A 3D printing modeling method based on digital human three-dimensional model

Abstract

The present invention discloses a 3D printing modeling method based on a digital human three-dimensional model, which relates to the field of 3D printing technology, including real-time collection of point cloud data and preprocessing, extracting features of the preprocessed point cloud data and constructing a connection network; generating a digital human three-dimensional model based on the connection network, slicing the digital human three-dimensional model to generate a printing path, monitoring the 3D printer operating parameters in real time and adjusting them; creating a visual interface to display the 3D printing modeling results of the digital human three-dimensional model in real time, and storing the digital human data generated during the collection and analysis process. The present invention collects point cloud data in real time, extracts features of the point cloud data and constructs a connection network to generate a digital human three-dimensional model, slices the digital human three-dimensional model to generate a printing path, and monitors the 3D printer operating parameters in real time and adjusts them, which not only improves data processing efficiency, but also ensures the quality and accuracy of the generated model, and significantly improves the stability of the printing process and the quality of the final product.

Description

Technical Field

[0001] The present invention relates to the technical field of 3D printing, and in particular to a 3D printing modeling method based on a digital human three-dimensional model. Background Art

[0002] In recent years, with the rapid development of 3D printing technology, its application in various fields such as medicine, manufacturing, and construction has become increasingly widespread. The technology of generating digital human models based on 3D scanning technology and 3D printing has gradually become a research hotspot. Traditional 3D printing technology is mostly used to manufacture mechanical parts, prototype models, etc., but with the improvement of scanning technology and data processing capabilities, it has become possible to use 3D scanning technology to generate high-precision digital human models and then materialize them through 3D printing technology. The development of this technology has brought new opportunities for personalized medicine in the medical field, customized prostheses, and virtual reality applications in the fields of education and entertainment.

[0003] Although the current 3D printing technology has made certain progress in realizing the printing of digital human models, there are still many shortcomings. The existing technology usually uses the ICP algorithm for point cloud registration. Although the algorithm is mature, it has high computational complexity and low efficiency when processing large-scale point cloud data, and lacks real-time monitoring and adjustment functions. The automation and intelligence level of the existing technology is low, and it is impossible to make timely and effective adjustments based on problems that arise during the printing process, resulting in difficulty in ensuring printing quality. Summary of the invention

[0004] In view of the above problems existing in the existing 3D printing modeling method based on digital human three-dimensional model, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is that the existing technology usually adopts the ICP algorithm for point cloud registration. Although the algorithm is mature, it has high computational complexity and low efficiency when processing large-scale point cloud data, and lacks real-time monitoring and adjustment functions. The automation and intelligence level of the existing technology is low, and it is impossible to make timely and effective adjustments according to problems arising during the printing process, resulting in difficulty in ensuring printing quality.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a 3D printing modeling method based on a digital human three-dimensional model, which includes collecting point cloud data in real time and preprocessing it, extracting the features of the preprocessed point cloud data and building a connection network; generating a digital human three-dimensional model based on the connection network, slicing the digital human three-dimensional model to generate a printing path, and monitoring and adjusting the operating parameters of the 3D printer in real time; creating a visual interface to display the 3D printing modeling results of the digital human three-dimensional model in real time, and storing the digital human data generated during the collection and analysis process.

[0007] As a preferred solution of the 3D printing modeling method based on the digital human three-dimensional model of the present invention, wherein: the real-time collection of point cloud data and pre-processing refers to using a full-body 360-degree high-precision 3D scanning device combining infrared and optical scanning technology to perform all-round scanning and collect point cloud data;

[0008] Use Gaussian filtering algorithm to denoise point cloud data;

[0009] The point cloud registration technology ICP algorithm is used to align and merge the point cloud data to form three-dimensional point cloud data.

