Cartoon image toy 3D scanning reverse modeling copying processing method

Abstract

The application discloses a cartoon image toy 3D scanning reverse modeling copying processing method, which comprises the following steps: selecting a cartoon image toy prototype to be copied, removing detachable parts, and spraying a matt auxiliary agent on the detail feature area of the cartoon image prototype; scanning the cartoon image toy prototype, and specially scanning the detachable parts and the detail feature area to obtain three-dimensional point cloud data; performing trajectory fitting processing on the edge extension position of the detachable parts; and performing weighted data splicing on the three-dimensional point cloud data. The application adopts differential high-precision scanning and layered progressive scanning, accurately captures complex structures such as detail feature areas, restores the arc, gap and limiting features, performs trajectory fitting and curvature continuity processing on the edge extension section of the detachable parts, solves the problems of misconnection, too large gap and poor cooperation, and greatly improves the assembly precision and smoothness.

Description

Technical Field

[0001] This invention relates to the field of technology, specifically to a method for reverse modeling and replicating 3D scanning of cartoon character toys. Background Technology

[0002] Cartoon character toys, with their vivid designs and distinctive personalities, are loved by consumers of all ages, especially those derived from anime IPs and customized cartoon dolls, where market demand continues to rise. Currently, the replication of cartoon character toys mainly employs two methods: traditional hand-modeling or forward design modeling.

[0003] Traditional hand-modeling and replication relies on the experience of technicians. The size and outline of the cartoon prototype are measured by hand, and then hand-carving or simple modeling is performed. This replication method is not only extremely inefficient, but also makes it difficult to accurately reproduce the details of the cartoon prototype. Especially for the complex connection positions of toys, manual measurement cannot capture the subtle gaps, joint curvature and fitting precision of the connection parts, resulting in problems such as misalignment, excessive gaps or unnatural joints in the replicated parts. The replication accuracy is poor. At the same time, it is difficult to guarantee the consistency of the connection parts of different batches of products, which cannot meet the needs of mass production.

[0004] While forward design modeling can improve processing accuracy to some extent, it requires redrawing the 3D model of the cartoon character, resulting in a long design cycle and a large workload. Especially for cartoon characters with complex shapes and numerous details, forward design struggles to accurately capture the three-dimensional outline and detailed texture of the prototype. Furthermore, it is difficult to accurately reproduce the connection relationships, tolerances, and subtle structures of the toy's complex connection positions. In particular, it cannot handle deep cavities and narrow slits such as joint extensions and internal hidden connection surfaces, leading to significant deviations between the designed connection positions and the prototype. This fails to meet the replication requirements of complex connection structures in cartoon toys. Therefore, we propose a 3D scanning reverse modeling replication processing method for cartoon character toys. Summary of the Invention

[0005] The purpose of this invention is to provide a method for reverse modeling and replicating 3D scanning of cartoon character toys, so as to solve the problems that need to be solved in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for 3D scanning reverse modeling and replication of cartoon character toys, the method comprising the following steps:

[0007] Step S1: Select the prototype of the cartoon character toy to be replicated, remove the detachable parts, and spray a matte auxiliary agent on the detailed feature areas of the cartoon character prototype.

[0008] Step S2: Using a 3D laser scanner, scan the prototype of the cartoon character toy, and specifically scan the detachable parts and detailed feature areas to obtain 3D point cloud data. Perform trajectory fitting processing on the edge extension positions of the detachable parts, and perform weighted data stitching on the 3D point cloud data.

[0009] Step S3: Import the 3D point cloud data obtained in step S2 into the reverse modeling system for reverse modeling processing;

[0010] Step S4: Based on the optimized 3D model obtained in step S3, select suitable processing materials and perform replication processing;

[0011] Step S5: After the surface treatment is completed, the key positions of the replica are scanned using a 3D scanner and compared with the scan data and 3D model of the original cartoon character toy to be replicated.

[0012] As a further description of the above technical solution:

[0013] The detailed feature region-specific scanning in step S2 includes the following steps:

[0014] Step S21: Adjust the scanning resolution according to the detailed feature area. Set the scanning resolution to 0.01-0.03mm. Set the scanning resolution of the regular surface of the cartoon character toy prototype after removing the detachable parts to 0.05-0.1mm.

[0015] Step S22: Use corresponding viewing angles to scan the detailed feature areas. Use a surround scan for the joint connections in the detailed feature areas, rotating the scanning angle around the joint to capture the joint's inner and outer connection arc, fitting gap, and limiting structure.

