Neural network-based slice model support effect prediction method and system

Abstract

The invention discloses a slice model supporting effect prediction method and system based on a neural network. The method comprises the steps that slice data of a model to be subjected to 3D printing and component operation of a user on the model are acquired; based on a support structure identification algorithm, identifying a support member in the model slice data; according to the component operation, based on a simulation algorithm, determining simulation change data corresponding to the model slice data; and predicting a support change effect of the model under the operation of the component based on a neural network algorithm according to the simulation change data and the support component. It can be seen that accurate supporting effect prediction based on user operation and data simulation can be achieved, the optimization accuracy and printing stability of the 3D printing model supporting structure are improved, and the supporting failure risk caused by misoperation is reduced.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for predicting the effect of slice model support based on neural networks. Background Technology

[0002] With the widespread application of 3D printing technology in high-precision manufacturing, enterprises and users are increasingly focusing on improving printing stability and success rates through optimized support structure design. Existing technologies typically acquire slice data from 3D models, identify support components using basic geometric analysis or fixed rules, and adjust the support design based on standard parameters to support the printing process under user operation. However, existing solutions lack dynamic analysis of user component operations and simulation prediction of support changes, making it difficult to accurately assess the impact of operations on the support structure. Commonly used static support design methods cannot adapt to complex operating scenarios, resulting in insufficient accuracy in support optimization. This makes them prone to support failure or printing defects due to improper operation, limiting the stability and printing efficiency of 3D printed models. Therefore, existing technologies have shortcomings that urgently need to be addressed. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method and system for predicting the support effect of sliced ​​models based on neural networks, which can realize accurate prediction of support effect based on user operation and data simulation, improve the optimization accuracy and printing stability of 3D printed model support structure, and reduce the risk of support failure caused by improper operation.

[0004] To address the aforementioned technical problems, the first aspect of this invention discloses a method for predicting the support effect of a slice model based on a neural network, the method comprising: Acquire slice data of the model to be 3D printed and user operations on the components of the model; Based on the support structure identification algorithm, the support components in the model slice data are identified; Based on the component operation and the simulation algorithm, the simulated change data corresponding to the model slice data is determined; Based on the simulated change data and the supporting components, a neural network algorithm is used to predict the support change effect of the model under the operation of the components.

[0005] As an optional implementation, in the first aspect of the present invention, the step of identifying support components in the model slice data based on the support structure identification algorithm includes: For each model slice in the model slice data, based on the image segmentation algorithm, multiple component cross-sectional envelope patterns belonging to different components in the model slice are segmented out; An image containing multiple cross-sectional envelope patterns of the components is input into a trained support component prediction neural network to obtain the probability of each cross-sectional envelope pattern of the components belonging to the support component. Based on the component probabilities corresponding to adjacent model slices, the supporting components in the model slice data are identified.

[0006] As an optional implementation, in the first aspect of the present invention, the supporting component prediction neural network is an RNN network architecture, which is trained using a training dataset that includes multiple cross-sectional images of training components and corresponding annotations indicating whether they belong to supporting components.

[0007] As an optional implementation, in a first aspect of the invention, identifying supporting components in the model slice data based on the component probabilities corresponding to adjacent model slices includes: The cross-sectional envelope pattern of the component whose probability is greater than a preset probability threshold is determined as the support component pattern; For any two adjacent model slices, calculate the graphic similarity between the supporting component graphics of the two model slices; Calculate the reciprocal of the positional difference between the support member graphics of the two model slices; Calculate the first probability average of the component probabilities between the support component graphics of the two model slices; The component correlation degree between the two model slices is obtained by summing the graphic similarity, the reciprocal, and the first probability average. Based on the supporting component graphic and the component correlation, the supporting components in the model slice data are identified.

[0008] As an optional implementation, in a first aspect of the invention, identifying the support member in the model slice data based on the support member graphic and the member correlation degree includes: For each preset component in the model slice data, determine the number of all the supporting component graphics corresponding to that preset component; Calculate the second probability average of the component probabilities of all the supporting component graphics corresponding to the preset component; Calculate the average correlation degree of the component correlation degree corresponding to all model slices containing the preset component; Calculate the product of the quantity, the second probability average, and the correlation average to obtain the possible support parameters corresponding to the preset component; The preset components whose possible support parameters are greater than a preset first parameter threshold are selected to obtain multiple support components in the model slice data.