[0010] As a preferred solution of the 3D printing modeling method based on the digital human three-dimensional model of the present invention, wherein: the extraction of features of the pre-processed three-dimensional point cloud data and the construction of a connection network refers to using a covariance matrix analysis method in a local area to calculate the eigenvalue and eigenvector of each point in the pre-processed three-dimensional point cloud data, and selecting the point with the largest eigenvalue as the feature point;

[0011] For each feature point identified by the ISS algorithm, the function in the point cloud library is used to calculate the curvature k of each feature point. The formula is:

[0012] ,

[0013] in For feature points Neighboring feature point area, For feature points Its neighboring points The Euclidean distance between Feature point area The number of feature points in

[0014] Randomly select sample points from the historical data of the 3D point cloud that has been scanned and feature point extracted;

[0015] Defining the objective function using multi-objective optimization , the formula is:

[0016] ,

[0017] Where M is the total number of connections, Mz is the number of correct connections, and Mc is the number of incorrect connections. is the balance parameter;

[0018] Randomly generate weights and The combination of is taken as the initial combination;

[0019] Set target thresholds based on historical connection data and calculate the objective function for each weight combination , select the weight combination with the largest objective function value for crossover and mutation, generate a new generation of weight combinations, repeat fitness evaluation, selection, crossover and mutation until the objective function value reaches the preset target threshold, stop iteration, and obtain the optimized weight combination and ;

[0020] The number of neighbors connected to each feature point is obtained based on spatial distance and curvature, and the KNN algorithm is used to calculate the similarity score between each pair of feature points. , the formula is:

[0021] ,

[0022] in and are the standard deviation parameters of spatial distance and curvature, and is the curvature value;

[0023] Based on similarity score The similarity score threshold is set to 2 times the standard deviation The similarity score Compare with the similarity score threshold. If the similarity score When the similarity score threshold is exceeded, a connection is established between two feature points, and a feature point connection network is created using the graph library NetworkX, where nodes represent feature points and edges represent connections.

[0024] As a preferred solution of the 3D printing modeling method based on the digital human 3D model of the present invention, wherein: the generating of the 3D model based on the connection network refers to extracting the coordinates of all connection points from the feature point connection network to generate coordinate point cloud data in the format of PLY;

[0025] Use the local plane fitting method to estimate the normal vector of each point cloud data point, and synthesize the gradient vectors of all points into a global gradient field V;

[0026] The gradient field is used as the right side of the Poisson equation, and the discrete Fourier transform is used to solve the Poisson equation. The formula is:

[0027] ,

[0028] Where f is the implicit surface function, is the vector differential operator, V is the global gradient field;

[0029] Extracting a three-dimensional surface from an implicit surface function f;

[0030] Use Marching Cubes algorithm to convert the 3D surface into a 3D mesh model of a digital human;

[0031] Use Laplace smoothing algorithm to smooth the 3D mesh model of digital human;

[0032] Detect and repair holes and thin-walled sections in your models using automatic repair algorithms.

[0033] As a preferred solution of the 3D printing modeling method based on the digital human 3D model of the present invention, wherein: the slicing process of the digital human 3D model to generate the printing path refers to converting the 3D mesh model into STL format using the 3D modeling software Autodesk Maya;

[0034] Use Cura to slice the 3D mesh model and generate the printing path .

[0035] As a preferred solution of the 3D printing modeling method based on the digital human three-dimensional model of the present invention, wherein: the real-time monitoring of the 3D printer operating parameters and adjustment refers to collecting the operating parameters of the 3D printer in real time during the printing process, calculating the operating status index , the formula is:

[0036] ,

[0037] in is the g-th real-time operation parameter, is the corresponding g-th real-time standard parameter value, G is the total number of real-time operating parameters;

[0038] The real-time collection of 3D printer operating parameters is standardized to obtain , calculate the current quality assessment value , the formula is:

[0039] ,

[0040] in is the standardized value of the gth parameter;

[0041] Collect the quality assessment data of historical digital human 3D models, pre-process the historical quality assessment data, use the statistical analysis tool SPSS to analyze the collected historical quality assessment data, and calculate the mean of the historical quality assessment data as the target quality assessment value ;

[0042] Based on historical data, set the threshold T. If the running status indicator > threshold T, the printing parameters are automatically adjusted, the formula is:

[0043] ,

[0044] in For the adjusted printing parameters, is the printing path, h is the adjustment coefficient, is the target quality assessment value, is the current quality assessment value;

[0045] Collect historical printing data including historical operating parameters and corresponding quality assessment data, and use moving average filtering for smoothing;

[0046] The processed historical printing data is analyzed by linear regression to obtain the quality of the historical operating parameters. As the dependent variable, the g-th historical operating parameter As the independent variable, the formula is:

[0047] ,

[0048] in is the intercept, is the regression coefficient;

[0049] Use the least squares method OLS to perform regression analysis and calculate the regression coefficients , the formula is:

[0050] ,

[0051] in is the quality corresponding to the g-th historical operating parameter, and is the average value of historical operating parameters and corresponding quality;

[0052] The regression coefficient As the adjustment factor h;

[0053] If the operating status indicator When the threshold value T is reached, monitoring continues.

