A three-dimensional orthodontic model generation method and system

Abstract

The present application relates to the technical field of model generation, and more particularly to a three-dimensional orthodontic model generation method and system, the present application proposes the following scheme, obtaining a target bearing object three-dimensional model, a local protrusion structure three-dimensional model and a reference shell three-dimensional model, determining the fitting path of the reference shell relative to the target bearing object, and detecting the path interference information under multiple intermediate fitting postures; according to the path interference information, determine the import area, the avoidance area and the fitting control area, difference modeling is carried out on the local groove body to be generated area, and a three-dimensional shell model with local groove body structure is generated. Thus, the temporary interference in the fitting process can be released, and the fitting control relationship in the final fitting state is retained.

Description

Technical Field

[0001] This invention relates to the field of model generation technology, and in particular to a method and system for generating three-dimensional orthodontic models. Background Technology

[0002] With the development of 3D scanning, computer-aided design, and digital manufacturing technologies, the method of generating fitting shell models based on the 3D model of the target object has been widely used in medical, industrial assembly, personalized protection, and positioning and clamping scenarios. For target objects with local protrusions on their surfaces, the shell model usually needs to form corresponding local grooves on its inner surface so that the shell can accommodate the local protrusions in the final fitted state and maintain the expected fitting, limiting, or covering relationship.

[0003] Existing methods for generating 3D shell models often use the final mating posture as the modeling benchmark. Based on the outer contour of the local protrusion at the final position, the inner surface of the benchmark shell is subjected to equidistant offset, Boolean subtraction, or uniform groove expansion to obtain the local groove structure. While this can meet the static containment requirements of the final position, for thin-walled shells, elastic shells, or shells with high fit, the local protrusion may experience procedural contact such as oblique entry, partial rubbing, screwing in, or mid-way abutment during the process of the shell moving from the unmating state to the final mating state. In this case, even if the local protrusion can be located within the groove in the final mating posture, it may still interfere with the groove opening or groove wall in the middle stage of the mating path.

[0004] If the aforementioned jamming problem is improved by increasing the overall gap between the slots, it may weaken the fit and restraint of the local slots on the local protruding structures in the final fitting state. This would result in the shell model being easy to fit, but difficult to maintain the original control precision. Taking invisible orthodontic appliances as an example, the attachment slots corresponding to the tooth surface attachments need to ensure that the attachments can be smoothly inserted during the appliance wearing process, and also need to maintain the ability to cover, traction, and torque control of the attachments after they are in place. Therefore, how to simultaneously consider path avoidance during the fitting process and fit control in the final state during the three-dimensional shell model generation stage has become a problem that needs to be solved in related model generation technologies.

[0005] To address the above issues, this application presents a method and system for generating three-dimensional orthodontic models. Summary of the Invention

[0006] The technical problem this invention aims to solve is to address the shortcomings of existing technologies by providing a method and system for generating three-dimensional orthodontic models. This method involves acquiring a three-dimensional model of the target support object, a three-dimensional model of the local protrusion structure, and a three-dimensional model of the reference shell. It determines the fitting path of the reference shell relative to the target support object and detects path interference information under multiple intermediate fitting postures. Based on the path interference information, it determines the introduction area, avoidance area, and fitting control area, and performs differentiated modeling on the local groove area to be generated, generating a three-dimensional shell model with local groove structures. This eliminates temporary interference during the fitting process while preserving the fitting control relationship in the final fitting state.

[0007] To achieve the above objectives, the present invention provides the following technical solution: A method for generating a three-dimensional orthodontic model, the method comprising: Obtain a three-dimensional model of the target load-bearing object, a three-dimensional model of the local protrusion structure set on the surface of the target load-bearing object, and a three-dimensional model of the reference shell corresponding to the three-dimensional model of the target load-bearing object; Based on the final fitting posture of the target bearing object 3D model and the reference shell 3D model, the fitting path of the reference shell 3D model relative to the target bearing object 3D model is determined, and the fitting path includes multiple intermediate fitting postures; Under each of the aforementioned intermediate fitting postures, the path interference information between the three-dimensional model of the local protrusion structure and the inner surface of the three-dimensional model of the reference shell is detected, and based on the path interference information, the introduction area, the avoidance area, and the fitting control area are determined in the local groove to be generated area corresponding to the three-dimensional model of the local protrusion structure. Based on the import area, the avoidance area, and the fitting control area, the local groove to be generated area is differentially modeled to generate a three-dimensional shell model with a local groove structure.

[0008] The area to be generated for the local groove is used to characterize the spatial range in which the local protrusion structure three-dimensional model forms the corresponding local groove on the inner surface of the reference shell three-dimensional model. The area to be generated for the local groove is determined based on the outer contour of the three-dimensional model of the local protrusion structure, the outward convex direction of the three-dimensional model of the local protrusion structure relative to the three-dimensional model of the target bearing object, and the surface transition area around the three-dimensional model of the local protrusion structure.

[0009] The method for determining the fitting path includes: Based on the fitting relationship between the target bearing object 3D model and the reference shell 3D model, the final fitting posture of the reference shell 3D model relative to the target bearing object 3D model is determined; Based on the main body extension direction and contour extension direction of the target bearing object three-dimensional model, the opening edge direction of the reference shell three-dimensional model, and the distribution position of the local protrusion structure three-dimensional model in the target bearing object three-dimensional model, the initial fitting posture of the reference shell three-dimensional model is determined. Based on the relative pose change between the initial fitting posture and the final fitting posture, the fitting path of the reference shell 3D model relative to the target bearing object 3D model is generated.

[0010] The intermediate overlay posture is determined based on the relative pose change, and the determination method of the intermediate overlay posture includes: The relative pose change between the initial fitting posture and the final fitting posture is decomposed to obtain the translational change and the rotational change. The attitude sampling interval is determined based on the translational change, the rotational change, the number of three-dimensional models of the local protrusion structure, and the outward convexity height of the three-dimensional model of the local protrusion structure relative to the three-dimensional model of the target bearing object. According to the attitude sampling interval, the relative pose change between the initial fitting pose and the final fitting pose is interpolated to obtain multiple intermediate fitting poses.

[0011] The methods for determining the path interference information include: For each intermediate fitting posture, determine the relative position between the three-dimensional model of the local protrusion structure and the inner surface of the three-dimensional model of the reference shell; When the distance between the three-dimensional model of the local protrusion structure and the inner surface of the three-dimensional model of the reference shell is less than the preset safety gap, the corresponding position is determined as the path interference position; The corresponding path interference depth is determined based on the difference between the path interference position and the preset safety gap. The path interference position, the path interference depth, and the corresponding intermediate overlay posture are determined as the path interference information.

[0012] The methods for determining the import area, avoidance area, and fitting control area include: Based on the path interference positions in each of the path interference information, the path interference positions that are located within the local groove to be generated area and are adjacent to each other are merged into candidate interference areas; Based on the arrangement order of the intermediate overlay postures corresponding to each path interference information in the overlay path, the interference occurrence stage corresponding to each candidate interference region is determined. The candidate interference region located at the beginning of the mating path during the corresponding interference occurrence stage is determined as the introductory region, and the candidate interference region located at the middle of the mating path during the corresponding interference occurrence stage is determined as the avoidance region. The mating control region is determined based on the mating position of the three-dimensional model of the local protrusion structure and the inner surface of the three-dimensional model of the reference shell under the final mating posture.

[0013] The differentiated modeling methods for the local groove to be generated area corresponding to the import area include: Based on the path interference information corresponding to the import area, determine the entry direction and path interference depth of the three-dimensional model of the local protrusion structure when it enters the local groove to be generated area; The slot expansion amount of the introductory area is determined based on the path interference depth, and the slot boundary corresponding to the introductory area is expanded according to the slot expansion amount.

[0014] The differentiated modeling methods for the local tank body to be generated corresponding to the avoidance zone include: Based on the path interference information corresponding to the avoidance zone, determine the intermediate overlay attitude associated with the avoidance zone; Obtain the occupied area of ​​the three-dimensional model of the local protrusion structure relative to the inner surface of the three-dimensional model of the reference shell under each associated intermediate overlay posture, and merge each occupied area to obtain the path sweep area corresponding to the avoidance area; Based on the path interference depth corresponding to the path sweep area and the avoidance area, candidate avoidance cavities are generated; Based on the distance relationship between the candidate avoidance cavity and the fitting control area, the candidate avoidance cavity is subjected to boundary constraint processing to obtain a restricted avoidance cavity; Based on the restricted avoidance cavity, a local concavity model is performed on the local groove area to be generated corresponding to the avoidance area.

[0015] The generation methods of the three-dimensional shell model include: Based on the differentiated modeling results corresponding to the import area, the avoidance area, and the fitting control area, respectively, the import groove surface, the avoidance groove surface, and the fitting groove surface are generated; Transitional connections are made between the adjacent boundaries of the inlet groove surface, the avoidance groove surface, and the fitting groove surface to obtain a local groove surface. The local groove surface is fused with the inner surface of the shell in the reference shell 3D model corresponding to the area where the local groove is to be generated, and the fusion boundary is smoothed to generate a 3D shell model with a local groove structure.

[0016] A three-dimensional orthodontic model generation system, the system comprising: The model acquisition module is used to acquire a three-dimensional model of the target bearing object, a three-dimensional model of the local protrusion structure set on the surface of the target bearing object, and a three-dimensional model of the reference shell corresponding to the three-dimensional model of the target bearing object. The fitting path determination module is used to determine the fitting path of the reference shell three-dimensional model relative to the target load-bearing object three-dimensional model based on the final fitting posture of the target load-bearing object three-dimensional model and the reference shell three-dimensional model. The fitting path includes multiple intermediate fitting postures. The interference partitioning module is used to detect the path interference information between the three-dimensional model of the local protrusion structure and the inner surface of the three-dimensional model of the reference shell under each of the intermediate fitting postures, and to determine the introduction area, avoidance area and fitting control area in the local groove to be generated area corresponding to the three-dimensional model of the local protrusion structure based on the path interference information. The tank modeling module is used to perform differentiated modeling on the local tank to be generated area based on the import area, the avoidance area and the fitting control area, and generate a three-dimensional shell model with local tank structure.