[0009] As an optional implementation, in the first aspect of the present invention, the component operation includes at least one of component position change operation, component physical parameter change operation, component replacement operation, component addition operation, and component deletion operation.

[0010] As an optional implementation, in the first aspect of the invention, determining the simulated change data corresponding to the model slice data based on the component operation and a simulation algorithm includes: Determine the target active component in the model slice data corresponding to the component operation; The component operation is input into the trained simulation prediction algorithm model to obtain the predicted change parameters corresponding to the component operation; the simulation prediction algorithm model is trained using a training dataset that includes multiple training component operations and corresponding component parameter change annotations. The predicted change parameters are applied to the target component to obtain the simulated change data corresponding to the model slice data.

[0011] As an optional implementation, in the first aspect of the present invention, the step of predicting the support change effect of the model under the operation of the component based on the simulated change data and the support component using a neural network algorithm includes: Determine at least one associated component by identifying the support component whose positional distance from the target component is less than a preset distance threshold; The predicted change parameters, along with the component parameters of the target component and all associated components, are input into a trained support effect change prediction algorithm model to obtain the support change effect of the model under the component operation. The support effect change prediction algorithm model is trained using a training dataset that includes multiple training component combination data and corresponding specific component change parameter annotations and support effect annotations. The component parameters include at least one of component position, component surface area, and component volume. The support change effect is an increase in support strength, an increase in support fragility, or a risk of support collapse.

[0012] A second aspect of this invention discloses a slice model support effect prediction system based on a neural network, the system comprising: The acquisition module is used to acquire slice data of the model to be 3D printed and user operations on the components of the model; The identification module is used to identify the supporting components in the model slice data based on the supporting structure identification algorithm; The determination module is used to determine the simulated change data corresponding to the model slice data based on the component operation and a simulation algorithm. The prediction module is used to predict the support change effect of the model under the operation of the component based on the simulated change data and the support component, using a neural network algorithm.

[0013] As an optional implementation, in a second aspect of the invention, the identification module identifies the specific method by which it identifies the support components in the model slice data based on a support structure identification algorithm, including: For each model slice in the model slice data, based on the image segmentation algorithm, multiple component cross-sectional envelope patterns belonging to different components in the model slice are segmented out; An image containing multiple cross-sectional envelope patterns of the components is input into a trained support component prediction neural network to obtain the probability of each cross-sectional envelope pattern of the components belonging to the support component. Based on the component probabilities corresponding to adjacent model slices, the supporting components in the model slice data are identified.

[0014] As an optional implementation, in a second aspect of the invention, the supporting component prediction neural network is an RNN network architecture, which is trained using a training dataset that includes multiple cross-sectional images of training components and corresponding annotations indicating whether they belong to supporting components.

[0015] As an optional implementation, in a second aspect of the invention, the specific method by which the identification module identifies supporting components in the model slice data based on the component probabilities corresponding to adjacent model slices includes: The cross-sectional envelope pattern of the component whose probability is greater than a preset probability threshold is determined as the support component pattern; For any two adjacent model slices, calculate the graphic similarity between the supporting component graphics of the two model slices; Calculate the reciprocal of the positional difference between the support member graphics of the two model slices; Calculate the first probability average of the component probabilities between the support component graphics of the two model slices; The component correlation degree between the two model slices is obtained by summing the graphic similarity, the reciprocal, and the first probability average. Based on the supporting component graphic and the component correlation, the supporting components in the model slice data are identified.

[0016] As an optional implementation, in a second aspect of the invention, the identification module identifies the specific method by which it identifies the support component in the model slice data based on the support component graphic and the component correlation degree, including: For each preset component in the model slice data, determine the number of all the supporting component graphics corresponding to that preset component; Calculate the second probability average of the component probabilities of all the supporting component graphics corresponding to the preset component; Calculate the average correlation degree of the component correlation degree corresponding to all model slices containing the preset component; Calculate the product of the quantity, the second probability average, and the correlation average to obtain the possible support parameters corresponding to the preset component; The preset components whose possible support parameters are greater than a preset first parameter threshold are selected to obtain multiple support components in the model slice data.