[0016] Step S23: After completing a specialized scan of each detailed feature area, collect specialized scan data simultaneously, use a point cloud stitching algorithm to stitch together different scan point cloud data, calibrate the point cloud data of the detailed feature area during cloud data stitching, and generate complete 3D point cloud data of the cartoon prototype, detachable parts and detailed feature areas after scanning is completed, and mark the detailed feature areas.

[0017] Step S23: Perform layered progressive scanning on the cartoon prototype, detachable parts, and detailed feature areas. During scanning, scan in layers along the joint axis, with each layer performed at a minimum scanning step distance d=0.008-0.01mm. For the hidden positions of detailed feature areas with a depth of 0.5-5mm, use a small-angle side-lit light source to illuminate and tighten the scanning position. Use a curvature fitting algorithm to complete the missing hidden points, and output a complete point cloud including extensions and hidden positions, which is marked as a high-precision modeling area.

[0018] As a further description of the above technical solution:

[0019] The reverse modeling process in step S3 includes the following steps:

[0020] Step S31: Extract the complete 3D point cloud data of the detailed feature area separately, remove noise points and isolated points from the 3D point cloud data, complete the missing 3D point cloud data, and simplify the 3D point cloud data of the cartoon prototype and detachable parts.

[0021] Step S32: Reconstruct the overall contour surface of the cartoon prototype and detachable parts, and then perform specific surface reconstruction on the detailed feature areas;

[0022] Step S33: Integrate the surface reconstruction data of the cartoon prototype, detachable parts, and detailed feature areas; adjust the relative positions of the cartoon prototype, detachable parts, and detailed feature areas; perform accuracy testing on the integrated 3D model; and make fine adjustments and corrections based on the scanning data of the cartoon character toy prototype to be replicated.

[0023] Step S34: Optimize the completed 3D model, and smooth the edge connection positions of detachable parts and detailed feature areas;

[0024] Step S35: Reconstruct the surface of the completed 3D model according to the principle of curvature continuity, and perform trajectory fitting on the extended segment.

[0025] As a further description of the above technical solution:

[0026] The replication method in step S4 includes the following steps:

[0027] Step S41: Adjust the parameters of the processing equipment according to the structural complexity and tolerance requirements of the cartoon prototype, detachable parts, and detailed feature areas;

[0028] Step S42: Decompose the 3D model into cartoon prototype, detachable parts, and detailed feature areas, and adjust the processing parameters to avoid processing deviations;

[0029] Step S43: Assemble the finished cartoon prototype, detachable parts, and detailed feature areas;

[0030] Step S44: Perform surface treatment on the assembled replica product, remove burrs, and polish the detailed feature areas.

[0031] As a further description of the above technical solution:

[0032] The trajectory fitting process for the edge extension position of the removable component in step S2 includes the following steps:

[0033] Step S221: Establish a local cylindrical coordinate system with the joint position center of the detachable component as the origin. Use a 3D laser scanner to identify the curvature center, rotation axis, tangential normal and transition fillet of the joint position of the detachable component. Extract the starting end face boundary contour, tangential vector and normal vector of the edge extension position of the detachable component as the forced constraint benchmark for trajectory fitting.

[0034] Step S222: Divide the joint position point cloud data of the detachable component into a head fitting segment, a main body extension segment, and a tail transition segment along the joint axis. Generate an initial trajectory according to the axial direction of the main body extension segment. Force the normal of the trajectory start end to be consistent with the normal of the joint end face to avoid abrupt changes in the angle. Calculate the continuous data of the first and second derivatives of the trajectory. If the continuous data of the first and second derivatives are not satisfied, automatically iterate and correct the control points.

[0035] Step S223: Identify the width of the fusion zone between the main body extension and the cartoon prototype, make the curvature at the entrance of the main body extension consistent with the curvature at the exit of the joint position, then perform a linear curvature gradient within the fusion zone, and perform a 360° rotational interference simulation with the center of the joint position.

[0036] Step S224: Using the fitted trajectory as a skeleton, sweep radially to generate an extended surface segment, and output the selectable surface segment.

