the rounded corners are reconstructed;

[0099] • Segmentation for class A, in which not only rounded regions but also curvature acceleration regions are isolated and included in the segmentation.

[0100] Therefore, users want to control and validate clustering levels according to their expectations and fine-tune the segmentation results.

[0101] In the example, the segmentation of the provided mesh may include detecting at least one primary boundary polyline. A primary boundary polyline is a boundary polyline with higher curvature compared to other polylines in the provided mesh. The mesh may include one or more primary boundary polylines. For each region in the provided mesh that has a high or highest curvature value (e.g., an extreme value of curvature), a boundary polyline can be obtained.

[0102] Now for reference Figure 6 An example of segmenting the provided mesh is discussed (S20). In this example, segmentation is performed by first calculating a contrast image of the provided mesh (S210). The contrast image is also called a contrast map. The purpose of the contrast image is to identify the extrema of the curvature evolution of the provided mesh. Other techniques aimed at identifying the extrema of curvature evolution can also be used. In the example, the image filtering process can use a process of brightening the contrast on the mesh image. The intensity value of a pixel can be compared with the average intensity value of its neighboring pixels. The difference value can be plotted using gray levels from white (0) to black (maximum difference). Here, the triangle normal is compared with the average polygon normal of its neighbors. The neighbor of a polygon is a polygon that shares at least one vertex with that polygon. The difference can be plotted on a logarithmic gray level from white (same normal = plane) to black (highest curvature).

[0103] In the example, the comparison image could be an adaptive map used to analyze the quality of the mesh. In this case, lower-quality regions might be extracted and isolated for separate processing from automated procedures. This improves the identification of boundary polylines in low-quality mesh regions.

[0104] Next, in S220, local line segments are extracted from the calculated contrast image. Local line segments are common edges between black and gray triangles, or common edges where the derivative of curvature is an extremum. The extracted local line segments are linked into the longest polyline. These polylines can be enriched with new polylines to define closed lines. New polylines that allow the definition of closed lines are selected to maximize the derivative of curvature. All the obtained polylines form a polyline network. In the example, new polylines can be calculated as follows: once polylines are obtained, they are propagated along a direction perpendicular to curvature to another polyline or boundary to assemble them. This yields closed regions (or closed lines).

[0105] The provided mesh is then divided into a polyline network (S230). Each closed line surrounds a set of triangles that define a submesh. A triangle can belong to a single submesh. The union of all submeshes corresponds to the initially provided mesh.

[0106] Next, in S240, the mesh partitioning is refined by applying normal clustering to each sub-mesh generated by S230. Returning to... Figures 10 to 12Their respective representations form the computed segmentation map: each region is presented in a different color. It should be understood that the colors are merely examples, and the markings may be invisible to the user.

[0107] Back Figure 6 In S250, a graph of the segmented mesh is calculated, or rather, a graph of the provided mesh is calculated from the refined segmentation. Therefore, the graph will include at least two regions and at least one boundary polyline between the at least two distinct regions. The term "graph" refers to a data scheme in which:

[0108] • Triangles belonging to a region are marked as belonging to that region;

[0109] Similarly, an edge belonging to one of the boundary polylines is marked as belonging to that boundary polyline and references the adjacent region; vertices belonging to multiple boundary polylines are marked as belonging to that boundary polyline and refer to the connecting polylines.

[0110] In this graph, there are no repetitions, the edges (which are vertices) are common, and they are shared between adjacent regions (which are polylines).

[0111] Referring again to segmentation S20, as already discussed, segmentation may reveal too many regions (depending on the clustering level). This can lead to decreased efficiency in identifying characteristic lines as the number of boundary polylines increases. In the example, two stages can be distinguished in the segmentation. The first stage may include computing a first segmentation, thus obtaining at least two first regions and at least one first boundary polyline between the provided mesh. The at least one first boundary polyline may be referred to as the primary boundary polyline of the mesh. The second stage may include computing a second segmentation performed at a higher level of segmentation optimization compared to the first segmentation. As a result of the second segmentation, at least two second regions and at least one second boundary polyline between the provided mesh are obtained, the at least two second regions belonging to at least one of the at least two first regions. The at least one second boundary polyline may be referred to as the secondary boundary polyline of the mesh.

[0112] Now for reference Figure 6 An example of two stages in the segmentation is discussed. As a result of this first stage (S230), at least two first regions from the provided mesh and at least one first boundary polyline between the at least two first regions are obtained. This first segmentation can be performed as previously discussed. One or more first boundary polylines are called the main boundary polylines of the mesh. The main boundary polylines are considered to be the lines that best represent the characteristics of the product.