[0017] Compared with the prior art, the beneficial effects of the present invention are: This application detects the path interference information between the local protrusion structure and the inner surface of the reference shell under multiple intermediate fitting postures, and divides the area to be generated of the local groove into an introduction zone, an avoidance zone, and a fitting control zone accordingly, so that the groove generation is no longer limited to static subtraction under the final posture. Therefore, the introduction zone improves the smoothness of the local protrusion structure entering the groove, the avoidance zone releases temporary resistance during the fitting process, and the fitting control zone retains the fitting, limiting, or holding relationships in the final fitting state, reducing the risk of jamming and overall groove expansion causing a decrease in control capability. Attached Figure Description

[0018] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 An exemplary application scenario diagram provided for an embodiment of this application; Figure 2 This is a schematic diagram of the structure of the three-dimensional model provided in the embodiments of this application; Figure 3 This is a flowchart illustrating a method for generating a three-dimensional orthodontic model, as provided in an embodiment of this application. Detailed Implementation

[0019] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0020] The term "embodiment" as used herein means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0021] The method for generating a 3D shell model with a local groove structure provided in this application embodiment can be applied to a type of 3D model generation scenario where a shell-like structure needs to be fitted onto the outside of a target load-bearing object, and the surface of the target load-bearing object has a local protrusion structure. A common feature of such scenarios is that the shell is not simply covered on a smooth surface, but rather requires the formation of grooves, cavities, or clearance spaces on its inner surface corresponding to the local protrusion structure, so that the shell can maintain the expected fit with the target load-bearing object in the final fitted state. The shell can be a thin-walled shell, an elastic shell, a transparent shell, a fitted sheath, a positioning clip, or other 3D structure that needs to be assembled onto the outside of the target load-bearing object along a certain path; the target load-bearing object can be an object with an irregular outer surface and local protrusions, positioning blocks, limiting bosses, functional accessories, or auxiliary structures.

[0022] In the process of generating relevant 3D shell models, a common approach is to first obtain the outer surface model of the target load-bearing object, then generate the inner surface of the shell according to a preset offset, and perform Boolean subtraction, equidistant enlargement, or local concavity processing on the inner surface of the shell based on the final position of the local protrusion structure, thereby forming a local groove structure. This approach has the advantages of being direct and easy to implement at the model design level, and is especially suitable for situations where the assembly path between the shell and the target load-bearing object is short, the local protrusion structure is low, or the groove only serves a static containment function. However, when the shell needs to undergo screwing, oblique pressing, local elastic opening, or multi-stage fitting to reach the final fitting state, generating the local groove only based on the final fitting posture often cannot accurately reflect the changes in space occupation of the local protrusion structure during the process of entering the groove.

[0023] Specifically, the fact that a local protrusion can fall into a local groove in the final mating posture does not necessarily mean that the local protrusion will smoothly enter the local groove during the mating path. For cases with narrow opening edges, significant changes in the curvature of the shell's inner surface, relatively large heights of local protrusions, or uneven distribution of multiple local protrusions, as the shell gradually transitions from the initial contact state to the final mating state, the local protrusion may first contact the groove edge, then rub against the middle section of the groove wall, or be blocked by a local area of ​​the shell's inner surface before approaching the final mating. This interference does not necessarily manifest as overlap in the final model, but rather as temporary path interference during the mating process. If the overall enlargement of the groove is still used to eliminate this type of interference, although the probability of assembly jamming can be reduced, it may also decrease the local groove's ability to cover, limit, or control the local protrusion in the final mating state. Without avoidance correction, it is easy for the model to appear to match in the final position, but to be difficult to position or to form a false positioning during actual assembly.

[0024] Based on the aforementioned practical difficulties, this application embodiment does not regard the local groove as a static cavity obtained by subtracting the outer contour of the local protrusion structure, but rather understands it as the restricted space that the local protrusion structure needs to pass through when entering the interior of the shell along the fitting path. In other words, the generation of the local groove not only needs to meet the geometric matching in the final fitting state, but also needs to take into account the dynamic passage relationship under multiple intermediate fitting postures in the fitting path. Therefore, after obtaining the three-dimensional model of the target bearing object, the three-dimensional model of the local protrusion structure, and the three-dimensional model of the reference shell, this application embodiment first reverse-engineers the fitting path of the shell relative to the target bearing object based on the final fitting posture, and sets multiple intermediate fitting postures on the fitting path; then, under each intermediate fitting posture, it detects whether there is path interference between the local protrusion structure and the inner surface of the shell, and then divides the area to be generated of the local groove into an introduction area, an avoidance area, and a fitting control area according to the location, depth, and fitting stage of the interference. Through this process, different areas of the local groove can perform different functions: the guide area is used to improve the guidance of the local protruding structure when it enters the groove, the avoidance area is used to release temporary jamming during the fitting process, and the fitting control area is used to retain the fitting or limiting relationship required in the final fitting state.

[0025] It should be noted that the method described in this application does not rely on the target bearing object having fixed structural partitions, nor does it require that the local protrusion structure be a regular geometry. As long as the shell model needs to form a final fit with the target bearing object through a certain fitting path, and the local protrusion structure on the surface of the target bearing object may temporarily interfere with the inner surface of the shell during the fitting process, the method described in this application can be used to model the local groove structure. The method is particularly suitable for digital design scenarios of thin-walled shells, elastic shells, or shells with high fit, because such shells usually require a smooth assembly process without sacrificing the final fit accuracy by significantly increasing the gap.

[0026] In a typical application, the target support object 3D model can be a target dentition 3D model, the local protrusion structure 3D model can be an attachment 3D model set on the tooth surface, the reference shell 3D model can be an orthodontic appliance reference shell model, and the local groove can be an attachment groove corresponding to the inner surface of the clear aligner. For clear aligners, the attachment groove needs to accommodate tooth surface attachments and maintain effective cooperation with the attachments during traction, torque control, indentation, or rotation control. Simultaneously, the aligner is typically not inserted along a single vertical direction during wear, but rather through a process involving local elastic opening, incisal edge introduction, slight rotation in the arch direction, and the sequential insertion of multiple attachments. Based on this, combining the attachment groove generation process with the intermediate posture interference in the aligner wearing path can more reasonably explain and address the contradictions of attachment groove final position matching but wearing process stagnation and weakened orthodontic force transmission after overall magnification, thus providing a more realistic modeling basis for the generation of the orthodontic appliance's 3D model.

[0027] It should be noted that during the wearing of orthodontic appliances, if there is a mismatch in the entry angle between the attachment slot and the tooth attachment, localized obstruction of the slot, mid-section jamming in the slot, or the attachment not fully entering the slot, the operator needs to manually grind, locally enlarge the slot, perform thermoforming, or re-fabricate the appliance in the corresponding area. While these methods can improve wearing difficulties to some extent, the correction position and amount largely depend on the operator's experience, which can easily lead to over-enlargement or under-correction of the slot. When the slot is over-enlarged, the attachment slot's wrapping and restraining effect on the tooth attachment is weakened, thus affecting the appliance's traction, torque control, and retention; when under-corrected, jamming or false positioning may still occur.

[0028] Therefore, if the path interference points during the attachment insertion process are identified during the 3D model generation stage, taking into account the wearing path of the orthodontic appliance relative to the dentition, and if introduction structures, avoidance structures, and fit control structures are formed for different interference areas, then the attachment slot can simultaneously ensure both wearing comfort and final fit control without relying on later manual refinishing. Thus, the orthodontic appliance model has already completed pre-correction of process-related jamming positions before entering the manufacturing stage, which helps reduce the probability of rework after trial fitting of the finished product and reduces the risk of insufficient orthodontic force transmission due to overall enlargement of the attachment slot.

[0029] In another typical application, the method of this application can also be applied to the generation of 3D models for fitted protective shells, positioning sleeves, or assembly sleeves. These products typically need to be fitted onto the outside of a target object with an irregular outer surface. The target object's surface may have positioning bosses, limiting protrusions, reinforcing ribs, snap-fit ​​blocks, or functional protrusions. In the final assembled state, the shell needs to maintain a high degree of fit with the target object, while simultaneously forming grooves or clearance cavities at corresponding positions of the local protrusions to ensure the shell can be smoothly fitted and maintain stable positioning.

[0030] In actual modeling, if the local groove is generated solely based on the outer contour of the local protrusion at the final assembly position, the path of the local protrusion as the shell gradually fits into the final position from the opening side can easily be overlooked. For example, the local protrusion may be able to fall into the groove in the final state, but it may first abut against the edge of the groove opening or the middle section of the groove wall during the fitting process. If this abutment is eliminated by enlarging the entire groove, the limiting fit strength between the shell and the local protrusion in the final assembly state may be reduced. By adopting the method of this application, the fitting path of the shell relative to the target object can be simulated in the model generation stage, and the local groove to be generated area can be modeled differently based on the path interference information under multiple intermediate fitting postures, including import, avoidance, and fitting control. This ensures the fitting ability of the shell while preserving the fitting and limiting effects required for the final assembly as much as possible.

[0031] refer to Figure 1 , Figure 1 This is an exemplary application scenario diagram provided for an embodiment of this application.

[0032] like Figure 1As shown, in the application scenario, the 3D model of the target bearing object, the 3D model of the local protrusion structure, and the 3D model of the reference shell can be used as input data for the model generation end. Specifically, the 3D model of the target bearing object is used to characterize the shape of the object to be fitted by the shell; the 3D model of the local protrusion structure is used to characterize the protruding parts set on the surface of the target bearing object that may affect the shell fitting process; and the 3D model of the reference shell is used to characterize the initial shell model before the formation of the local groove structure or before the groove correction is completed.

[0033] After acquiring the aforementioned model data, the model generation end can first perform a fitting path analysis to determine the relative motion process of the reference shell 3D model relative to the target load-bearing object 3D model from the unfitted state to the final fitted state. Since the local protrusion structure can be located inside the shell in the final fitted state, this does not mean that it will not abut or get stuck against the inner surface of the reference shell 3D model during the fitting process. Therefore, the model generation end further analyzes the path interference between the local protrusion structure 3D model and the inner surface of the reference shell 3D model by combining the intermediate fitting postures in the fitting path.

[0034] After completing the fitting path analysis, the model generation end can perform groove generation processing based on the path interference information. That is, a local groove structure is generated in the 3D model of the reference shell at the position corresponding to the 3D model of the local protrusion structure. The local groove structure is not simply obtained by equidistant enlargement or direct subtraction based on the final position of the local protrusion structure. Instead, it combines the entry process of the local protrusion structure in the fitting path and performs differentiated modeling of different regions of the groove. This allows it to provide the necessary entry and avoidance space during the fitting process and retain the required fitting control relationship in the final fitting state.

[0035] Therefore, the model generation end can output a three-dimensional shell model with local groove structure. The three-dimensional shell model can be used for subsequent digital design verification, processing data generation, mold generation, or solid shell manufacturing. Figure 1 This application only illustrates one possible scenario of the method described in this application. The specific forms of the three-dimensional model of the target bearing object, the three-dimensional model of the local protruding structure, and the three-dimensional model of the reference shell do not constitute a limitation on the scope of protection of this application.

[0036] Taking orthodontic appliances as an example, refer to Figure 2 , Figure 2 This is a schematic diagram of the structure of a three-dimensional model provided in an embodiment of this application.

[0037] like Figure 2As shown, in the digital modeling scenario of orthodontic appliances, the 3D model of the dentition can correspond to the aforementioned 3D model of the target bearing object, the 3D model of the attachments can correspond to the aforementioned 3D model of the local protrusion structure, the 3D model of the orthodontic appliance can correspond to the aforementioned 3D model of the reference shell, and the 3D model of the shell can correspond to the aforementioned 3D shell model with the local groove structure.