[0054] As a preferred solution of the 3D printing modeling method based on the digital human 3D model of the present invention, wherein: the creating a visual interface to display the 3D printing modeling result of the digital human 3D model in real time refers to creating an interactive interface using the visualization tool Python;

[0055] Transmit real-time data to the visualization interface through the API interface;

[0056] The setting visualization interface includes the printing parameter display area, adjustment control area and printing progress area.

[0057] As a preferred solution of the 3D printing modeling method based on the digital human three-dimensional model described in the present invention, wherein: the data generated in the storage collection and analysis process refers to storing the collected point cloud data and the digital human data generated by the analysis in a database, the database classifies and stores the digital human data according to the timestamp data and implements encryption protection measures, synchronously uploads the stored data to the cloud for backup, and regularly performs integrity checks on the stored data and backup data.

[0058] A computer device comprises: a memory and a processor; the memory stores a computer program, and the processor implements the steps of the above-mentioned 3D printing modeling method based on a digital human three-dimensional model when executing the computer program.

[0059] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned 3D printing modeling method based on a digital human three-dimensional model.

[0060] The beneficial effects of the present invention are as follows: the present invention collects point cloud data in real time and performs preprocessing, extracts features of the preprocessed point cloud data and constructs a connection network, generates a three-dimensional model of a digital human based on the connection network, slices the three-dimensional model of the digital human to generate a printing path, and monitors and adjusts the operating parameters of the 3D printer in real time, which not only improves the data processing efficiency, but also ensures the quality and accuracy of the generated model, significantly improving the stability of the printing process and the quality of the final product. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0062] Figure 1 It is a flowchart of the 3D printing modeling method based on the digital human three-dimensional model.

[0063] Figure 2 It is a structural schematic diagram of the 3D printing modeling method based on the digital human three-dimensional model. DETAILED DESCRIPTION

[0064] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings.

[0065] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0066] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0067] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and which provides a 3D printing modeling method based on a digital human 3D model. The 3D printing modeling method based on a digital human 3D model includes:

[0068] S1, collect point cloud data in real time and preprocess it, extract the features of the preprocessed point cloud data and build a connection network;

[0069] Specifically, real-time collection of point cloud data and pre-processing refers to using a full-body 360-degree high-precision 3D scanning device that combines infrared and optical scanning technologies to perform all-round scanning and collect point cloud data;

[0070] Use Gaussian filtering algorithm to denoise point cloud data;

[0071] The point cloud registration technology ICP algorithm is used to align and merge the point cloud data to form three-dimensional point cloud data.

[0072] The use of full-body 360-degree high-precision 3D scanning equipment that combines infrared and optical scanning technology not only improves the accuracy and comprehensiveness of data collection, but also significantly reduces blind areas in the data collection process, ensuring the basic data quality for subsequent processing. By smoothing the point cloud data, the random noise generated during the scanning process can be effectively removed, the purity of the data can be improved, the edge detail information in the data can be retained, and the loss of details caused by excessive smoothing can be avoided, ensuring the reliability and accuracy of the data, providing high-quality basic data for subsequent point cloud alignment and model construction. The formation of consistent three-dimensional point cloud data not only improves the overall consistency of the data, but also enhances the accuracy and integrity of the model. The ICP algorithm can effectively handle the displacement and rotation differences between different perspectives, ensuring that the final generated three-dimensional point cloud data has high accuracy and consistency.