[0037] As a further description of the above technical solution:

[0038] The cartoon prototype, detachable parts, and detailed feature areas are weighted during cloud stitching. The weighting formula is as follows:

[0039] ;

[0040] in, The region weights are the region weights of the detailed feature regions. 3-5;

[0041] Area weighting of cartoon prototypes and detachable parts =1;

[0042] The total registration error of the weighted point cloud data;

[0043] R represents the rotational transformation relationship of the point cloud data in three-dimensional space;

[0044] T represents the translational transformation relationship of the point cloud data in three-dimensional space;

[0045] Calculate and accumulate the data for all point cloud data to be stitched together, numbered from 1 to n.

[0046] i is the index of a single point in the point cloud data;

[0047] n represents the total number of valid point clouds participating in the splicing calculation;

[0048] Let i be the i-th 3D coordinate in the target point cloud data;

[0049] Let be the three-dimensional coordinates of the i-th point in the source point cloud data.

[0050] As a further description of the above technical solution:

[0051] The minimum scan step distance of the extended curved surface satisfies the following formula:

[0052] ;

[0053] Where L is the effective length of the extension section, which ranges from 1 to 10 mm;

[0054] N is the number of single-turn scanning angles in the extended section, ranging from 24 to 72.

[0055] K represents the number of axial scanning layers in the extended section, ranging from 8 to 32.

[0056] As a further description of the above technical solution:

[0057] In step S1, the matte scanning aid is a nano-sized silica matte spray, with the spray thickness controlled at 0.01-0.05 mm. After spraying, it is left to stand for 3-5 minutes until completely dry.

[0058] As a further description of the above technical solution:

[0059] In step S5, the comparison detection uses 3D comparison analysis. When the color difference map shows a deviation exceeding ±0.02mm, the abnormal area is automatically marked and the modeling step is returned for correction.

[0060] As a further description of the above technical solution:

[0061] An electronic device, comprising:

[0062] One or more processors;

[0063] Memory, used to store one or more programs;

[0064] When the one or more programs are executed by the one or more processors, the one or more processors implement a method for reverse modeling and replicating 3D scanning of cartoon character toys.

[0065] Compared with the prior art, the present invention has the following beneficial effects:

[0066] (1) First, the present invention adopts differentiated high-precision scanning and layered progressive scanning to accurately capture complex structures such as detailed feature areas, restore the curvature, gap and limit features, and perform trajectory fitting and curvature continuity processing on the edge extension of detachable parts to solve the problems of misalignment, excessive gap and poor fit, and greatly improve the assembly accuracy and smoothness of movement.

[0067] (2) Secondly, special scanning combined with small-angle side-emitting light source and curvature fitting completion can fully acquire structural data that is difficult to collect by traditional methods, such as hidden positions, extension segments, and inner and outer sides of joints. Furthermore, by prioritizing the registration accuracy of key areas through weighted point cloud stitching, the model has better integrity and consistency. This setting can replace manual measurement, carving and forward redrawing, making the reverse modeling process shorter and the workload lower. Complex shapes can also be quickly molded. This setting allows for regional modeling and processing, standardized parameters and tolerance control, and high consistency of multiple batches of products, meeting the needs of industrial mass production.

[0068] (3) In addition, the nano-matte spray pretreatment of the present invention eliminates reflection and refraction, avoiding scanning distortion; point cloud denoising, completion, simplification and surface reconstruction improve the quality of the model, and with the formulaic calculation of the extended segment scanning step distance, it ensures that the trajectory is smooth, without interference and without jagged abrupt changes. After replication, 3D comparative analysis and color difference map are used for automatic detection. If the deviation is exceeded, it is immediately marked and repaired. The quality is traceable and the defects can be quickly corrected. The final product's appearance, feel and assembly effect are highly close to the prototype. Attached Figure Description

[0069] Figure 1 This is a schematic diagram of the system principle of the present invention. Detailed Implementation

[0070] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0071] Example 1:

[0072] Please see Figure 1 This invention provides a technical solution: a method for 3D scanning reverse modeling and replication of cartoon character toys, the method comprising the following steps:

[0073] Step S1: Select the prototype of the cartoon character toy to be replicated, remove detachable parts, and clean the surface of the prototype to remove dust, stains, and attached debris. Spray a matte auxiliary agent on the detailed feature areas of the cartoon character prototype. Remove detachable parts that are easily obscured by complex connection points, joint extensions, and hidden locations using a non-destructive disassembly method, and mark and store them separately. Evenly spray a matte scanning auxiliary agent on the reflective areas, transparent areas, complex connection points, joint extensions, and hidden gaps on the surface of the prototype to avoid scanning distortion caused by reflection and refraction during the scanning process, and ensure that fine structures can be accurately captured. Fix the pre-treated prototype and the disassembled parts on the scanning worktable, adjust the fixing angle to ensure that the joint extensions and hidden connection surfaces are unobstructed and can be completely covered by the scanning equipment. The matte scanning auxiliary agent uses nano-grade silica matte spray, and the spray thickness is controlled at 0.01-0.05mm. After spraying, let it stand for 3-5 minutes until completely dry.