[0038] Among them, the 3D model of the dentition is used to represent the 3D external surface morphology of the object to be fitted, which in the orthodontic scenario can specifically be the outer surface model of the crown during the target treatment stage; the 3D model of attachments is used to represent the local protruding structures set on the surface of the object to be fitted, which in the orthodontic scenario can specifically be rectangular attachments, inclined attachments, elliptical attachments or other orthodontic attachments set on the tooth surface; the 3D model of treatment is used to represent the reference shell 3D model adapted to the 3D model of the dentition, which in the orthodontic scenario can specifically be the initial aligner shell model that has not yet formed attachment slots for attachment positions or has not yet completed the differential correction of attachment slots; the 3D model of the shell is used to represent the 3D shell model obtained after local slot modeling, which in the orthodontic scenario can specifically be the 3D model of the orthodontic aligner with attachment slots formed on the inner surface.

[0039] Based on the above correspondence, the path interference between the attachment 3D model and the inner surface of the orthodontic 3D model can be analyzed in the fitting path of the orthodontic 3D model relative to the dental arch 3D model. Based on the path interference, an attachment slot corresponding to the attachment 3D model can be formed in the shell 3D model.

[0040] Next, with reference to the accompanying drawings, the method for generating a three-dimensional orthodontic model provided in the embodiments of this application will be further elaborated. Figure 3 The methods shown include: S1: Obtain the three-dimensional model of the target load-bearing object, the three-dimensional model of the local protrusion structure set on the surface of the target load-bearing object, and the three-dimensional model of the reference shell corresponding to the three-dimensional model of the target load-bearing object.

[0041] S2: Based on the final fitting posture of the target bearing object 3D model and the reference shell 3D model, determine the fitting path of the reference shell 3D model relative to the target bearing object 3D model, the fitting path including multiple intermediate fitting postures.

[0042] In this embodiment, the final fitting posture is used to characterize the relative positional relationship between the reference shell 3D model and the target bearing object 3D model when they reach the expected fitting state.

[0043] It is important to note that the fact that the shell can finally fit into place does not equate to the absence of local jamming during the fitting process. For thin-walled shells, elastic shells, or shells with high fit, the shell typically needs to undergo processes such as oblique entry, partial opening, screwing in, or gradual pressing before reaching the final fitting state. The space occupied by local protrusions during the intermediate processes often differs from their final positions. Based on this understanding, the fitting path of the reference shell 3D model from its unfinished to its final fitting state can be determined by reverse engineering the final fitting posture, and multiple intermediate fitting postures can be set along the fitting path. The intermediate fitting postures can be obtained through translation interpolation, rotation interpolation, or a combination of translation and rotation interpolation, and can also be adjusted in conjunction with the shell opening direction, the extension direction of the target load-bearing object, and the distribution position of local protrusions.

[0044] S3: Under each of the intermediate fitting postures, detect the path interference information between the three-dimensional model of the local protrusion structure and the inner surface of the three-dimensional model of the reference shell, and determine the introduction area, avoidance area and fitting control area in the local groove to be generated area corresponding to the three-dimensional model of the local protrusion structure according to the path interference information.

[0045] In this embodiment, the path interference information may include at least the path interference position, the path interference depth, and the corresponding intermediate fitting posture. Specifically, during detection, the distance relationship between the 3D model of the local protruding structure and the inner surface of the 3D model of the reference shell can be determined under each intermediate fitting posture. When the distance between them is less than a preset safety gap, the corresponding position is identified as the path interference position, and the path interference depth is determined based on the degree of insufficient safety gap. The preset safety gap can be determined by combining model scanning error, mesh reconstruction error, manufacturing error, and the allowable elastic deformation of the shell. For example, in orthodontic appliance design, the safety gap range can be determined by comprehensively considering intraoral scanning error, appliance forming error, and accessory processing error. Furthermore, different interference positions have different functions in the local groove. Areas located on the fitting entry side and interfering earlier are more suitable for forming an introductory area; areas interfering midway through fitting but not ultimately requiring tight fitting are more suitable for forming an avoidance area; and areas requiring fitting or limitation during final fitting are more suitable for forming a fitting control area.

[0046] S4: Based on the import area, the avoidance area and the fitting control area, perform differentiated modeling on the local groove to be generated area to generate a three-dimensional shell model with local groove structure.

[0047] In this embodiment, differentiated modeling can be understood as using different geometric processing methods for different functional areas within the local tank to be generated area.

[0048] It is important to emphasize that avoidance does not simply mean enlarging the entire slot, but rather releasing only the space in the fitting path that actually causes procedural jamming, while preserving the contact control relationship required for final fitting. Taking orthodontic appliances as an example, the accessory slot can thus reduce the probability of accessory jamming during wear and maintain the ability to wrap, pull, and control torque on the accessories in the final wearing state, thereby reducing the loss of precision caused by manual grinding, local slot enlargement, or re-fabrication.

[0049] Before elaborating on the specific processing procedures involved in the embodiments of this application, it should be noted that for three-dimensional shell models with high fitting requirements, the local groove cannot be simply understood as a static recess on the inner surface of the shell used to accommodate the protruding structure. Especially when the shell thickness is thin, the opening edge constraint is strong, and the outer surface of the target load-bearing object has continuous curvature changes, the spatial relationship between the local protruding structure and the inner surface of the shell will continuously change when the shell moves from a not fully fitted state to the final fitted state. Even if the local protruding structure can fall into the preset groove in the final fitted posture, it may first locally abut against the groove opening, groove wall, or groove bottom transition area in the intermediate fitted posture. Therefore, the rationality of the local groove model depends not only on the geometric matching in the final posture, but also on whether there is a continuous space for the local protruding structure to pass through during the fitting process.

[0050] Conventional modeling methods typically use the final fitting posture as the sole benchmark, projecting the outer contour of the local protrusion onto the inner surface of the shell, and then forming the groove through equidistant offsets, Boolean subtraction, or uniform gap expansion. This method can meet basic modeling requirements when the protrusion height is small, the fitting path is approximately straight, and the shell allows for large deformation. However, when the local protrusion has a certain outward protrusion height, or when the shell needs to be fitted gradually along the oblique, arc, or screw-in direction, the groove contour in the final posture cannot reflect the sweeping area occupied by the protrusion during the entry process. If the groove is enlarged as a whole to avoid jamming, the area in the groove that should bear the functions of fitting, limiting, or holding will also be weakened; if a small gap is still maintained, fitting difficulties will easily be exposed after processing or wearing. The essence of the above contradiction is that traditional groove generation only considers the final accommodating space, without distinguishing between the process passage space and the final holding space.

[0051] Based on this understanding, this application's embodiments divide the area to be generated for the local groove into regions with different modeling attributes. The entry region corresponds to the area the local protrusion structure first approaches when entering the groove, primarily used to improve entry smoothness; the avoidance region corresponds to the area where the local protrusion structure temporarily occupies space and may cause contact during the intermediate stages of the fitting path, primarily used to release procedural interference; the fitting control region corresponds to the area where the fitting or limiting relationship still needs to be maintained in the final fitting state, primarily used to prevent the local groove from losing control due to excessive expansion. These partitions are not directly specified based on human experience, but are derived from the distance changes and interference positions between the local protrusion structure and the inner surface of the shell under multiple intermediate fitting postures. Therefore, the generation of the local groove can be transformed from static contour subtraction to spatial constraint modeling oriented towards the fitting path.

[0052] Furthermore, the avoidance process in this embodiment is not equivalent to enlarging the entire hole of the local groove. A more realistic approach to the actual mating behavior is to use the area occupied by the local protrusion structure in the relevant intermediate mating posture as the path sweeping area, and to determine the candidate avoidance space by combining the path interference depth; simultaneously, the mating control area is used as the avoidance boundary constraint to limit the intrusion of the candidate avoidance space into the final holding area. In this way, the local groove can provide the necessary passage margin for the mating process without compromising the mating accuracy in the final mating state due to blind enlargement.

[0053] Therefore, the technical focus of this application's embodiments is not simply to generate a groove with a shape similar to the local protrusion structure, but rather to utilize the intermediate posture interference relationship in the fitting path to assign different geometric functions to different areas of the local groove. For subsequent model manufacturing or assembly, the resulting three-dimensional shell model has already pre-corrected the possible procedural abutment positions before entering solid processing, while retaining the fitting control area required for final fitting, thus enabling the local groove to simultaneously satisfy both accessibility and retention constraints.

[0054] Next, we will further elaborate on the technical content of the three-dimensional model in this application.

[0055] It should be noted that the target load-bearing object 3D model, the local protrusion structure 3D model and the reference shell 3D model in this application are all pre-constructed 3D models. The modeling data corresponding to the target load-bearing object 3D model and the local protrusion structure 3D model can be collected by 3D scanning, structured light scanning, laser scanning, photogrammetry or computer-aided design data import, while the reference shell 3D model can be constructed based on the target load-bearing object 3D model.

[0056] Those skilled in the art will understand that the specific modeling method for the 3D model can employ existing 3D modeling methods such as point cloud reconstruction, mesh reconstruction, surface fitting, solid modeling, parametric modeling, or Boolean operation modeling. The key is to obtain the 3D geometric data used to represent the outer surface of the target load-bearing object, the outer surface of the local protruding structure, and the inner surface of the reference shell. This application does not further limit the specific modeling software, modeling format, or model generation algorithm. For example, the 3D model can use STL, OBJ, PLY, STEP, or other data formats that can be used to represent 3D geometric shapes.

[0057] In one optional implementation, the 3D model of the target bearing object can be obtained by first denoising, filling holes, and reconstructing the surface of the acquired point cloud data. The 3D model of the local protruding structure can be imported and generated according to the preset installation position on the surface of the target bearing object, and a unified coordinate relationship is established with the 3D model of the target bearing object. The 3D model of the reference shell can be obtained by offsetting, thickening, and boundary trimming the outer surface of the 3D model of the target bearing object, so that the reference shell 3D model has an inner surface contour corresponding to the 3D model of the target bearing object.

[0058] Next, we will further elaborate on the technical content of the method of this application regarding the region to be generated in the local tank.

[0059] It should be noted that the region to be generated for the local groove can be understood as a candidate modeling area on the inner surface of the reference shell 3D model that has a spatial correspondence with the 3D model of the local protrusion structure and is used for subsequent generation of the local groove structure. In other words, the region to be generated for the local groove is not the already formed groove itself, but rather the groove generation range determined in advance before the division of the import area, avoidance area, and fitting control area. By determining the region to be generated for the local groove in advance, the subsequent path interference detection results can be limited to the inner surface of the shell that is actually related to the local protrusion structure, avoiding unnecessary geometric corrections to other irrelevant areas of the reference shell 3D model.

[0060] Specifically, the area to be generated for the local groove can be determined based on the outer contour of the 3D model of the local protrusion structure, the outward convex direction of the 3D model of the local protrusion structure relative to the 3D model of the target load-bearing object, and the surface transition area on the periphery of the 3D model of the local protrusion structure. The outer contour of the 3D model of the local protrusion structure defines the basic coverage area of ​​the local groove on the inner surface of the shell; the outward convex direction determines the main protrusion direction of the local protrusion structure relative to the surface of the target load-bearing object, ensuring that the area to be generated for the groove covers the direction area where the local protrusion structure may make contact when it enters the inner surface of the shell; the surface transition area on the periphery reflects the connection boundary between the local protrusion structure and the outer surface of the target load-bearing object, preventing the groove boundary from being generated only along the outer contour of the local protrusion structure, which could result in an excessively narrow groove opening or abrupt edge changes.