[0113] Following the first stage, a second segmentation (S240) is performed. The second segmentation achieves a higher level of subdivision optimization compared to the first segmentation. As a result of the second segmentation, at least two second regions are obtained from the provided mesh. These at least two second regions belong to at least one of the at least two first regions. At least one second boundary polyline is still obtained between the at least two second regions. One or more other boundary polylines are discovered. These additional boundary polylines are called the mesh's secondary boundary polylines.

[0114] In the example, the designer can trigger the splitting or merging of regions. User intervention is only required if the automatic segmentation process fails to identify certain characteristic lines. This might be the case when the mesh is noisy. The user can select regions they want to split. The curvature map is then used to automatically split the regions after the curvature map. The user can select at least two regions they want to merge or expand, and then merge the selected regions. Interestingly, the system provides assistance to the user during the region splitting process. The curvature map can be a second-order curvature map. Mesh portions with significant curvature can be identified. The curvature map is a curvature texture from a high-polygon mesh, producing accurate results. Graph calculations are performed as known in the art. Therefore, the user can use the provided mesh's curvature map to select one of the regions and split the selected region, and / or they can select at least two regions and merge them into a single region.

[0115] Figure 18 and Figure 19 The splitting and merging of regions are shown. Figure 18 Regions 1800 and 1802 have been merged, thus forming Figure 19 The new region 1900 is shown. Regions 1810 and 1830 have been split using a curvature diagram, and the newly discovered region has been merged with regions 1810 and 1820 to form new regions 1910 and 1920. A portion of the previously split region 1830 has been merged with regions 1810 and 1820.

[0116] Back Figure 5S30. In this step of the method, the wireframe of the provided mesh object can be reproduced well. The characteristic lines of the provided mesh are polylines following sharp and smooth edges. However, polylines cannot be directly utilized by product designers and CAD systems implementing this method. Therefore, the boundary polylines are converted into smooth curves. The conversion involves placing control points on the wireframe to control deviations relative to the curves and skeleton. Points can be added until the curves are within the tolerance zone from the mesh. Curves can be created with as few points as possible to obtain high-quality spikes along each curve. Interestingly, the mathematical descriptions of the created curves may differ. If the quality of the smooth curves is dominant, the number of line segments and / or the number of control points and / or the degree of smoothness of the curves can be minimized. The resulting surfaces resting on these curves will be tighter. This improves the design and characteristic lines of the object to reproduce the surface of the prototype. Local variations in the curves are minimized, and the view orientation has no effect on the results. Only the quality of the segmentation view affects the obtained curves.

[0117] Therefore, the transformation (S30) includes calculating a smoothing curve for each boundary line of the previously calculated (S20). The smoothing curve represents the feature lines of the digitized prototype. As is known in the art, the set of points is transformed into a smoothing curve. Points can be placed on the calculated smoothing curve to minimize deviations from the smoothing curve and the provided mesh. This improves the tolerance of the wireframe relative to the shape of the prototype. Additionally or alternatively, the number of control points is minimized. Therefore, the quality of the needles along each smoothing curve is improved. Additionally or alternatively, the degree of smoothing is minimized; this helps to improve the quality of the smoothing curve.

[0118] The properties of the smooth curve can be selected by the user or left as default. Different categories of surfaces can be selected to lie on the calculated smooth curve. A "Class A" surface can be defined as any surface supported by a property line, and with curvature continuity mathematically of G2 or even G3. Curvature continuity refers to the continuity between surfaces sharing the same boundary. Curvature continuity means that every point on every surface along the common boundary has the same radius of curvature, so the boundary is blended, and therefore no physical joints exist or are not visible. A Class A surface has continuous curvature without unwanted ripples.

[0119] In the automotive industry, Class A surfaces are primarily used on all visible exterior surfaces (e.g., body panels, bumpers, grilles, lights, etc.) and all visible surfaces of interior tactile and sensory parts (e.g., dashboards, seats, door mats, etc.). This may also include hoods, fenders, trunk panels, and carpets. Class A surfaces are also used in high-tech industries and everyday consumer goods. In product design, Class A surfaces can be applied to things such as the housings of injection-molded industrial equipment, household appliances, plastic packaging defined by highly organic surfaces, toys, or furniture. Aerospace also uses Class A surfaces, for example, when designing interior trim such as baffles for vents and lighting, roof racks, seats, and cabin areas.

[0120] Class B and Class C surfaces can also be used. They have lower quality and lower levels of continuity between adjacent surfaces. Any surface supported by property lines has mathematically G0 or G1 curvature continuity.