[0074] Step S2: Using a 3D laser scanner, scan the prototype of the cartoon character toy, and specifically scan the detachable parts and detailed feature areas to obtain 3D point cloud data. Perform trajectory fitting processing on the edge extension positions of the detachable parts, and perform weighted data stitching on the 3D point cloud data.

[0075] Step S3: Import the 3D point cloud data obtained in step S2 into the reverse modeling system for reverse modeling processing;

[0076] Step S4: Based on the optimized 3D model obtained in step S3, select suitable processing materials and perform replication processing;

[0077] Step S5: The key positions of the surface-processed replica are scanned using a 3D scanner and compared with the scan data and 3D model of the original cartoon character toy to be replicated. The comparison and detection adopts 3D comparison analysis. When the color difference map shows a deviation exceeding ±0.02mm, the abnormal area is automatically marked and the modeling step is returned for correction.

[0078] Among these methods, differentiated high-precision scanning and layered progressive scanning are employed to accurately capture complex structures such as detailed feature areas, restoring curvature, gaps, and limiting features. Trajectory fitting and curvature continuity processing are performed on the edge extensions of detachable components, resolving issues such as misalignment, excessive gaps, and poor fit, significantly improving assembly accuracy and smoothness of movement. Secondly, specialized scanning, combined with small-angle side-emitting light sources and curvature fitting completion, fully acquires structural data that is difficult to obtain using traditional methods, such as hidden locations, extensions, and the inner and outer sides of joints. Furthermore, weighted point cloud stitching prioritizes registration accuracy in key areas, resulting in a complete model. With improved integrity and consistency, this setup can replace manual measurement, carving, and forward redrawing, resulting in a shorter reverse modeling process, lower workload, and faster mold creation even for complex shapes. This setup features regional modeling and processing, standardized parameters, and tolerance control, ensuring high consistency across multiple batches of products and meeting the needs of industrial mass production. In addition, the invention's nano-matte spray pretreatment eliminates reflection and refraction, avoiding scanning distortion. Point cloud denoising, completion, simplification, and surface reconstruction improve model quality, and the formulaic calculation of the extended segment scanning step distance ensures smooth, interference-free, and jagged abrupt changes in trajectory. After replication, 3D comparative analysis and color difference maps are used for automatic detection, and out-of-tolerance items are immediately marked and reworked, ensuring quality traceability and rapid correction of defects. The final product's appearance, feel, and assembly effect are highly close to the prototype.

[0079] Example 2:

[0080] The detailed feature region-specific scanning in step S2 includes the following steps:

[0081] Step S21: Adjust the scanning resolution according to the detailed feature area. Set the scanning resolution to 0.01-0.03mm. Set the scanning resolution of the regular surface of the cartoon character toy prototype after removing the detachable parts to 0.05-0.1mm. Adjust the scanning light source intensity to avoid the light source being too strong, which will cause shadows at the connection gaps, and the light source being too weak, which will cause the loss of details. Ensure that the scanning light can evenly cover every fine structure at the connection position.

[0082] Step S22: Use corresponding viewing angles to scan the detailed feature areas. Use a surround scan for the joint connections in the detailed feature areas, rotating the scanning angle around the joint to capture the joint's inner and outer connection arc, fitting gap, and limiting structure.

[0083] Step S23: After completing a specialized scan of each detailed feature area, collect specialized scan data simultaneously. Use a point cloud stitching algorithm to stitch together different scan point cloud data. During cloud data stitching, calibrate the point cloud data of the detailed feature area to remove redundant and noise points, ensuring that the point cloud data at the connection points is continuous, complete, without discontinuities or deviations. After scanning, generate complete 3D point cloud data of the cartoon prototype, detachable parts, and detailed feature areas. Mark the detailed feature areas to facilitate subsequent modeling processing.