[0061] In one optional implementation, the connection boundary between the 3D model of the local protrusion structure and the 3D model of the target load-bearing object can be identified first, and the base contour of the local protrusion structure can be obtained based on the connection boundary. Then, the outer contour envelope of the 3D model of the local protrusion structure can be determined based on the top contour, sidewall contour, and base contour of the local protrusion structure. Subsequently, the outward convex direction of the 3D model of the local protrusion structure can be determined based on the local surface normal of the 3D model of the target load-bearing object near the connection boundary, or based on the direction from the base center of the local protrusion structure to the top center. After mapping the outer contour envelope along the outward convex direction and its adjacent tangential direction to the inner surface of the reference shell 3D model, an initial candidate region can be obtained. Then, combined with the surface transition region within a preset range around the local protrusion structure, the initial candidate region is expanded and the boundary is smoothed to obtain the local groove to be generated region.

[0062] Furthermore, the expansion range of the area to be generated for the local slot can be determined based on the protrusion height of the local protrusion structure, model errors, and shell manufacturing errors. For example, the greater the protrusion height of the local protrusion structure, the easier it is to generate abutment at the slot opening or wall during fitting, and the area to be generated for the local slot can be increased accordingly towards the entry side and sidewall. If the 3D model is derived from scan reconstruction, the expansion range can also be determined by combining scan errors, mesh reconstruction errors, and subsequent processing errors. For example, in the orthodontic appliance scenario, when the protrusion height of the tooth surface attachment is 0.8mm to 1.2mm, the area to be generated for the attachment slot can be expanded circumferentially by 0.10mm to 0.30mm based on the outer contour projection range of the attachment; when the protrusion height of the attachment is greater than 1.2mm or the attachment is located in an area with a large change in crown curvature, the expansion range can be increased to 0.30mm to 0.50mm. The above values ​​are only used to illustrate one possible implementation method, and those skilled in the art can adjust them according to model accuracy, material forming errors, and allowable gaps.

[0063] Taking orthodontic appliance application scenario as an example, the target support object's 3D model is the dentition 3D model, the local protruding structure's 3D model is the attachment 3D model set on the tooth surface, and the reference shell 3D model is the appliance's reference shell model. In this case, the area to be generated for the local slot can be the region corresponding to the attachment's 3D model on the inner surface of the appliance's reference shell model. Its determination process can include: extracting the bottom edge contour and outer surface contour of the attachment's 3D model on the tooth surface, determining the attachment's outward convex direction relative to the tooth surface; mapping the attachment's outer contour to the inner surface of the appliance's reference shell model according to the final fitting relationship of the appliance; and appropriately expanding the mapping result by combining the transition area of ​​the attachment's peripheral tooth surface to form the area to be generated for the attachment slot. The resulting area to be generated for the attachment slot not only covers the space required for the final placement of the attachment but also reserves the basic range for subsequent import, avoidance, and fit control modeling based on wearing path interference.

[0064] Next, we will further elaborate on the technical content of the fitting path in this application.

[0065] In one example, the method for determining the fitting path includes: S2.1: Based on the fitting relationship between the target bearing object 3D model and the reference shell 3D model, determine the final fitting posture of the reference shell 3D model relative to the target bearing object 3D model.

[0066] Specifically, the final fitting posture can be understood as the target pose of the reference shell 3D model relative to the target carrier 3D model after fitting. It is not an arbitrarily chosen relative position, but rather determined by the fit relationship between the inner surface of the reference shell 3D model and the outer surface of the target carrier 3D model. For shell models requiring the generation of local groove structures, only by first determining where the shell should ultimately be located can it be determined whether the local protrusions will prematurely contact the inner surface of the shell during entry. In other words, the final fitting posture provides the endpoint reference of the fitting path and also the spatial reference for determining whether the fitting control area needs to be retained. If the final fitting posture is not accurately determined, the subsequently generated entry and avoidance areas, while capable of releasing procedural interference, may deviate from the final fitting position that the shell should maintain.

[0067] As can be seen from the foregoing embodiments, the reference shell three-dimensional model is obtained by offsetting, thickening and boundary trimming the target bearing object three-dimensional model. Therefore, the two usually have an initial coordinate correspondence, which can be used as the initial estimate of the final fitting posture.

[0068] Furthermore, multiple fitting detection points are selected on the inner surface of the reference shell 3D model, and the distance from each fitting detection point to the outer surface of the target load-bearing object 3D model is calculated. When the average fitting gap, the local maximum fitting gap, and the shell opening edge deviation all meet the preset fitting conditions, the corresponding relative pose is determined as the final fitting posture. The preset fitting conditions can be determined based on model acquisition error, model reconstruction error, processing error, and the allowable small deformation of the shell material. For example, when the target load-bearing object 3D model is reconstructed from scanned data, the scanning error can be taken as 0.03 mm, the mesh reconstruction error can be taken as 0.02 mm, and the shell forming or printing error can be taken as 0.05 mm. After the three are superimposed, the average fitting gap threshold can be set to 0.10 mm to 0.15 mm. For areas with large local curvature changes, the local maximum fitting gap threshold can be widened to 0.25 mm to 0.30 mm to avoid misjudging normal gaps caused by sudden changes in model curvature as non-fitting.

[0069] S2.2: Determine the initial fitting posture of the reference shell three-dimensional model based on the main body extension direction and contour extension direction of the target bearing object three-dimensional model, the opening edge direction of the reference shell three-dimensional model, and the distribution position of the local protrusion structure three-dimensional model in the target bearing object three-dimensional model.

[0070] Specifically, the initial fitting posture characterizes the initial relative position of the reference shell 3D model before it is fully fitted onto the target load-bearing object 3D model. It, along with the final fitting posture, defines both ends of the fitting path. For shell-type structures, the fitting process is typically not completed by direct pressing along a single normal direction. The target load-bearing object may have curved contours, inclined outer surfaces, or locally narrowed areas, and the reference shell 3D model may also have opening edges, thin-walled boundaries, and local elastic deformation constraints. If the initial fitting posture is generated simply by outward movement, the shell's entry path may differ from the actual assembly or wearing method, causing subsequent intermediate fitting postures to fail to accurately reflect the true spatial path of locally protruding structures.

[0071] In this embodiment, the main body extension direction and the contour extension direction can be extracted first from the three-dimensional model of the target bearing object. The main body extension direction can be determined by the main axis direction, centerline direction, or the direction of the line connecting the center points of multiple sections of the target bearing object, and is used to characterize the direction in which the target bearing object extends from one end to the other. The contour extension direction can be determined by the outer contour boundary, arc boundary, or opening projection boundary, and is used to characterize the main direction of the outer periphery of the target bearing object. For the reference shell three-dimensional model, its opening edge direction can be extracted. Specifically, the boundary points on the shell opening boundary can be identified, and the opening orientation and opening tangent can be determined based on the boundary ring formed by the boundary points. Subsequently, the reference shell three-dimensional model is moved outward from the final fitting posture in the opposite direction of the opening orientation, so that an initial separation distance is formed between the shell opening edge and the outer surface of the target bearing object. The initial separation distance can be determined based on the maximum outward convex height of the local protrusion structure three-dimensional model, the thickness of the shell opening edge, and the safety margin. For example, when the maximum outward protrusion height of the local protrusion structure is 1.2 mm, the thickness of the shell opening edge is 0.75 mm, and the safety margin is 0.30 mm, the initial separation distance can be set to not less than 2.25 mm; if the maximum outward protrusion height of the local protrusion structure is less than 0.6 mm and the distribution is small, the initial separation distance can be set to 1.2 mm to 1.8 mm; if the maximum outward protrusion height of the local protrusion structure is greater than 1.5 mm or located in an area where the target load-bearing object is significantly curved, the initial separation distance can be increased to 2.5 mm to 3.5 mm.

[0072] S2.3: Based on the relative pose change between the initial fitting posture and the final fitting posture, generate the fitting path of the reference shell 3D model relative to the target bearing object 3D model.

[0073] Specifically, the fitting path is used to characterize the continuous relative motion process of the reference shell 3D model as it gradually transitions from the initial fitting posture to the final fitting posture. Relative pose changes include not only spatial translation but also rotational changes. For models with many local protrusions or narrow shell opening edges, the impact of rotational changes on path interference is often no less than that of translational changes. If the fitting path is represented only by a linear translation between the initial and final positions, the actual sweep range of the local protrusions on the slot opening and wall will be compressed, and the avoidance area may also be incorrectly reduced.

[0074] It should be noted that the specific data form of the relative pose change can be obtained by calculating the difference after establishing pose description data for the initial fitting posture and the final fitting posture respectively. The pose description data can include position data and attitude data. The position data can be represented by the coordinates of a preset reference point in the three-dimensional coordinate system of the reference shell three-dimensional model. The preset reference point can be the geometric center, the center of the opening edge, the center of the inner surface envelope of the shell, or the center point obtained by fitting multiple positioning feature points. The attitude data can be represented by rotation matrix, Euler angle, rotation vector, or quaternion, as long as it can characterize the orientation change of the reference shell three-dimensional model relative to the target bearing object three-dimensional model. This application does not impose further limitations on the specific expression format of the attitude data.

[0075] In this embodiment, the coordinate system of the target bearing object's 3D model can be used as a reference coordinate system. The first pose description data of the reference shell 3D model in the initial fitting posture and the second pose description data in the final fitting posture are recorded. Subsequently, the translational change is obtained based on the positional difference between the first pose description data and the second pose description data; the rotational change is obtained based on the posture difference between the first pose description data and the second pose description data.

[0076] For example, when the reference point of the 3D model of the base shell deviates by 2.40 mm from the final fitting posture along the opening direction in the initial fitting posture, and there is a 5° posture difference around the contour extension direction of the target bearing object, the 2.40 mm can be used as the translational change and the 5° as the rotational change, for subsequently generating multiple intermediate fitting postures that gradually transition from the initial fitting posture to the final fitting posture. The above relative pose changes can be directly read from the pose transformation record in the modeling software, or calculated by rigidly registering the corresponding feature points, boundary points, or surface sampling points in the initial and final fitting postures.