[0121] When using G0 continuity, the distance between every point on the edge of two adjacent facets must meet the following constraints:

[0122] - For Class A: Not exceeding 0.01 mm;

[0123] - For Class B: Not exceeding 0.02 mm;

[0124] - For Class C: No more than 0.05 mm.

[0125] When using G1 continuity (also known as tangent continuity), the angle between the edges of two adjacent facets and the surface tangent must meet the following constraints:

[0126] - For Category A: no more than 6 points (0.1°);

[0127] - For Category B: no more than 12 points (0.2°);

[0128] - For Category C: no more than 30 points (0.5°).

[0129] For G2 continuity (also known as curvature continuity), the control parameter is the surface curvature along its contour. The curvature must meet the following constraints:

[0130] - For Class A surfaces, the curvature of the contours of at least two adjacent facets must be consistent for every 100 mm.

[0131] There are no applicable rules for classes B and C. Only the point of maximum curvature or inflection point is allowed to be drawn along the patch contour of a class A surface.

[0132] Figure 20 yes Figure 19The boundary polylines of the mesh are transformed into a smooth curve representation. The automatically generated smooth curves are taut and do not wobble, so they represent the characteristic lines of the provided mesh very well. Figure 21 It is a representation of the spikes in the automatically generated characteristic lines. The set of smooth curves forms the wireframe of the 3D modeling object containing the characteristic lines.

[0133] Next, in S40, a network of at least one characteristic line is calculated. The term network refers to the interconnected characteristic lines that form the wireframe, such that each endpoint of a smooth curve is connected to at least one other smooth curve. The connected smooth curve network (i.e., the characteristic line network) forms the wireframe of the 3D modeled object. The action of connecting the endpoints of the smooth curves is called snapping.

[0134] In the example, the construction (computation) of the network can depend on the definition of the master curve and the slave curve. The definition of the master curve and the slave curve depends on the definition of the master boundary polyline and the slave boundary polyline, performed as discussed in the two stages of the reference segmentation process. The slave curve is obtained by calculating a smoothed curve of the slave boundary polyline. The master curve is obtained by calculating a smoothed curve of the master boundary polyline. The ends of the slave curve are slightly adjusted so that they lie on and connect to the master curve. For example, if the master curve deforms, the ends of the slave curve change accordingly. Thus, at least one first boundary polyline emanating from the first segmentation is achieved. This improves the preservation of the characteristic lines of the wireframe forming the provided 3D mesh.

[0135] In the example, a set of master smooth curves is created; these are the longest smooth curves. If at least one endpoint of a smooth curve is not connected to another smooth curve, then the smooth curves in this master curve set will connect to another smooth curve. When it is necessary to connect smooth curves, the closest smooth curve in the master curve set can be selected. Distance can be used to select the closest smooth curve, such as Euclidean distance or geodesic distance.

[0136] The endpoints of the smooth curves to be connected are moved to the point of minimum distance on the main curve.

[0137] In the example, the master smoothing curve group comprises the master smoothing curves of the wireframe. The master smoothing curves are obtained after transforming the main boundary polyline of the mesh. The master smoothing curves are considered to best represent the product feature lines. A first iteration is performed to connect the curves in this master group. Within this group, the master curve of each connected pair of curves is the longest curve. A second iteration is then performed using the slave smoothing curves (smooth curves obtained after transforming the boundary polyline into a smoothing curve). Slave curves with at least one endpoint not connected to a smoothing curve will preferably be captured by one of the master smoothing curves. The closest master smoothing curve can be selected. Distance can be used to select the closest smoothing curve, such as Euclidean distance or geodesic distance. If the slave smoothing curves are still not connected after the second iteration, a similar process to that described above can be applied to the slave smoothing curve group.

[0138] Therefore, the system automatically calculates the master curves that other curves can capture. The aim is to support longer taut curves. In this case, the creation of the long master curve can precede the creation of master curves that are not captured. Similarly, the first boundary polyline emanating from the first segment (now converted into a master curve) is supported.

[0139] In one example, the snap point from the curve can be moved along the main smoothing curve through simple editing. It's worth noting that designers may want to control, adjust, or fine-tune the network through manual editing.

[0140] Figure 22 The example shown consists of three curves (curve 1, curve 2, and curve 3) that are not the main curves. These three curves are interconnected here with their respective endpoints 2200 and minimum continuity (G0).