[0084] Step S23: Perform layered progressive scanning on the cartoon prototype, detachable parts, and detailed feature areas. During scanning, scan in layers along the joint axis, with each layer executed at a minimum scanning step distance d=0.008-0.01mm to ensure the integrity of the extended segment outline. For the hidden positions of detailed feature areas with a depth of 0.5-5mm, use a small-angle side-lit light source to illuminate and tighten the scanning position. Use a curvature fitting algorithm to fill in the missing hidden points, outputting a complete point cloud containing the extended segment and hidden positions, and mark it as a high-precision modeling area. The limitation of each step distance, etc., can ensure more accurate modeling.

[0085] The reverse modeling process in step S3 includes the following steps:

[0086] Step S31: Extract the complete 3D point cloud data of the detailed feature area separately, remove noise points and isolated points from the 3D point cloud data, and complete the missing 3D point cloud data to ensure that the point cloud data at the connection position accurately reflects the prototype structure; simplify the 3D point cloud data of the cartoon prototype and detachable parts to reduce the amount of data and improve modeling efficiency without affecting the overall shape.

[0087] Step S32: Reconstruct the overall outline and curved surface of the cartoon prototype, detachable parts, and detailed feature areas. Then, perform specific curved surface reconstruction on the detailed feature areas. For joint connections, reconstruct the curved surface structure of the joint balls and joint grooves to accurately restore the joint's mating arc, gap size, and limiting structure, ensuring that the replicated joints can move flexibly without looseness. For the connection between clothing and the body, reconstruct the curved transition of the connecting surface to restore the natural arc and fit of the folds, avoiding problems such as misalignment and excessive gaps. For the connection between accessories and the main body, reconstruct the positioning structure, fitting surface, and mating tolerance of the connecting surface to ensure that accessories and the main body can be accurately assembled.

[0088] Step S33: Integrate the surface reconstruction data of the cartoon prototype, detachable parts, and detailed feature areas; adjust the relative positions of the cartoon prototype, detachable parts, and detailed feature areas; perform accuracy testing on the integrated 3D model; and make fine adjustments based on the scanning data of the cartoon character toy prototype to be replicated until the error of the complex connection position between the model and the prototype is controlled within ±0.02mm.

[0089] Step S34: Optimize the completed 3D model, smooth the edges of detachable parts and detailed feature areas to avoid burrs and sharp corners, and ensure the feel and aesthetics of the replicated product; add machining baselines and positioning holes to the model according to the processing requirements, and add assembly positioning marks at complex connection positions and hidden assembly surfaces to facilitate subsequent processing and assembly.

[0090] Step S35: Reconstruct the surface of the completed 3D model according to the principle of curvature continuity, and perform trajectory fitting on the extended segment.

[0091] Among these methods, the separate extraction of detailed feature point clouds, denoising and completion, and hierarchical simplification of main point clouds are employed. This ensures accurate restoration of key features such as connection positions, joints, and hidden structures while significantly reducing the amount of modeling data and improving the efficiency and stability of reverse modeling. Furthermore, through a layered reconstruction approach that combines overall contour surface reconstruction and detailed feature-specific surface reconstruction, the model is integrated and then subjected to accuracy testing and in-situ fine-tuning. This ensures that the error in complex connection positions is strictly controlled within ±0.02mm, guaranteeing that the 3D model is highly consistent with the cartoon prototype. The surface is reconstructed according to the principle of curvature continuity, and the trajectory of the extension segments is fitted, resulting in smooth surface transitions without sharp abrupt changes and high structural matching of the extension segments. This enhances the product's structural strength, assembly stability, and smoothness of movement.

[0092] The replication method in step S4 includes the following steps:

[0093] Step S41: Adjust the parameters of the processing equipment according to the structural complexity and tolerance requirements of the cartoon prototype, detachable parts, and detailed feature areas;

[0094] Step S42: Decompose the 3D model into cartoon prototype, detachable parts, and detailed feature areas, and adjust the processing parameters to avoid processing deviations;

[0095] Step S43: Assemble the finished cartoon prototype, detachable parts, and detailed feature areas;

[0096] Step S44: Perform surface treatment on the assembled replica product, remove burrs, and polish the detailed feature areas.

[0097] Among them, the processing parameters are adjusted according to the structural complexity and the difference in fit tolerance, and the component disassembly processing mode is used to reduce the processing difficulty of complex structures, avoid deformation, deviation and interference problems that are easy to occur in the overall processing, ensure the stability of the size and position accuracy of each component, and uniform assembly after component processing to ensure accurate alignment and uniform fit gap of detachable components and detailed feature areas, so as to meet the activity and assembly requirements of cartoon toys.