[0017] As an optional implementation, in a second aspect of the present invention, the component operation includes at least one of component position change operation, component physical parameter change operation, component replacement operation, component addition operation, and component deletion operation.

[0018] As an optional implementation, in a second aspect of the invention, the determining module determines the specific method by which, based on the component operation and a simulation algorithm, the simulated change data corresponding to the model slice data is determined, including: Determine the target active component in the model slice data corresponding to the component operation; The component operation is input into the trained simulation prediction algorithm model to obtain the predicted change parameters corresponding to the component operation; the simulation prediction algorithm model is trained using a training dataset that includes multiple training component operations and corresponding component parameter change annotations. The predicted change parameters are applied to the target component to obtain the simulated change data corresponding to the model slice data.

[0019] As an optional implementation, in a second aspect of the invention, the prediction module predicts the specific method by which, based on the simulated change data and the supporting component, and using a neural network algorithm, the support change effect of the model under the operation of the component includes: Determine at least one associated component by identifying the support component whose positional distance from the target component is less than a preset distance threshold; The predicted change parameters, along with the component parameters of the target component and all associated components, are input into a trained support effect change prediction algorithm model to obtain the support change effect of the model under the component operation. The support effect change prediction algorithm model is trained using a training dataset that includes multiple training component combination data and corresponding specific component change parameter annotations and support effect annotations. The component parameters include at least one of component position, component surface area, and component volume. The support change effect is an increase in support strength, an increase in support fragility, or a risk of support collapse.

[0020] A third aspect of this invention discloses another neural network-based slice model support effect prediction system, the system comprising: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute some or all of the steps in the neural network-based slice model support effect prediction method disclosed in the first aspect of the present invention.

[0021] The fourth aspect of the present invention discloses a computer storage medium storing computer instructions, which, when invoked, are used to execute some or all of the steps in the neural network-based slice model support effect prediction method disclosed in the first aspect of the present invention.

[0022] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: This invention acquires slice data of 3D printed models and user component operations, identifies support components based on a support structure recognition algorithm, generates simulated change data by combining simulation algorithms, and predicts the support change effect under component operations through neural networks. This enables accurate prediction of support effect based on user operations and data simulation, improves the optimization accuracy and printing stability of 3D printed model support structure, and reduces the risk of support failure due to improper operation. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used 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 those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart illustrating a method for predicting the support effect of a slice model based on a neural network, as disclosed in an embodiment of the present invention.

[0025] Figure 2 This is a schematic diagram of the structure of a slice model support effect prediction system based on neural networks disclosed in an embodiment of the present invention.

[0026] Figure 3 This is a schematic diagram of another neural network-based slice model support effect prediction system disclosed in an embodiment of the present invention. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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.

[0028] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0029] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0030] This invention discloses a method and system for predicting the support effect of a sliced ​​model based on a neural network. By acquiring sliced ​​data of a 3D printed model and user component operations, a support structure recognition algorithm is used to identify support components. Simulation algorithms are combined to generate simulated change data, and a neural network is used to predict the support change effect under component operations. This enables accurate prediction of support effects based on user operations and data simulation, improving the optimization accuracy and printing stability of the 3D printed model's support structure, and reducing the risk of support failure due to improper operation. Detailed explanations follow.

[0031] Example 1 Please see Figure 1 , Figure 1This is a flowchart illustrating a method for predicting the support effect of a slice model based on a neural network, as disclosed in an embodiment of the present invention. Figure 1 The described neural network-based slicing model support effect prediction method can be applied to data processing systems / data processing equipment / data processing servers (wherein, the server includes local processing servers or cloud processing servers). For example... Figure 1 As shown, the neural network-based slice model support effect prediction method can include the following operations: 101. Obtain the slice data of the model to be 3D printed and the user's component operations on the model.

[0032] Optionally, the slice data can be an STL slice file, a G-code file, or an AMF slice data; this invention does not impose any limitations.

[0033] Optionally, the component operation may include component removal, component adjustment, component addition, or component parameter modification, and the present invention does not limit this.