[0073] Further, extracting the features of the preprocessed three-dimensional point cloud data and constructing a connection network means using a covariance matrix analysis method in a local area to calculate the eigenvalue and eigenvector of each point in the preprocessed three-dimensional point cloud data, and selecting the point with the largest eigenvalue as the feature point;

[0074] For each feature point identified by the ISS algorithm, the function in the point cloud library is used to calculate the curvature k of each feature point. The formula is:

[0075] ,

[0076] in For feature points Neighboring feature point area, For feature points Its neighboring points The Euclidean distance between Feature point area The number of feature points in

[0077] Randomly select sample points from the historical data of the 3D point cloud that has been scanned and feature point extracted;

[0078] Defining the objective function using multi-objective optimization , the formula is:

[0079] ,

[0080] Where M is the total number of connections, Mz is the number of correct connections, and Mc is the number of incorrect connections. is the balance parameter;

[0081] Randomly generate weights and The combination of is taken as the initial combination;

[0082] Set target thresholds based on historical connection data and calculate the objective function for each weight combination , select the weight combination with the largest objective function value for crossover and mutation, generate a new generation of weight combinations, repeat fitness evaluation, selection, crossover and mutation until the objective function value reaches the preset target threshold, stop iteration, and obtain the optimized weight combination and ;

[0083] The number of neighbors connected to each feature point is obtained based on spatial distance and curvature, and the KNN algorithm is used to calculate the similarity score between each pair of feature points. , the formula is:

[0084] ,

[0085] in and are the standard deviation parameters of spatial distance and curvature, and is the curvature value;

[0086] Based on similarity score The similarity score threshold is set to 2 times the standard deviation The similarity score Compare with the similarity score threshold. If the similarity score When the similarity score threshold is exceeded, a connection is established between two feature points, and a feature point connection network is created using the graph library NetworkX, where nodes represent feature points and edges represent connections.

[0087] Extracting the salient features of point cloud data ensures that the selection of feature points is highly accurate and representative, thereby improving the reliability of subsequent processing. Accurate extraction of feature points is crucial to building a high-quality connection network because it directly affects the accuracy and stability of the connection network. Accurate calculation of curvature values ​​helps to improve the accuracy of the connection network. By defining the objective function and optimizing it, the optimal weight combination can be determined to maximize the correct connection ratio and minimize the incorrect connection ratio, ensuring the accuracy and robustness of the connection network. The calculation of the similarity score takes into account the standard deviation parameters of spatial distance and curvature, and can fully reflect the geometric relationship between feature points. The connection network constructed in this way has high accuracy and stability, and can build a reliable three-dimensional model in complex three-dimensional point cloud data.

[0088] S2, generating a three-dimensional model of a digital human based on the connection network, slicing the three-dimensional model of the digital human to generate a printing path, and monitoring and adjusting the operating parameters of the 3D printer in real time;

[0089] Specifically, generating a three-dimensional model based on a connection network refers to extracting the coordinates of all connection points from the feature point connection network and generating coordinate point cloud data in a PLY format;

[0090] Use the local plane fitting method to estimate the normal vector of each point cloud data point, and synthesize the gradient vectors of all points into a global gradient field V;

[0091] The gradient field is used as the right side of the Poisson equation, and the discrete Fourier transform is used to solve the Poisson equation. The formula is:

[0092] ,

[0093] Where f is the implicit surface function, is the vector differential operator, V is the global gradient field;

[0094] Extracting a three-dimensional surface from an implicit surface function f;

[0095] Use Marching Cubes algorithm to convert the 3D surface into a 3D mesh model of a digital human;

[0096] Use Laplace smoothing algorithm to smooth the 3D mesh model of digital human;

[0097] Detect and repair holes and thin-walled sections in your models using automatic repair algorithms.

[0098] By synthesizing the gradient field and using it as the right side of the Poisson equation, the Poisson equation is solved using discrete Fourier transform and implicit surface functions are efficiently generated, which not only improves the accuracy of 3D model generation, but also significantly reduces the calculation time. It is suitable for large-scale 3D data processing and solves the problem of high computational complexity in traditional methods. The Marching Cubes algorithm searches for isosurfaces in the 3D grid and generates triangular patches to represent the surface, ensuring the geometric details and integrity of the model. When processing complex surfaces and 3D data with rich details, it can provide high-quality models and provide a good foundation for subsequent smoothing and repair. The Laplace smoothing algorithm is used to smooth the 3D grid model of the digital human, which can eliminate noise and jagged edges on the model surface and make the model surface smoother. The automatic repair algorithm is used to detect and repair holes and thin-walled parts in the model to ensure the integrity and stability of the model, which not only improves the visual quality of the model, but also enhances the operability and printing quality of the model in the actual 3D printing process, solving the problem of low model quality and easy defects in traditional methods.