[0077] In this embodiment, the relative pose change between the initial and final fitting postures can be decomposed to obtain translational and rotational changes. The translational change can be used to determine the direction and distance of the shell opening towards the target load-bearing object, while the rotational change can be used to determine whether the shell screws in, straightens, or tilts back during the approach process. Then, a fitting path is constructed based on the translational and rotational changes. For the translational portion, interpolation can be performed along the displacement direction from the initial to the final fitting posture; for the rotational portion, Euler angle interpolation, quaternion interpolation, or angle interpolation based on the rotation axis can be used to ensure a continuous transition in the rotational changes of each intermediate fitting posture. The spacing between intermediate fitting postures can be adapted to the accuracy of subsequent path interference detection. For example, for models with no more than three local protrusions and a maximum protrusion height of no more than 1.0 mm, the translation sampling interval can be set to 0.25 mm to 0.35 mm, and the rotation sampling interval can be set to 2° to 3°. For models with more than three local protrusions or a maximum protrusion height greater than 1.0 mm, the translation sampling interval can be tightened to 0.10 mm to 0.25 mm, and the rotation sampling interval can be tightened to 1° to 2°. The sampling interval can be determined based on the model error and a preset safety gap. For example, when the overall model error is 0.10 mm, the translation sampling interval should not be greater than 0.30 mm to avoid adjacent poses crossing narrow interference positions and not being detected.

[0078] Next, we will further elaborate on the technical aspects of path partitioning in this application.

[0079] It should be noted that the import area, avoidance area, and fitting control area in this application correspond to areas with different geometric functions in the local groove to be generated area, and are not fixed areas pre-specified on the 3D model. For the local groove, the area near the groove opening, the middle section of the groove wall, and the final fitting surface play different roles in the fitting process: the area near the groove opening is more likely to affect whether the local protrusion structure can enter the groove smoothly; the middle section of the groove wall is more likely to generate short-term abutment during the relative movement of the shell; and the final fitting surface is related to whether the shell can still form a stable covering, limit, or retain the local protrusion structure after the shell is fitted.

[0080] In this application, path interference information can be understood as a spatial record formed when insufficient spacing, increased contact tendency, or local abutment occurs between the 3D model of a local protruding structure and the inner surface of the 3D model of the reference shell and the target load-bearing object during the process of the reference shell 3D model gradually approaching the target load-bearing object 3D model along the fitting path. Path interference information may include path interference position, path interference depth, corresponding intermediate fitting posture, the local groove region where the interference position is located, and the stage at which the interference occurs. Among them, the path interference position is used to characterize the specific location of concern in the local groove region to be generated; the path interference depth is used to characterize the degree of insufficiency relative to the preset safety gap; and the corresponding intermediate fitting posture is used to characterize the sequence of stages in which the interference occurs in the fitting path. The preset safety clearance can be determined based on the 3D scanning error, mesh reconstruction error, model registration error, and shell manufacturing error. For example, if the scanning error is 0.03mm, the mesh reconstruction error is 0.02mm, the registration error is 0.03mm, and the manufacturing error is 0.05mm, then the preset safety clearance can be set to about 0.13mm. When the model accuracy is high or the shell material has a large elastic margin, it can be 0.10mm to 0.15mm. When the curvature of the model surface changes greatly or the processing error is high, it can be 0.15mm to 0.25mm.

[0081] In one example, the path interference information is determined in the following ways: For each intermediate fitting posture, determine the relative position between the three-dimensional model of the local protrusion structure and the inner surface of the three-dimensional model of the reference shell; The relative position is obtained by placing the three-dimensional model of the local protrusion structure and the three-dimensional model of the reference shell in the same coordinate system according to the corresponding intermediate overlay posture, and by performing a nearest distance search between the sampling points on the outer surface of the three-dimensional model of the local protrusion structure and the inner surface of the three-dimensional model of the reference shell. Specifically, multiple surface sampling points can be selected on the outer surface of the three-dimensional model of the local protrusion structure. For each surface sampling point, the nearest corresponding surface point is found on the inner surface of the reference shell three-dimensional model. The relative position between the three-dimensional model of the local protrusion structure and the inner surface of the reference shell three-dimensional model is determined based on the distance between the surface sampling point and the corresponding surface point, the direction of the connecting line, and the inner surface normal at the corresponding surface point.

[0082] When the distance between the three-dimensional model of the local protrusion structure and the inner surface of the three-dimensional model of the reference shell is less than the preset safety gap, the corresponding position is determined as the path interference position; The method for determining the spacing includes: Multiple protruding surface sampling points are selected on the outer surface of the three-dimensional model of the local protruding structure, and the nearest inner surface point corresponding to each protruding surface sampling point is determined on the inner surface of the three-dimensional model of the reference shell. Based on the spatial distance between each sampling point on the raised surface and its corresponding nearest inner surface point, multiple candidate spacings are obtained; The distance between the three-dimensional model of the local protrusion structure and the inner surface of the three-dimensional model of the reference shell is determined based on the minimum value among the multiple candidate distances. When the direction of the line connecting the sampling point on the protruding surface corresponding to the minimum value and the nearest inner surface point satisfies the normal constraint of the inner surface of the reference shell three-dimensional model, the minimum value is taken as the effective spacing.

[0083] In this embodiment, the sampling points on the raised surface can be determined according to the vertices of the triangular mesh, the center points of the facets, or the equidistant resampling points of the 3D model of the local raised structure. For local raised structures with small changes in surface curvature, mesh vertices can be directly used as sampling points on the raised surface; for local raised structures with inclined surfaces, rounded corners, or abrupt changes in local curvature, facet center points or edge midpoints can be added to the original mesh vertices to avoid missing the top abutment position of the groove wall in sparse meshes. The nearest inner surface point can be determined through nearest neighbor search, bounding box accelerated search, or spatial indexing structure, as long as the nearest distance from the sampling point on the raised surface to the inner surface of the reference shell 3D model can be obtained.

[0084] Furthermore, normal constraints are used to exclude back distances that are unrelated to mating interference. For example, based on the inner surface normal at the nearest inner surface point, it can be determined whether the line connecting the sampling point on the protruding surface to the nearest inner surface point is oriented towards the accessible side of the inner surface of the shell; when the direction of the line is consistent with the direction of the accessible side, the candidate spacing can be used as the effective spacing; when the direction of the line points to the outside of the shell or crosses a non-corresponding surface, it can be excluded from the path interference judgment criteria.

[0085] The corresponding path interference depth is determined based on the difference between the path interference position and the preset safety gap. The path interference position, the path interference depth, and the corresponding intermediate overlay posture are determined as the path interference information.

[0086] In yet another example, the determination of the import area, avoidance area, and fitting control area includes: S3.1: Based on the path interference positions in each of the path interference information, merge the path interference positions that are located within the local groove to be generated area and are adjacent to each other into candidate interference areas.

[0087] Specifically, path interference positions typically exist as discrete points, local patches, or sets of sampling points. Directly partitioning the local trench area based on a single path interference position is susceptible to the effects of 3D mesh density, surface sampling noise, and model registration errors. Subsequent modeling of the local trench requires continuously processable surface regions, not isolated points. Therefore, after obtaining path interference information under multiple intermediate overlay postures, the path interference positions can be mapped to the inner surface of the shell containing the local trench area. Then, using the local trench area as a boundary constraint, the path interference positions falling within this area can be aggregated.

[0088] In this embodiment, path interference positions located outside the area to be generated in the local groove can be first eliminated, and then merged according to the spatial distance between the path interference positions, the adjacency relationship of the mesh on the inner surface of the shell, and the adjacency relationship of the corresponding intermediate fitting postures. Specifically, the side length of the triangular mesh in the area to be generated in the local groove can be used as a scale reference to determine the adjacency judgment distance; when the spatial distance between two path interference positions is not greater than the preset adjacency distance, and there is a direct adjacency relationship between the two faces or a relationship that can be connected through a few intermediate faces, they can be considered to be spatially adjacent. The preset adjacency distance can be determined according to the average mesh side length and model error of the area to be generated in the local groove. For example, if the average mesh side length of the area to be generated in the local groove is 0.08mm, and the total scanning error and mesh reconstruction error is about 0.05mm, then the preset adjacency distance can be about 0.24mm, or it can be selected in the range of 0.20mm to 0.30mm.

[0089] It should be noted that although the rejected path interference positions may still have some impact on the local contact state between the shell and the local protrusion structure during the actual mating process—for example, in cases of local material springback, superimposed manufacturing errors, or large deviations in manual assembly angles—they may still manifest as slight scratches, short-term contact, or local obstruction. However, since the focus of this solution is to generate local groove structures based on path interference information that appears stably and has regional continuity, directly incorporating path interference positions that only appear in a single intermediate mating posture, have an interference depth close to the model noise level, or do not form a continuous interference region into the groove correction could easily cause unnecessary indentations or excessive avoidance in the local groove, thus weakening the fitting control relationship in the final mating state. Therefore, in this embodiment, further modeling processing of such rejected path interference positions is temporarily omitted.

[0090] S3.2: Determine the interference occurrence stage corresponding to each candidate interference region based on the arrangement order of the intermediate overlay postures corresponding to each path interference information in the overlay path.

[0091] Specifically, the candidate interference regions, even within the same local groove to be generated area, have different geometric meanings depending on the stage of the mating process. Interference appearing near the beginning of the mating path is often related to the groove opening abutment when the local protrusion just enters the local groove; interference appearing in the middle of the mating path often corresponds to the process-related jamming caused by the local protrusion passing along the groove wall or inside the groove; and approaching or fitting closer to the final mating posture may be related to the final holding, limiting, or fitting control of the shell.

[0092] In this embodiment, intermediate fitting postures can be numbered sequentially according to their order from furthest from the final fitting posture to closest to it. For each candidate interference region, the intermediate fitting posture numbers corresponding to the interference positions along the path can be counted, and the interference occurrence stage of the candidate interference region can be determined based on the posture number where interference first occurs, the range of postures during which interference lasts, and the posture interval with the largest interference depth. If a candidate interference region experiences interference in the early part of the fitting path and the corresponding interference depth gradually increases as it enters the local groove, its occurrence stage can be considered to be biased towards the early part; if a candidate interference region mainly appears in the middle part of the fitting path and gradually disappears near the final fitting posture, its occurrence stage can be considered to be biased towards the middle part; if a candidate interference region remains close to the end of the fitting path, it is necessary to further determine whether it belongs to the fitting control-related area based on the final fitting position.

[0093] Furthermore, the division of the mating path into the front, middle, and final stages can be determined based on the total number of intermediate mating postures. For example, when the mating path includes 20 intermediate mating postures, the stages corresponding to the 1st to 7th intermediate mating postures can be considered the front stage, the stages corresponding to the 8th to 17th intermediate mating postures can be considered the middle stage, and the stages corresponding to the 18th to 20th intermediate mating postures can be considered the final approach stage. If the total number of intermediate mating postures is small, the division can also be based on a ratio of 35% for the front, 50% for the middle, and 15% for the rear. If the protrusion height of a local protrusion is large, the front entry process is more sensitive to jamming, and the proportion of the front stage can be adjusted to around 40%. If the shell opening is wide and the entry process is relatively gentle, the proportion of the front stage can be controlled at around 30%. These proportions can be selected based on the shell opening edge width, the protrusion height of the local protrusion, and the sampling density of the mating path, as long as the initial entry, process passage, and final approach stages can be stably distinguished.

[0094] S3.3: The candidate interference region located at the front of the mating path during the corresponding interference occurrence stage is determined as the introductory region, and the candidate interference region located at the middle of the mating path during the corresponding interference occurrence stage is determined as the avoidance region. The mating control region is determined based on the mating position of the three-dimensional model of the local protrusion structure and the inner surface of the three-dimensional model of the reference shell under the final mating posture.