[0098] The trajectory fitting process for the edge extension position of the removable component in step S2 includes the following steps:

[0099] Step S221: Establish a local cylindrical coordinate system with the joint position center of the detachable component as the origin. Use a 3D laser scanner to identify the curvature center, rotation axis, tangential normal and transition fillet of the joint position of the detachable component. Extract the starting end face boundary contour, tangential vector and normal vector of the edge extension position of the detachable component as the forced constraint benchmark for trajectory fitting.

[0100] Step S222: Divide the joint position point cloud data of the detachable component into a head fitting segment, a main body extension segment, and a tail transition segment along the joint axis. Generate an initial trajectory according to the axial direction of the main body extension segment. Force the normal of the trajectory start end to be consistent with the normal of the joint end face to avoid abrupt changes in the angle. Calculate the continuous data of the first and second derivatives of the trajectory. If the continuous data of the first and second derivatives are not satisfied, automatically iterate and correct the control points.

[0101] Step S223: Identify the width of the fusion zone between the main body extension and the cartoon prototype, make the curvature at the entrance of the main body extension consistent with the curvature at the exit of the joint position, then perform a linear curvature gradient within the fusion zone, and perform a 360° rotational interference simulation with the center of the joint position to ensure that the main body extension does not collide with other main body components throughout its entire stroke.

[0102] Step S224: Using the fitted trajectory as a skeleton, sweep radially to generate an extended surface segment, perform Laplacian smoothing on the transition positions, retain features while eliminating jagged edges and abrupt changes; output an optional surface.

[0103] Specifically, a local cylindrical coordinate system is established at the joint center, and the boundary contour, vector and curvature parameters are accurately extracted as a forced constraint benchmark to ensure the geometric accuracy of trajectory fitting from the source, avoiding trajectory offset and misalignment. Furthermore, the segmented fitting of the extension segment and the forced constraint of trajectory continuity ensure the continuity of the first and second derivatives, eliminate abrupt angle changes and surface creases, making the extension segment smooth in shape and stable in structure. Moreover, through consistent curvature transition and linear gradient of the fusion zone, the extension segment can be naturally integrated with the joint and the prototype, completely avoiding assembly and motion interference, and ensuring smooth and reliable movement. In addition, the surface generated by sweeping the fitted trajectory and smoothing it with Laplacian polishing completely preserves the detailed features while eliminating jagged edges, burrs and surface abrupt changes, which greatly improves the surface quality and modeling accuracy of the extension segment.

[0104] The cartoon prototype, detachable parts, and detailed feature areas are weighted during cloud stitching. The weighting formula is as follows:

[0105] ;

[0106] in, The region weights are the region weights of the detailed feature regions. 3-5;

[0107] Area weighting of cartoon prototypes and detachable parts =1;

[0108] The total registration error of the weighted point cloud data;

[0109] R represents the rotational transformation relationship of the point cloud data in three-dimensional space;

[0110] T represents the translational transformation relationship of the point cloud data in three-dimensional space;

[0111] Calculate and accumulate the data for all point cloud data to be stitched together, numbered from 1 to n.

[0112] i is the index of a single point in the point cloud data;

[0113] n represents the total number of valid point clouds participating in the splicing calculation;

[0114] Let i be the i-th 3D coordinate in the target point cloud data;

[0115] Let be the three-dimensional coordinates of the i-th point in the source point cloud data.

[0116] The point cloud registration strategy employs a weighted priority approach based on detailed feature regions, assigning higher weights to high-precision regions such as joints, connecting points, and hidden structures. This significantly improves the stitching accuracy of key parts and ensures the fidelity of complex structures. The hierarchical weighted calculation of regular and detailed regions ensures that high-precision features are not distorted while maintaining overall point cloud stitching efficiency and stability. This avoids local errors affecting the overall registration result. Furthermore, the optimal registration is performed based on a weighted error function, resulting in point cloud stitching that more closely matches the real prototype structure. This effectively reduces problems such as stitching misalignment, discontinuity, and large deviations, providing a reliable data foundation for subsequent high-precision reverse modeling. The weighted algorithm can adaptively balance the alignment accuracy of multiple point clouds, resulting in stronger consistency during batch scanning and stitching, meeting the data requirements for high-precision replication of cartoon toys and industrial mass production.

[0117] The minimum scan step distance of the extended curved surface satisfies the following formula:

[0118] ;

[0119] Where L is the effective length of the extension section, which ranges from 1 to 10 mm;

[0120] N is the number of single-turn scanning angles in the extended section, ranging from 24 to 72.