[0099] Further, slicing the digital human three-dimensional model to generate a printing path refers to converting the three-dimensional mesh model into an STL format using 3D modeling software Autodesk Maya;

[0100] Use Cura to slice the 3D mesh model and generate the printing path .

[0101] Through the efficient processing capabilities of Maya, the accuracy and integrity of the model conversion process are ensured, which not only improves the efficiency of data conversion, but also ensures the operability of the model during the printing process, and solves the compatibility issues between different software and devices. The efficient slicing algorithm of Cura software can automatically optimize the printing path of each layer, reduce printing time and improve printing quality. Through the generated printing path, the 3D printer can build the model layer by layer to ensure the printing accuracy and consistency of each layer, solving the problems of insufficient path optimization and low printing efficiency in traditional slicing methods, and significantly improving the overall performance of 3D printing.

[0102] Furthermore, real-time monitoring and adjustment of 3D printer operating parameters means collecting the operating parameters of the 3D printer in real time during the printing process and calculating the operating status indicators. , the formula is:

[0103] ,

[0104] in is the g-th real-time operation parameter, is the corresponding g-th real-time standard parameter value, G is the total number of real-time operating parameters;

[0105] The real-time collection of 3D printer operating parameters is standardized to obtain , calculate the current quality assessment value , the formula is:

[0106] ,

[0107] in is the standardized value of the gth parameter;

[0108] Collect the quality assessment data of historical digital human 3D models, pre-process the historical quality assessment data, use the statistical analysis tool SPSS to analyze the collected historical quality assessment data, and calculate the mean of the historical quality assessment data as the target quality assessment value ;

[0109] Based on historical data, set the threshold T. If the running status indicator > threshold T, the printing parameters are automatically adjusted, the formula is:

[0110] ,

[0111] in For the adjusted printing parameters, is the printing path, h is the adjustment coefficient, is the target quality assessment value, is the current quality assessment value;

[0112] Collect historical printing data including historical operating parameters and corresponding quality assessment data, and use moving average filtering for smoothing;

[0113] The processed historical printing data is analyzed by linear regression to obtain the quality of the historical operating parameters. As the dependent variable, the g-th historical operating parameter As the independent variable, the formula is:

[0114] ,

[0115] in is the intercept, is the regression coefficient;

[0116] Use the least squares method OLS to perform regression analysis and calculate the regression coefficients , the formula is:

[0117] ,

[0118] in is the quality corresponding to the g-th historical operating parameter, and is the average value of historical operating parameters and corresponding quality;

[0119] The regression coefficient As the adjustment factor h;

[0120] If the operating status indicator When the threshold value T is reached, monitoring continues.

[0121] Through this real-time monitoring, anomalies in the printing process can be quickly discovered to prevent quality problems from occurring. The standardized parameters can comprehensively reflect the overall status of the printer and provide accurate operating status evaluation. Calculating operating status indicators and quality evaluation values ​​not only provides an objective evaluation standard, but also sets scientific and reasonable quality target values ​​through analysis of historical data, solving the problem of traditional quality control methods relying on manual judgment and lacking objective standards, significantly improving the accuracy and efficiency of quality control. The automatic adjustment mechanism avoids quality problems caused by delayed human intervention, ensuring the continuity of the printing process and the high quality of printed products. The linear regression analysis model can accurately predict printing quality and guide parameter adjustment. This data-driven approach enhances the controllability and predictive power of the printing process and significantly improves the stability and consistency of printing quality.

[0122] S3, create a visual interface to display the 3D printing modeling results of the digital human three-dimensional model in real time, and store the digital human data generated during the collection and analysis process;

[0123] Further, creating a visualization interface to display the 3D printing modeling results of the digital human three-dimensional model in real time refers to creating an interactive interface using a visualization tool Python;

[0124] Transmit real-time data to the visualization interface through the API interface;

[0125] The setting visualization interface includes the printing parameter display area, adjustment control area and printing progress area.