[0095] Specifically, the division into the introductory zone, the avoidance zone, and the fitting control zone essentially assigns different geometric processing attributes to the local groove area to be generated. The candidate interference region, located at the beginning of the fitting path, typically corresponds to the position where the local protrusion is first restricted when entering from the groove entrance side. This is designated as the introductory zone, where flared openings, oblique transition surfaces, or rounded guide surfaces can be formed during subsequent modeling, providing a smoother introductory space for the local protrusion. The candidate interference region, located in the middle of the fitting path, typically corresponds to the procedural resistance that occurs when the local protrusion has entered the vicinity of the groove but has not yet reached the final fitting position. This is designated as the avoidance zone, where local avoidance cavities can be formed during subsequent modeling to release the temporary space occupied during the intermediate fitting stage. The fitting control zone, however, cannot be simply determined based on the interference occurrence stage. It needs to be determined by combining the correspondence between the local protrusion and the inner surface of the shell under the final fitting posture, to avoid mistakenly treating the final fitting position as an avoidance space.

[0096] It should be noted that the specific division of the front and middle sections of the fitting path can be determined by referring to the arrangement order of the intermediate fitting postures in the fitting path. Specifically, the process of the reference shell 3D model gradually approaching the final fitting posture from a position away from it can be used as the sorting direction, and the fitting path can be divided into stages according to the sequence number of each intermediate fitting posture or the path progress. For example, when the fitting path includes 20 intermediate fitting postures, the path range corresponding to the 1st to 7th intermediate fitting postures can be determined as the front section of the fitting path, the path range corresponding to the 8th to 17th intermediate fitting postures can be determined as the middle section of the fitting path, and the path range corresponding to the 18th to 20th intermediate fitting postures can be considered as the final section approaching the final fitting posture. The above division method can also be determined proportionally, for example, the first 30% of the fitting path approaching the initial entry process can be determined as the front section of the fitting path, the next 30% to 60% can be determined as the middle section of the fitting path, and the remaining part approaching the final fitting posture can be considered as the final section.

[0097] Next, we will further elaborate on the technical content of the method of this application regarding the three-dimensional shell model.

[0098] It should be noted that, in the specific technical content corresponding to the foregoing embodiments, this application has already completed the determination of the local groove to be generated area, the construction of the fitting path, the identification of path interference information, and the division of the introduction area, the avoidance area, and the fitting control area. Therefore, it is necessary to further perform differentiated modeling of the above three areas in conjunction with the three-dimensional model of the local protrusion structure, so as to release the entry interference and process abutment of the local protrusion structure during the fitting process, while retaining the fitting, limiting, or holding effect of the inner surface of the shell on the local protrusion structure in the final fitting posture.

[0099] In one example, the differentiated modeling method for the local groove to be generated area corresponding to the import area includes: determining the entry direction and path interference depth of the three-dimensional model of the local protrusion structure when it enters the local groove to be generated area based on the path interference information corresponding to the import area; determining the groove expansion amount of the import area based on the path interference depth; and expanding the groove boundary corresponding to the import area according to the groove expansion amount.

[0100] Specifically, the differentiated modeling of the inlet area described in this application refers to reconstructing the groove geometry at the location corresponding to the inlet area on the inner surface of the reference shell 3D model, so that the local groove forms a flared structure or transition surface that can guide entry on the side where the local protrusion structure enters. In other words, the inlet area modeling does not change the 3D model of the target bearing object or the 3D model of the local protrusion structure, nor does it uniformly enlarge the entire local groove, but rather performs local geometric correction on the inlet boundary of the local groove on the reference shell 3D model. The corrected geometry may include the expanded groove boundary, the inlet groove surface that transitions from the expanded boundary to the inner boundary of the groove, and the inlet surface patch used to replace the original local area of ​​the inner surface of the reference shell 3D model. The inlet area is represented in the 3D model as a directional groove inlet, rather than a simple enlarged pit.

[0101] In this embodiment, the entry direction of the local protrusion structure 3D model into the local groove to be generated area can be determined first based on the path interference information corresponding to the import area. The entry direction can be determined by the positional changes of the local protrusion structure 3D model relative to the local groove to be generated area under two or more adjacent intermediate nesting postures. For example, by selecting the intermediate nesting posture of the local protrusion structure 3D model before the initial interference and the intermediate nesting posture when the first path interference occurs, and comparing the positional changes of the outer contour center or surface sampling point center of the local protrusion structure 3D model under the two postures, the approximate direction of the local protrusion structure entering the local groove to be generated area can be obtained. For cases where the surface morphology of the local protrusion structure is irregular, several surface sampling points where interference occurs can also be selected, and the movement trend of these sampling points in adjacent intermediate nesting postures can be statistically analyzed. The direction with the more concentrated movement trend is taken as the entry direction. The entry direction obtained in this way can reflect which side the local protrusion structure actually approaches the groove opening from, and the subsequent groove expansion can be carried out along the entry side, without having to uniformly expand the entire groove circumferentially.

[0102] Furthermore, the path interference depth can be used to determine the extent to which the slot needs to be expanded. Specifically, the path interference depths corresponding to multiple path interference positions can be statistically analyzed within the import area, and the slot expansion amount can be determined based on the maximum path interference depth, the average path interference depth, or a preset quantile depth. To avoid over-expansion caused by individual outliers, a higher quantile depth can be prioritized as the expansion basis. For example, the depth value located at the 80th to 90th percentile position after sorting the path interference depths within the import area can be selected, and then a safety margin can be added to obtain the slot expansion amount. The safety margin can be determined by combining model acquisition error, mesh reconstruction error, and manufacturing error. For example, if the multiple path interference depths within the import area are 0.06mm, 0.08mm, 0.10mm, 0.12mm, and 0.18mm, where 0.18mm may come from local noise or edge sharp points, 0.12mm can be selected as the reference interference depth; if the overall model error is approximately 0.05mm, the slot expansion amount can be determined to be 0.17mm. For orthodontic appliance accessory slots, the slot expansion can usually be controlled within the range of 0.10mm to 0.35mm. When the accessory protrudes significantly or the entry angle is steep, it can be appropriately increased to about 0.40mm, but it should not be directly expanded to the fit control area to avoid weakening the accessory slot's retention effect on the accessory in the final wearing state.

[0103] In the specific modeling process, the original slot boundary corresponding to the import area can be extracted first. The original slot boundary can be the boundary line of the local slot area to be generated near the entry side of the local protrusion structure, or it can be a spatial curve fitted by several boundary points. Then, the original slot boundary is moved by the slot expansion amount along the opposite direction of the entry direction or along the tangential outward expansion direction of the inner surface of the shell to obtain the expanded slot boundary. For the inner surface of the shell with relatively gentle curvature, the boundary can be directly expanded along the local tangential plane; for the inner surface of the shell with large curvature changes, the boundary can be expanded first along the geodesic direction of the inner surface of the shell or the propagation direction of adjacent mesh patches, and then the expanded boundary points are projected back to the modelable area near the inner surface of the shell. The expanded slot boundary obtained in this way can fit the original shell surface curvature, reducing the occurrence of hanging, folding, or local abrupt changes in the slot boundary in the 3D model.

[0104] Furthermore, after obtaining the extended slot boundary, an introduction transition surface can be generated between the extended slot boundary and the local tank interior boundary. The introduction transition surface can be constructed using surface fitting, spline surfaces, mesh reconstruction, or multi-segment smooth surface patch splicing, as long as it allows the extended slot to gradually transition into the local tank interior space. The slope or transition length of the introduction transition surface can be determined based on the slot extension amount and the available length of the introduction area along the entry direction. For example, when the slot extension amount is 0.17mm and the available length of the introduction area along the entry direction is 0.80mm, a gradual transition surface can be generated within the range of 0.60mm to 0.80mm to avoid abrupt steps between the slot boundary and the tank interior boundary. If the available length is less than 0.50mm, a rounded transition surface can be used instead of a longer slope, with the rounded radius set to 0.10mm to 0.20mm. After completing the construction of the introduction transition surface, it is replaced or merged with the corresponding original internal surface local area in the 3D model of the reference shell to obtain the tank modeling result corresponding to the introduction area.

[0105] In another example, the processing method for the bonding control area is similar. The bonding control area belongs to the bonding retention area in the final fitting state, and its corresponding modeling method can be determined by referring to the bonding position, bonding direction, and preset bonding control gap between the local protrusion structure 3D model and the inner surface of the reference shell 3D model in the final fitting posture.

[0106] In this embodiment, modeling the fitting control area may include: extracting the target fitting surface corresponding to the inner surface of the reference shell three-dimensional model from the outer surface of the local protruding structure three-dimensional model in the final fitting posture; generating the fitting control surface according to the target fitting surface and the preset fitting control gap; and connecting the fitting control surface with the groove surfaces corresponding to the guide area and the avoidance area through a boundary transition connection. The preset fitting control gap can be determined based on the shell manufacturing error, model registration error, and final retention requirements. For example, for areas requiring strong limiting or retaining functions, the preset fitting control gap can be set to 0.03mm to 0.08mm; for areas that only need to accommodate the local protruding structure and do not bear the main limiting function, the preset fitting control gap can be set to 0.08mm to 0.15mm.

[0107] It should be noted that the problem with the differentiated modeling of the local groove area to be generated corresponding to the avoidance zone in this application is that the interference corresponding to the avoidance zone is not the stable space that the local protrusion structure should occupy in the final fitting posture, but rather the space that the local protrusion structure temporarily passes through during the intermediate fitting stage as the reference shell 3D model gradually approaches the target bearing object 3D model along the fitting path. This type of space may no longer maintain a corresponding relationship with the local protrusion structure in the final fitting posture, or may not bear the fitting, limiting, and holding functions of the local protrusion structure. If the groove is still directly generated according to the outer contour of the local protrusion structure in the final fitting posture, the procedural abutment in the avoidance zone cannot be fully released; if the local groove is enlarged as a whole in order to eliminate the abutment, the fitting control area that should have been retained may also be weakened, reducing the control capability of the inner surface of the shell after final fitting over the local protrusion structure.

[0108] In one example, the differentiated modeling methods for the local trench area to be generated corresponding to the avoidance zone include: S4.1: Determine the intermediate overlay attitude associated with the avoidance zone based on the path interference information corresponding to the avoidance zone.

[0109] Specifically, the avoidance zone reflects the procedural contact formed when the 3D model of the local protrusion structure passes through the area to be generated in the local groove during the middle of the fitting path. Therefore, the intermediate fitting postures associated with the avoidance zone should not be understood as a single posture at which the maximum interference depth occurs, but should cover the continuous posture range of the local protrusion structure entering, passing through, and leaving the avoidance zone. If only one intermediate fitting posture is selected to generate the avoidance space, the subsequently obtained avoidance cavity is prone to bias towards a single instantaneous position and cannot cover the complete passage range of the local protrusion structure during the intermediate fitting stage; if all intermediate fitting postures are included without screening, the range of the avoidance cavity will be expanded by irrelevant postures, causing excessive concavity of the local groove. Based on this, this embodiment uses the path interference information corresponding to the avoidance zone as a clue to screen the intermediate fitting postures related to the actual contact process of the avoidance zone, so that the subsequent path sweep area can correspond to the procedural interference itself, rather than the entire fitting path.