[0121] K represents the number of axial scanning layers in the extended section, ranging from 8 to 32.

[0122] Specifically, the scanning step size quantization calculation formula enables the standardization and precise setting of the scanning parameters for the extension segment, avoiding the loss of detail or data redundancy caused by manual experience-based settings. Furthermore, the step size is dynamically calculated by combining three factors: extension segment length, number of scanning angles per circle, and number of axial scanning layers, accurately matching the scanning requirements of extension segments of different sizes, balancing detail reproduction and scanning efficiency. The parameter range setting ensures complete acquisition of the extension segment surface, providing a complete point cloud foundation for high-precision modeling. In addition, quantizing the minimum scanning step size can avoid over-scanning, reducing the amount of point cloud data while ensuring accuracy, improving subsequent processing speed, and adapting to the efficient replication needs of complex extension structures in cartoon toys.

[0123] Example 3:

[0124] An electronic device, comprising:

[0125] One or more processors;

[0126] Memory, used to store one or more programs;

[0127] When the one or more programs are executed by the one or more processors, the one or more processors implement a method for reverse modeling and replicating 3D scanning of cartoon character toys.

[0128] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for reverse modeling and replicating a cartoon character toy using 3D scanning, characterized in that: The method for 3D scanning reverse modeling and replication of cartoon character toys includes the following steps: Step S1: Select the prototype of the cartoon character toy to be replicated, remove the detachable parts, and spray a matte auxiliary agent on the detailed feature areas of the cartoon character prototype. Step S2: Using a 3D laser scanner, scan the prototype of the cartoon character toy, and specifically scan the detachable parts and detailed feature areas to obtain 3D point cloud data. Perform trajectory fitting processing on the edge extension positions of the detachable parts, and perform weighted data stitching on the 3D point cloud data. Step S3: Import the 3D point cloud data obtained in step S2 into the reverse modeling system for reverse modeling processing; Step S4: Based on the optimized 3D model obtained in step S3, select suitable processing materials and perform replication processing; Step S5: After the surface treatment is completed, the key positions of the replica are scanned using a 3D scanner and compared with the scan data and 3D model of the original cartoon character toy to be replicated.

2. The method for reverse modeling and replicating a cartoon character toy using 3D scanning according to claim 1, characterized in that: The detailed feature region-specific scanning in step S2 includes the following steps: Step S21: Adjust the scanning resolution according to the detailed feature area. Set the scanning resolution to 0.01-0.03mm. Set the scanning resolution of the regular surface of the cartoon character toy prototype after removing the detachable parts to 0.05-0.1mm. Step S22: Use corresponding viewing angles to scan the detailed feature areas. Use a surround scan for the joint connections in the detailed feature areas, rotating the scanning angle around the joint to capture the joint's inner and outer connection arc, fitting gap, and limiting structure. Step S23: After completing a specialized scan of each detailed feature area, collect specialized scan data simultaneously, use a point cloud stitching algorithm to stitch together different scan point cloud data, calibrate the point cloud data of the detailed feature area during cloud data stitching, and generate complete 3D point cloud data of the cartoon prototype, detachable parts and detailed feature areas after scanning is completed, and mark the detailed feature areas. Step S23: Perform layered progressive scanning on the cartoon prototype, detachable parts, and detailed feature areas. During scanning, scan in layers along the joint axis, with each layer performed at a minimum scanning step distance d=0.008-0.01mm. For the hidden positions of detailed feature areas with a depth of 0.5-5mm, use a small-angle side-lit light source to illuminate and tighten the scanning position. Use a curvature fitting algorithm to complete the missing hidden points, and output a complete point cloud including extensions and hidden positions, which is marked as a high-precision modeling area.

3. The method for reverse modeling and replicating a cartoon character toy using 3D scanning according to claim 2, characterized in that: The reverse modeling process in step S3 includes the following steps: Step S31: Extract the complete 3D point cloud data of the detailed feature area separately, remove noise points and isolated points from the 3D point cloud data, complete the missing 3D point cloud data, and simplify the 3D point cloud data of the cartoon prototype and detachable parts. Step S32: Reconstruct the overall contour surface of the cartoon prototype and detachable parts, and then perform specific surface reconstruction on the detailed feature areas; Step S33: Integrate the surface reconstruction data of the cartoon prototype, detachable parts, and detailed feature areas; adjust the relative positions of the cartoon prototype, detachable parts, and detailed feature areas; perform accuracy testing on the integrated 3D model; and make fine adjustments and corrections based on the scanning data of the cartoon character toy prototype to be replicated. Step S34: Optimize the completed 3D model, and smooth the edge connection positions of detachable parts and detailed feature areas; Step S35: Reconstruct the surface of the completed 3D model according to the principle of curvature continuity, and perform trajectory fitting on the extended segment.