[0126] Using Python's visualization tools, complex data can be graphically displayed and user interacted with, which not only improves the intuitiveness and aesthetics of data display, but also enhances the interactivity between users and the printing process, solving the problems of non-intuitive information presentation and complex operations in traditional display methods. The API interface can efficiently and accurately transmit data, ensuring that the information displayed on the visualization interface is the latest, so that users can understand the equipment operation status and printing progress in real time during the printing process, and promptly discover and solve potential problems. The print parameter display area is used to display the current print settings and status information, which is convenient for users to monitor the printing process in real time. The adjustment control area provides users with an interface for adjusting print parameters, so that users can adjust print settings in time according to actual conditions. The print progress area displays the printing progress, which is convenient for users to understand the printing completion status. This design not only improves the user's control over the printing process, but also improves the success rate and quality of printing, solving the problems of scattered information and inconvenient operation in traditional methods.

[0127] Furthermore, storing the data generated during the collection and analysis process means storing the collected point cloud data and the digital human data generated by the analysis in a database. The database classifies and stores the digital human data according to timestamp data and implements encryption protection measures, simultaneously uploads the stored data to the cloud for backup, and regularly performs integrity checks on the stored data and backup data.

[0128] The use of databases improves the efficiency of data storage and retrieval, ensuring efficient use of data. For large-scale point cloud data and complex digital human data, databases provide a reliable data management foundation, avoid data confusion and difficulty in retrieval, and ensure the validity and availability of data. The combination of timestamps and classified storage improves the organization and management efficiency of data, ensuring that data can be quickly located and used when needed. Encryption protection is an important means of data security management. By encrypting sensitive data, the confidentiality and integrity of data are ensured. Cloud backup ensures high availability and security of data, and provides a data disaster recovery solution. Integrity testing is an important means to ensure the long-term validity and reliability of data. Through regular testing, data corruption or loss problems can be discovered and repaired in a timely manner.

[0129] Embodiment 2, if the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0130] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0131] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0132] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit with a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit with a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

Claims

1. A 3D printing modeling method based on a digital human three-dimensional model, characterized in that: include, Collect and preprocess point cloud data in real time, extract features of preprocessed point cloud data and build a connection network; Generate a 3D model of a digital human based on the connection network, slice the 3D model of the digital human to generate a printing path, and monitor the operating parameters of the 3D printer in real time and make adjustments; Create a visual interface to display the 3D printing modeling results of the digital human three-dimensional model in real time, and store the digital human data generated during the collection and analysis process; The real-time collection of point cloud data and pre-processing refers to using a full-body 360-degree high-precision 3D scanning device that combines infrared and optical scanning technologies to perform all-round scanning and collect point cloud data; Use Gaussian filtering algorithm to denoise point cloud data; Use the ICP algorithm of point cloud registration technology to align and merge point cloud data to form three-dimensional point cloud data; Extracting the features of the pre-processed three-dimensional point cloud data and constructing a connection network refers to using a covariance matrix analysis method in a local area to calculate the eigenvalue and eigenvector of each point in the pre-processed three-dimensional point cloud data, and selecting the point with the largest eigenvalue as the feature point; For each feature point identified by the ISS algorithm, the function in the point cloud library is used to calculate the curvature k of each feature point. The formula is: ， in For feature points Neighboring feature point area, For feature points Its neighboring points The Euclidean distance between Feature point area The number of feature points in Randomly select sample points from the historical data of the 3D point cloud that has been scanned and feature point extracted; Defining the objective function using multi-objective optimization , the formula is: ， Where M is the total number of connections, Mz is the number of correct connections, and Mc is the number of incorrect connections. is the balance parameter; Randomly generate weights and The combination of is taken as the initial combination; Set target thresholds based on historical connection data and calculate the objective function for each weight combination , select the weight combination with the largest objective function value for crossover and mutation, generate a new generation of weight combinations, repeat fitness evaluation, selection, crossover and mutation until the objective function value reaches the preset target threshold, stop iteration, and obtain the optimized weight combination and ; The number of neighbors connected to each feature point is obtained based on spatial distance and curvature, and the KNN algorithm is used to calculate the similarity score between each pair of feature points. , the formula is: ， in and are the standard deviation parameters of spatial distance and curvature, and is the curvature value; Based on similarity score The similarity score threshold is set to 2 times the standard deviation. Compare with the similarity score threshold. If the similarity score When the similarity score threshold is exceeded, a connection is established between two feature points, and a feature point connection network is created using the graph library NetworkX, where nodes represent feature points and edges represent connections.