[0110] In this embodiment, path interference information where the path interference position falls within the avoidance zone can be used as initial association records, and the intermediate overlay posture sequence number corresponding to each initial association record can be extracted. If the path interference positions in multiple intermediate overlay postures all fall within the same avoidance zone, and the corresponding path interference depth is greater than a preset effective depth, then these intermediate overlay postures are determined as candidate association postures. The preset effective depth can be determined based on the model synthesis error, which may include scanning error, mesh reconstruction error, registration error, and processing error. For example, when the scanning error is 0.03 mm, the mesh reconstruction error is 0.02 mm, and the registration error is 0.03 mm, 0.08 mm can be used as the effective depth reference value; when the path interference depth is less than 0.08 mm and only appears in a single intermediate overlay posture, it can be regarded as a weak interference record and is not included in the main modeling basis of the avoidance zone. For regions where there is effective interference in two or more consecutive intermediate overlay postures, even if the path interference depth of one posture is slightly lower than the effective depth, it can still be retained within the association posture range to avoid sweep range breaks caused by posture sampling fluctuations.

[0111] S4.2: Obtain the occupied area of ​​the local protrusion structure 3D model relative to the inner surface of the reference shell 3D model under each associated intermediate overlay posture, and merge each occupied area to obtain the path sweep area corresponding to the avoidance area.

[0112] Specifically, the occupied area represents the local spatial range traversed or approached by the 3D model of the local protrusion structure relative to the inner surface of the reference shell 3D model in a certain intermediate fitting posture. Since the interference of the avoidance zone originates from the dynamic passage relationship during the intermediate fitting stage, simply extracting the projected contour of the local protrusion structure in the final fitting posture cannot reflect the spatial envelope of the local protrusion structure during its entry, grazing, and exit from the avoidance zone. This embodiment obtains the path sweep area formed by the local protrusion structure near the avoidance zone by acquiring multiple associated occupied areas in intermediate fitting postures and merging these occupied areas. The path sweep area is equivalent to the continuous passage range left by the local protrusion structure in the middle of the fitting process. Subsequent avoidance cavities can be generated around this passage range, thereby avoiding the expansion of areas unrelated to process interference.

[0113] In this embodiment, the inner surface of the reference shell 3D model can be used as a reference surface. For each associated intermediate fitting posture, the 3D model of the local protrusion structure is transformed to the position under the corresponding posture, and the outer surface sampling points facing the avoidance area in the 3D model of the local protrusion structure are extracted. For each outer surface sampling point, the nearest inner surface point can be determined on the inner surface of the reference shell 3D model. If the distance between the outer surface sampling point and the nearest inner surface point is less than a preset sweep recognition distance, or the direction of the line connecting the two satisfies the contactable direction constraint of the inner surface of the reference shell, then the patch area near the nearest inner surface point is determined as the occupied area under the corresponding posture. The preset sweep recognition distance can be determined according to the preset safety gap and the path interference depth range. For example, when the preset safety gap is 0.12mm and the larger path interference depth in the avoidance area is 0.18mm, the sweep recognition distance can be set to 0.30mm to 0.35mm, so that adjacent areas that have already interfered and those that are about to interfere are all included in the occupied area. If the curvature of the outer surface of the local protrusion structure is large, the sampling density of the outer surface can be increased by using grid vertices, patch center points and edge midpoints as sampling points to reduce the gaps at the boundary of the swept area.

[0114] Furthermore, the occupied regions obtained under the various associated intermediate overlay postures can be merged using patch union, boundary envelope, or projection overlay methods. Specifically, multiple occupied regions can be mapped to the inner surface of the shell where the local trench to be generated region is located, and occupied regions that overlap or whose boundary distance is less than a preset connectivity distance can be merged into a continuous region. The preset connectivity distance can be determined according to the average mesh side length of the local trench to be generated region. For example, when the average mesh side length is 0.08mm, the preset connectivity distance can be 0.16mm to 0.24mm. When there are gaps between multiple occupied regions that are smaller than this connectivity distance, bridging can be performed to fill them, so as to prevent the path sweep region from being artificially split by the discrete mesh interval. The merged path sweep region can also undergo boundary smoothing and small island removal. The small island area threshold can be determined according to the area of ​​the local trench to be generated region. For example, isolated regions that are less than one percent of the area of ​​the local trench to be generated region and are not connected to the main interference region can be excluded from the path sweep region.

[0115] S4.3: Generate candidate avoidance cavities based on the path interference depth corresponding to the path sweep area and the avoidance area.

[0116] Specifically, the path sweep area defines the plane or curved surface range that the clearance cavity should cover, while the path interference depth defines the degree to which the clearance cavity needs to be recessed or released into the inner surface of the shell. The candidate clearance cavity is not simply formed by a uniform depth throughout the path sweep area, but rather by creating locally recessed spaces with varying depths based on the path interference depth at different locations within the clearance area. For locations with larger path interference depths, the clearance cavity can provide a greater clearance depth; for locations with smaller path interference depths or located at the edge of the sweep area, the clearance cavity can gradually decrease in size to smoothly connect with the original inner surface of the shell. The resulting candidate clearance cavity can provide the necessary space around the procedural abutment position without indiscriminately hollowing out the entire area surrounding the clearance area.

[0117] In this embodiment, a depth distribution can be established within the path sweep region. The depth distribution can be determined based on the path interference depth at each path interference position, the number of interference durations of the corresponding intermediate overlay posture, and the distance from the boundary of the path sweep region. For a patch corresponding to multiple path interference positions, a larger path interference depth can be selected as the basic avoidance depth for that patch; for a patch located between multiple interference positions, interpolation can be performed based on the depth of adjacent interference positions to make the avoidance depth continuously change. To avoid local cusps causing the avoidance cavity to be too deep, a higher quantile depth can be used as the main depth basis. For example, the path interference depths in a certain avoidance region are 0.09mm, 0.12mm, 0.15mm, 0.17mm, and 0.28mm, respectively. Among them, 0.28mm only appears near a single mesh cusp. 0.17mm can be selected as the main avoidance depth, and a safety margin of 0.04mm to 0.06mm can be superimposed to obtain a candidate avoidance depth of approximately 0.21mm to 0.23mm. The safety margin can be determined based on the overall error of the model. If the scanning error, registration error and manufacturing error combined are about 0.05 mm, then the safety margin can be taken as about 0.05 mm.

[0118] S4.4: Based on the distance relationship between the candidate avoidance cavity and the fitting control area, the candidate avoidance cavity is subjected to boundary constraint processing to obtain a restricted avoidance cavity.

[0119] Specifically, while the candidate clearance cavity can release procedural interference during the intermediate fitting stage, if its boundary approaches or encroaches excessively on the fitting control area, it will weaken the fitting, limiting, or retaining effect of the inner surface of the shell on the local protrusion structure in the final fitting posture. The clearance area and the fitting control area may be spatially adjacent, or even partially overlap, which is a common geometric relationship when the local protrusion structure transitions from the intermediate passing state to the final fitting state. Therefore, after the candidate clearance cavity is formed, its boundary needs to be constrained according to the fitting control area so that the clearance cavity only retains the part used to release the procedural abutment, without destroying the fitting control surface that needs to be retained in the final fitting posture.

[0120] In this embodiment, the distance relationship between the candidate avoidance cavity boundary and the fitting control area boundary can be calculated. This distance relationship can be the surface distance along the inner surface of the shell or the shortest distance in three-dimensional space. For models with a large curvature of the inner surface of the shell, the surface distance along the inner surface of the shell is preferred, making the constraint result more consistent with the actual transition relationship of the groove boundary. If the distance between the candidate avoidance cavity boundary and the fitting control area is greater than the preset retention spacing, the candidate avoidance cavity can remain unchanged; if it is less than the preset retention spacing, or if the candidate avoidance cavity overlaps with the fitting control area, the candidate avoidance cavity is subjected to boundary back-off, depth reduction, or partial truncation. The preset retention spacing can be determined based on the shell thickness, manufacturing error, and fitting control gap. For example, when the shell thickness is 0.75mm, the manufacturing error is 0.05mm, and the fitting control gap is 0.08mm, the preset retention gap can be set to 0.15mm to 0.25mm; when the local protrusion structure ultimately needs to be strongly limited, it can be taken to be about 0.25mm; when the local protrusion structure only needs to accommodate and does not bear the main holding function, it can be taken to be about 0.15mm.

[0121] Furthermore, boundary constraint processing can be performed using a graded contraction method. For cases where the candidate avoidance cavity is only slightly close to the fitting control area, the boundary of the candidate avoidance cavity can be retracted along the direction away from the fitting control area, restoring the preset retention distance between them. For cases where the candidate avoidance cavity partially intrudes into the fitting control area, the intruding portion can be removed, and a gradual transition surface can be generated at the removal boundary, allowing the avoidance cavity depth to gradually transition from the original candidate depth to the control gap of the fitting control area. For cases where the candidate avoidance cavity extensively covers the fitting control area, the avoidance depth in that area can be reduced, while retaining the fitting surface corresponding to the fitting control area as the priority boundary. For example, when the candidate avoidance cavity depth is 0.22 mm, the fitting control area requires a control gap of no more than 0.08 mm, and the boundary distance between the two is only 0.05 mm, the avoidance depth on the side closer to the fitting control area can be reduced to 0.10 mm to 0.12 mm, and a transition width of approximately 0.20 mm can be set, allowing the avoidance cavity to gradually converge to the boundary of the fitting control area. The restricted clearance cavity obtained after boundary constraint treatment not only retains the clearance space required in the intermediate fitting stage, but also leaves a stable boundary for the fitting control area in the final fitting state.

[0122] S4.5: Based on the restricted avoidance cavity, perform local concavity modeling on the local groove area to be generated corresponding to the avoidance area.

[0123] Specifically, local concavity modeling involves transforming the constrained clearance cavity into the actual geometric structure on the inner surface of the reference shell's 3D model. This modeling does not involve regenerating the entire reference shell 3D model; instead, it replaces or modifies the original inner surface with the curved surface corresponding to the constrained clearance cavity within the local groove area to be generated in the clearance zone, creating a clearance structure with local concavity, gradual transition, and continuous boundaries.