[0141] Figure 23 This illustrates an example where curve 2 is the master curve. The master curve 2 is captured from the endpoint of curve 1. We maintain complete continuity between the two parts of curve 2.

[0142] Further design operations can now be performed on the 3D-modeled wireframe. In the example, surface components are calculated based on the wireframe of the 3D-modeled object. Furthermore, the system can also calculate the continuity constraints of the 3D-modeled object. To generate a surface from the wireframe, continuity (e.g., G0: simple continuity, or G1: tangent continuity) can be associated with each common boundary curve between two adjacent regions, and the generated surface will conform to that common boundary curve. Continuity is provided to the user after geometric analysis of concurrent curves in the network nodes. The possibility of reducing the provided continuity can be offered to the user, for example, from G1 to G0.

[0143] refer to Figures 13 to 17 Now we will discuss an example of calculating the segmentation map obtained from segmentation (S20). Figure 13These represent characteristic lines of, for example, a digitized prototype (physical model), such as a point cloud generated by laser scanning. Figure 14 The main curve on the prototype to be detected is shown. The first segmentation (S230) is performed to obtain the outer and inner boundary polylines and other key characteristic lines. These are, in fact, the longest and tightest lines. Figure 15 The boundary polyline extracted from the mesh is shown. Figure 16 The remaining characteristic lines 1600, 1610, 1620, 1630, and 1640 of the mesh to be extracted are shown. These remaining characteristic lines are from the boundary polylines. These polylines represent important details of the mesh and are local polylines. In fact, they connect to the main polylines. A second segmentation is performed at a higher clustering level, resulting in a refinement of the region detection due to the first segmentation.

Claims

1. A computer-implemented method for a 3D modeling object used to design a physical prototype of a product, the 3D modeling object comprising a wireframe based on at least one characteristic line, the method comprising: S10: Provide the mesh for the 3D modeled object; S20: Calculate the division of the provided mesh to obtain at least two regions and at least one boundary polyline between the at least two regions; S30: Transform each of at least one boundary polyline into at least one characteristic line; as well as S40: Calculate the network of the at least one characteristic line, the network of the at least one characteristic line forming the wireframe of the 3D modeling object. The mesh segmentation provided by the calculation includes: Detect at least one principal boundary polyline of the provided mesh, which is a polyline with higher curvature compared to other polylines of the provided mesh. Detecting at least one main boundary polyline includes: A contrast map of the grid is calculated by applying image filtering to the grid. Calculate the curvature evolution and identify the extreme values ​​of the curvature evolution; The at least one main boundary polyline is calculated by linking local line segments extracted from the contrast map.

2. The computer-implemented method according to claim 1, wherein, The mesh segmentation provided by the calculation includes: Calculate the first segment to obtain: At least two first regions in the provided grid; and At least one first boundary broken line between the at least two first regions; Calculate a second segmentation performed at a higher optimized segmentation level compared to the first segmentation, thereby obtaining: The provided grid contains at least two second regions, which belong to at least one of the at least two first regions; and At least one second boundary broken line between the at least two second regions.

3. The computer-implemented method according to claim 2, wherein, The network for calculating the at least one characteristic line includes: Select at least one of the first boundary polylines; A second smooth curve calculated from at least one second boundary polyline is connected to a first smooth curve calculated from at least one selected first boundary polyline by performing a snapping operation, wherein the end of the second smooth curve is adjusted to lie on the first smooth curve.

4. The computer-implemented method according to claim 1, wherein, The computation of the provided mesh segmentation also includes fine segmentation by calculating normal clustering on each region of the mesh.

5. The computer-implemented method according to claim 1, further comprising: After calculating the division of the provided mesh: Select one of the at least two regions and divide the selected region using the curvature map of the provided grid; and / or Select at least two regions and merge them into one region.

6. The computer-implemented method according to claim 5, wherein, The transformation includes calculating a smooth curve for each of the at least one boundary polyline.

7. The computer-implemented method according to claim 6, further comprising: Place the control points of the smooth curve, wherein the placement is performed to minimize the following: The deviation relative to the smooth curve and the provided grid; and / or The number of control points; and / or The degree of smoothness of the curve.

8. The computer-implemented method according to any one of claims 1 to 7, further comprising: The surface components are calculated based on the continuity constraints of the wireframe and the 3D modeled object.

9. A computer program product comprising instructions for performing the method according to any one of claims 1-8.

10. A computer-readable storage medium having instructions recorded thereon for performing the method according to any one of claims 1-8.

11. A computer system comprising a processor coupled to a memory and a graphical user interface, the memory storing instructions for performing the method according to any one of claims 1-8.

Images