4. The method for reverse modeling and replicating a cartoon character toy using 3D scanning according to claim 3, characterized in that: The replication method in step S4 includes the following steps: Step S41: Adjust the parameters of the processing equipment according to the structural complexity and tolerance requirements of the cartoon prototype, detachable parts, and detailed feature areas; Step S42: Decompose the 3D model into cartoon prototype, detachable parts, and detailed feature areas, and adjust the processing parameters to avoid processing deviations; Step S43: Assemble the finished cartoon prototype, detachable parts, and detailed feature areas; Step S44: Perform surface treatment on the assembled replica product, remove burrs, and polish the detailed feature areas.

5. The method for reverse modeling and replicating a cartoon character toy using 3D scanning according to claim 4, characterized in that: The trajectory fitting process for the edge extension position of the removable component in step S2 includes the following steps: Step S221: Establish a local cylindrical coordinate system with the joint position center of the detachable component as the origin. Use a 3D laser scanner to identify the curvature center, rotation axis, tangential normal and transition fillet of the joint position of the detachable component. Extract the starting end face boundary contour, tangential vector and normal vector of the edge extension position of the detachable component as the forced constraint benchmark for trajectory fitting. Step S222: Divide the joint position point cloud data of the detachable component into a head fitting segment, a main body extension segment, and a tail transition segment along the joint axis. Generate an initial trajectory according to the axial direction of the main body extension segment. Force the normal of the trajectory start end to be consistent with the normal of the joint end face to avoid abrupt changes in the angle. Calculate the continuous data of the first and second derivatives of the trajectory. If the continuous data of the first and second derivatives are not satisfied, automatically iterate and correct the control points. Step S223: Identify the width of the fusion zone between the main body extension and the cartoon prototype, make the curvature at the entrance of the main body extension consistent with the curvature at the exit of the joint position, then perform a linear curvature gradient within the fusion zone, and perform a 360° rotational interference simulation with the center of the joint position. Step S224: Using the fitted trajectory as a skeleton, sweep radially to generate an extended surface segment, and output the selectable surface segment.

6. The method for reverse modeling and replicating a cartoon character toy using 3D scanning according to claim 5, characterized in that: The cartoon prototype, detachable parts, and detailed feature areas are weighted during cloud stitching. The weighting formula is as follows: ； in, The region weights are the region weights of the detailed feature regions. 3-5; Area weighting of cartoon prototypes and detachable parts =1; The total registration error of the weighted point cloud data; R represents the rotational transformation relationship of the point cloud data in three-dimensional space; T represents the translational transformation relationship of the point cloud data in three-dimensional space; Calculate and accumulate the data for all point cloud data to be stitched together, numbered from 1 to n. i is the index of a single point in the point cloud data; n represents the total number of valid point clouds participating in the stitching calculation; Let i be the i-th 3D coordinate in the target point cloud data; Let be the three-dimensional coordinates of the i-th point in the source point cloud data.

7. The method for reverse modeling and replicating a cartoon character toy using 3D scanning according to claim 6, characterized in that: The minimum scan step distance of the extended curved surface satisfies the following formula: ； Where L is the effective length of the extension section, which ranges from 1 to 10 mm; N is the number of single-turn scanning angles in the extended section, ranging from 24 to 72. K represents the number of axial scanning layers in the extended section, ranging from 8 to 32.

8. The method for reverse modeling and replicating a cartoon character toy using 3D scanning according to claim 7, characterized in that: In step S1, the matte scanning aid is a nano-sized silica matte spray, with the spray thickness controlled at 0.01-0.05 mm. After spraying, it is left to stand for 3-5 minutes until completely dry.

9. The method for reverse modeling and replicating a cartoon character toy using 3D scanning according to claim 8, characterized in that: In step S5, the comparison detection uses 3D comparison analysis. When the color difference map shows a deviation exceeding ±0.02mm, the abnormal area is automatically marked and the modeling step is returned for correction.

10. An electronic device, characterized in that, include: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement a method for reverse modeling and replicating a cartoon character toy using 3D scanning as described in any one of claims 1-9.

Images