[0124] In this embodiment, the boundary curve and the concave bottom surface of the restricted avoidance cavity can be extracted first. The boundary curve can be determined by the intersection of the restricted avoidance cavity and the inner surface of the original reference shell 3D model, or by the outer contour of the restricted avoidance cavity on the local groove to be generated area; the concave bottom surface can be generated according to the depth distribution of the restricted avoidance cavity. Subsequently, an avoidance transition surface is constructed between the boundary curve of the restricted avoidance cavity and the concave bottom surface. The avoidance transition surface can be generated by spline surface fitting, mesh patch reconstruction, or a gradual offset method based on the normal of the original shell inner surface. To avoid the formation of sharp steps at the edge of the avoidance cavity, a transition width can be set. For example, when the avoidance depth is about 0.20 mm, the transition width can be set to 0.20 mm to 0.40 mm; when the avoidance depth exceeds 0.30 mm, the transition width can be expanded to 0.40 mm to 0.60 mm, so that the local concavity transitions smoothly from the original shell inner surface to the concave bottom surface. For thin-walled shells such as orthodontic appliances, the transition surface can also be checked in conjunction with the shell thickness to avoid local depressions that would cause the remaining thickness between the outer and inner surfaces to be lower than the manufacturing allowable range.

[0125] It is understandable that the generation of a 3D shell model is, in practice, a process of splicing and merging local model patches with the original shell model. For those skilled in the field of 3D modeling, surface splicing, mesh fusion, boundary bridging, local remeshing, and surface smoothing are all relatively mature 3D geometric processing methods. This application does not impose any particular limitation on the specific model splicing algorithm used. As long as the imported groove surface, the avoided groove surface, and the fitted groove surface can be included as components of the local groove surface, and this local groove surface can be embedded into the corresponding local groove to be generated region in the reference shell 3D model, so that the generated 3D shell model remains geometrically continuous and can express the local groove structure, the implementation requirements of this application can be met.

[0126] In one optional implementation, the original inner surfaces of the base shell corresponding to the area to be generated in the local groove can be deleted or marked in the 3D model of the base shell to form the fusion boundary. Then, the import groove surface, the avoidance groove surface, and the fitting groove surface are spliced ​​together according to their respective spatial positions to obtain the local groove surface. For cases where there are boundary gaps, boundary overlaps, or abrupt changes in normal between adjacent groove surfaces, boundary point matching, boundary line resampling, triangular patch bridging, spline surface transition, or local mesh reconstruction can be used for connection. After the connection is completed, the fusion boundary between the local groove surface and the base shell 3D model can be further smoothed to ensure that there are no sharp corners, isolated patches, or non-manifold boundaries when the local groove surface transitions from the import area and the avoidance area to the fitting control area.

[0127] Furthermore, surface smoothing can be achieved using existing methods such as Laplacian smoothing, curvature constraint smoothing, normal consistency adjustment, or local spline fitting, and this application does not impose further limitations.

[0128] Therefore, the final generation of the 3D shell model does not depend on a specific modeling software or meshing algorithm, but can be accomplished based on existing 3D geometric processing techniques. It is important to emphasize that the focus of this application is on importing the channel surfaces, avoiding the channel surfaces, and fitting the channel surfaces from the differentiated modeling results after the interference partitioning of the fitting path. The model splicing and fusion itself can be achieved using existing surface fusion and mesh splicing methods in this field. Through this processing, without changing the overall shape of the baseline 3D shell model, only the local channel areas to be generated can be structurally replaced and boundary-fused to obtain a 3D shell model with local channel structures.

[0129] This application embodiment also provides a three-dimensional orthodontic model generation system, the system comprising: The model acquisition module is used to acquire a three-dimensional model of the target bearing object, a three-dimensional model of the local protrusion structure set on the surface of the target bearing object, and a three-dimensional model of the reference shell corresponding to the three-dimensional model of the target bearing object. The fitting path determination module is used to determine the fitting path of the reference shell three-dimensional model relative to the target load-bearing object three-dimensional model based on the final fitting posture of the target load-bearing object three-dimensional model and the reference shell three-dimensional model. The fitting path includes multiple intermediate fitting postures. The interference partitioning module is used to detect the path interference information between the three-dimensional model of the local protrusion structure and the inner surface of the three-dimensional model of the reference shell under each of the intermediate fitting postures, and to determine the introduction area, avoidance area and fitting control area in the local groove to be generated area corresponding to the three-dimensional model of the local protrusion structure based on the path interference information. The tank modeling module is used to perform differentiated modeling on the local tank to be generated area based on the import area, the avoidance area and the fitting control area, and generate a three-dimensional shell model with local tank structure.

[0130] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for generating a three-dimensional orthodontic model, characterized in that, The method includes: Obtain a three-dimensional model of the target load-bearing object, a three-dimensional model of the local protrusion structure set on the surface of the target load-bearing object, and a three-dimensional model of the reference shell corresponding to the three-dimensional model of the target load-bearing object; Based on the final fitting posture of the target bearing object 3D model and the reference shell 3D model, the fitting path of the reference shell 3D model relative to the target bearing object 3D model is determined, and the fitting path includes multiple intermediate fitting postures; Under each of the aforementioned intermediate fitting postures, the path interference information between the three-dimensional model of the local protrusion structure and the inner surface of the three-dimensional model of the reference shell is detected, and based on the path interference information, the introduction area, the avoidance area, and the fitting control area are determined in the local groove to be generated area corresponding to the three-dimensional model of the local protrusion structure. Based on the import area, the avoidance area, and the fitting control area, the local groove to be generated area is differentially modeled to generate a three-dimensional shell model with a local groove structure.

2. The method of claim 1, wherein, The area to be generated for the local groove is used to characterize the spatial range in which the local protrusion structure three-dimensional model forms the corresponding local groove on the inner surface of the reference shell three-dimensional model. The area to be generated for the local groove is determined based on the outer contour of the three-dimensional model of the local protrusion structure, the outward convex direction of the three-dimensional model of the local protrusion structure relative to the three-dimensional model of the target bearing object, and the surface transition area around the three-dimensional model of the local protrusion structure.

3. The method of claim 1, wherein, The method for determining the fitting path includes: Based on the fitting relationship between the target bearing object 3D model and the reference shell 3D model, the final fitting posture of the reference shell 3D model relative to the target bearing object 3D model is determined; Based on the main body extension direction and contour extension direction of the target bearing object three-dimensional model, the opening edge direction of the reference shell three-dimensional model, and the distribution position of the local protrusion structure three-dimensional model in the target bearing object three-dimensional model, the initial fitting posture of the reference shell three-dimensional model is determined. Based on the relative pose change between the initial fitting posture and the final fitting posture, the fitting path of the reference shell 3D model relative to the target bearing object 3D model is generated.

4. The method of claim 3, wherein, The intermediate overlay posture is determined based on the relative pose change, and the determination method of the intermediate overlay posture includes: The relative pose change between the initial fitting posture and the final fitting posture is decomposed to obtain the translational change and the rotational change. The attitude sampling interval is determined based on the translational change, the rotational change, the number of three-dimensional models of the local protrusion structure, and the outward convexity height of the three-dimensional model of the local protrusion structure relative to the three-dimensional model of the target bearing object. According to the attitude sampling interval, the relative pose change between the initial fitting pose and the final fitting pose is interpolated to obtain multiple intermediate fitting poses.

5. The method of claim 2, wherein, The methods for determining the path interference information include: For each intermediate fitting posture, determine the relative position between the three-dimensional model of the local protrusion structure and the inner surface of the three-dimensional model of the reference shell; When the distance between the three-dimensional model of the local protrusion structure and the inner surface of the three-dimensional model of the reference shell is less than the preset safety gap, the corresponding position is determined as the path interference position; The corresponding path interference depth is determined based on the difference between the path interference position and the preset safety gap. The path interference position, the path interference depth, and the corresponding intermediate overlay posture are determined as the path interference information.

6. The method of claim 5, wherein, The methods for determining the import area, avoidance area, and fitting control area include: Based on the path interference positions in each of the path interference information, the path interference positions that are located within the local groove to be generated area and are adjacent to each other are merged into candidate interference areas; Based on the arrangement order of the intermediate overlay postures corresponding to each path interference information in the overlay path, the interference occurrence stage corresponding to each candidate interference region is determined. The candidate interference region located at the beginning of the mating path during the corresponding interference occurrence stage is determined as the introductory region, and the candidate interference region located at the middle of the mating path during the corresponding interference occurrence stage is determined as the avoidance region. The mating control region is determined based on the mating position of the three-dimensional model of the local protrusion structure and the inner surface of the three-dimensional model of the reference shell under the final mating posture.

7. The method of claim 6, wherein, The differentiated modeling methods for the local groove to be generated area corresponding to the import area include: Based on the path interference information corresponding to the import area, determine the entry direction and path interference depth of the three-dimensional model of the local protrusion structure when it enters the local groove to be generated area; The slot expansion amount of the introductory area is determined based on the path interference depth, and the slot boundary corresponding to the introductory area is expanded according to the slot expansion amount.

8. The method of claim 6, wherein, The differentiated modeling methods for the local tank body to be generated corresponding to the avoidance zone include: Based on the path interference information corresponding to the avoidance zone, determine the intermediate overlay attitude associated with the avoidance zone; Obtain the occupied area of ​​the three-dimensional model of the local protrusion structure relative to the inner surface of the three-dimensional model of the reference shell under each associated intermediate overlay posture, and merge each occupied area to obtain the path sweep area corresponding to the avoidance area; Based on the path interference depth corresponding to the path sweep area and the avoidance area, candidate avoidance cavities are generated; Based on the distance relationship between the candidate avoidance cavity and the fitting control area, the candidate avoidance cavity is subjected to boundary constraint processing to obtain a restricted avoidance cavity; Based on the restricted avoidance cavity, a local concavity model is performed on the local groove area to be generated corresponding to the avoidance area.

9. The method according to claim 1, characterized in that, The generation methods of the three-dimensional shell model include: Based on the differentiated modeling results corresponding to the import area, the avoidance area, and the fitting control area, respectively, the import groove surface, the avoidance groove surface, and the fitting groove surface are generated; Transitional connections are made between the adjacent boundaries of the inlet groove surface, the avoidance groove surface, and the fitting groove surface to obtain a local groove surface. The local groove surface is fused with the inner surface of the shell in the reference shell 3D model corresponding to the area where the local groove is to be generated, and the fusion boundary is smoothed to generate a 3D shell model with a local groove structure.

10. A three-dimensional orthodontic model generation system, implemented based on the three-dimensional orthodontic model generation method as described in any one of claims 1-9, characterized in that, The system includes: The model acquisition module is used to acquire a three-dimensional model of the target bearing object, a three-dimensional model of the local protrusion structure set on the surface of the target bearing object, and a three-dimensional model of the reference shell corresponding to the three-dimensional model of the target bearing object. The fitting path determination module is used to determine the fitting path of the reference shell three-dimensional model relative to the target load-bearing object three-dimensional model based on the final fitting posture of the target load-bearing object three-dimensional model and the reference shell three-dimensional model. The fitting path includes multiple intermediate fitting postures. The interference partitioning module is used to detect the path interference information between the three-dimensional model of the local protrusion structure and the inner surface of the three-dimensional model of the reference shell under each of the intermediate fitting postures, and to determine the introduction area, avoidance area and fitting control area in the local groove to be generated area corresponding to the three-dimensional model of the local protrusion structure based on the path interference information. The tank modeling module is used to perform differentiated modeling on the local tank to be generated area based on the import area, the avoidance area and the fitting control area, and generate a three-dimensional shell model with local tank structure.

Images