2. The 3D printing modeling method based on the digital human three-dimensional model according to claim 1, characterized in that: Generating a three-dimensional model based on a connection network refers to extracting the coordinates of all connection points from the feature point connection network to generate coordinate point cloud data in a PLY format; Use the local plane fitting method to estimate the normal vector of each point cloud data point, and synthesize the gradient vectors of all points into a global gradient field V; The gradient field is used as the right side of the Poisson equation, and the discrete Fourier transform is used to solve the Poisson equation. The formula is: ， Where f is the implicit surface function, is the vector differential operator, V is the global gradient field; Extracting a three-dimensional surface from an implicit surface function f; Use Marching Cubes algorithm to convert the 3D surface into a 3D mesh model of a digital human; Use Laplace smoothing algorithm to smooth the 3D mesh model of digital human; Detect and repair holes and thin-walled sections in your models using automatic repair algorithms.

3. The 3D printing modeling method based on the digital human three-dimensional model according to claim 2, characterized in that: The slicing of the digital human 3D model to generate a printing path refers to converting the 3D mesh model into an STL format using the 3D modeling software Autodesk Maya; Use Cura to slice the 3D mesh model and generate the printing path .

4. The 3D printing modeling method based on the digital human three-dimensional model according to claim 3, characterized in that: The real-time monitoring and adjustment of the operating parameters of the 3D printer refers to collecting the operating parameters of the 3D printer in real time during the printing process and calculating the operating status indicators. , the formula is: ， in is the g-th real-time operation parameter, is the corresponding g-th real-time standard parameter value, G is the total number of real-time operating parameters; The real-time collection of 3D printer operating parameters is standardized to obtain , calculate the current quality assessment value , the formula is: ， in is the standardized value of the gth parameter; Collect the quality assessment data of historical digital human 3D models, pre-process the historical quality assessment data, use the statistical analysis tool SPSS to analyze the collected historical quality assessment data, and calculate the mean of the historical quality assessment data as the target quality assessment value ; Based on historical data, set the threshold T. If the running status indicator > threshold T, the printing parameters are automatically adjusted, the formula is: ， in For the adjusted printing parameters, is the printing path, h is the adjustment coefficient, is the target quality assessment value, is the current quality assessment value; Collect historical printing data including historical operating parameters and corresponding quality assessment data, and use moving average filtering for smoothing; The processed historical printing data is analyzed by linear regression to obtain the quality of the historical operating parameters. As the dependent variable, the g-th historical operating parameter As the independent variable, the formula is: ， in is the intercept, is the regression coefficient; Use the least squares method OLS to perform regression analysis and calculate the regression coefficients , the formula is: ， in is the quality corresponding to the g-th historical operating parameter, and is the average value of historical operating parameters and corresponding quality; The regression coefficient As the adjustment factor h; If the operating status indicator When the threshold value T is reached, monitoring continues.

5. The 3D printing modeling method based on the digital human three-dimensional model according to claim 4, characterized in that: The creating of a visualization interface to display the 3D printing modeling result of the digital human 3D model in real time refers to creating an interactive interface using the visualization tool Python; Transmit real-time data to the visualization interface through the API interface; The setting visualization interface includes the printing parameter display area, adjustment control area and printing progress area.

6. The 3D printing modeling method based on the digital human three-dimensional model according to claim 5, characterized in that: The data generated during the storage collection and analysis process refers to storing the collected point cloud data and the digital human data generated by the analysis in a database. The database classifies and stores the digital human data according to timestamp data and implements encryption protection measures, simultaneously uploads the stored data to the cloud for backup, and regularly performs integrity checks on the stored data and backup data.

7. A computer device comprising: Memory and processor; The memory stores a computer program, characterized in that: when the processor executes the computer program, the steps of the 3D printing modeling method based on a digital human three-dimensional model according to any one of claims 1 to 6 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the 3D printing modeling method based on a digital human three-dimensional model according to any one of claims 1 to 6 are implemented.

Images