[0034] 102. Based on the support structure identification algorithm, identify the support components in the model slice data. Optionally, the support structure recognition algorithm can be a neural network algorithm, a geometric analysis algorithm, or an image processing algorithm; this invention does not limit the algorithm.

[0035] Optionally, the support member can be a single support column, a grid support structure, or a tree-like support structure; the present invention does not limit this.

[0036] 103. Based on the component operation and the simulation algorithm, determine the simulated change data corresponding to the model slice data. Optionally, the simulation algorithm can be a finite element analysis algorithm, a geometric transformation algorithm, or a machine learning prediction algorithm; this invention does not limit the algorithm.

[0037] Optionally, the simulated change data may include changes in component position, shape, or material distribution; this invention does not impose any limitations.

[0038] 104. Based on the simulated change data and supporting components, and using a neural network algorithm, predict the support change effect of the model under component operation.

[0039] Optionally, the supporting change effect can be a qualitative classification result, a quantitative score, or a probability distribution; this invention does not limit this.

[0040] As can be seen, the above-described embodiments of the invention acquire slice data of the 3D printed model and user component operations, identify support components based on a support structure identification algorithm, generate simulated change data by combining simulation algorithms, and predict the support change effect under component operations through neural networks. This enables accurate prediction of support effect based on user operations and data simulation, improves the optimization accuracy and printing stability of the 3D printed model support structure, and reduces the risk of support failure due to improper operation.

[0041] As an optional embodiment, the step above, identifying support components in the model slice data based on the support structure identification algorithm, includes: For each model slice in the model slice data, based on the image segmentation algorithm, the cross-sectional envelope graphics of multiple components belonging to different components in the model slice are segmented out; An image containing the envelope patterns of multiple component cross sections is input into a trained support component prediction neural network to obtain the probability of each component belonging to a support component corresponding to its cross section envelope pattern. Based on the component probabilities corresponding to adjacent model slices, the supporting components in the model slice data are identified.

[0042] Optionally, the image segmentation algorithm can be the U-Net algorithm, the Mask R-CNN algorithm, or the DeepLab algorithm; this invention does not limit the algorithm.

[0043] Optionally, the cross-sectional envelope of the component can be a two-dimensional contour, a three-dimensional projected contour, or a simplified geometric shape; the present invention does not impose any limitations on this.

[0044] Optionally, the predictive neural network of the support component can be a convolutional neural network, a classification neural network, or a probability prediction model; the present invention does not limit this.

[0045] Optionally, the component probability can be a binary probability, a multi-class probability, or a confidence score; this invention does not impose any limitations.

[0046] As can be seen, through the above optional embodiments, by segmenting the component cross-sectional envelope pattern in the model slice based on the image segmentation algorithm, combining the trained support component prediction neural network to obtain the component probability, and identifying the support component based on the component probability of adjacent slices, accurate support component identification based on image segmentation and probability analysis is achieved, thereby improving the accuracy and reliability of 3D printing support structure detection and reducing the risk of printing defects caused by component misjudgment.

[0047] As an optional embodiment, in the above steps, the supporting component prediction neural network is an RNN network architecture, which is trained using a training dataset that includes multiple training component cross-sectional images and corresponding annotations indicating whether they belong to supporting components.

[0048] As can be seen, the model details of the predictive neural network for the support component are defined through the above optional embodiments, so as to accurately predict the shape of the support component in the future, assist in realizing accurate prediction of support effect based on user operation and data simulation, improve the optimization accuracy and printing stability of the 3D printed model support structure, and reduce the risk of support failure due to improper operation.

[0049] As an optional embodiment, the step of identifying supporting components in the model slice data based on the component probabilities corresponding to adjacent model slices includes: The cross-sectional envelope pattern of components with a probability greater than a preset probability threshold is identified as the support component pattern; For any two adjacent model slices, calculate the graphic similarity between the supporting component graphics of the two model slices; Calculate the reciprocal of the positional difference between the support member graphics of the two model slices; Calculate the first probability average of the component probabilities between the support component graphics of the two model slices; The component correlation degree between the two model slices is obtained by summing the graphic similarity, the reciprocal, and the first probability average. Based on the supporting component graphics and component correlation, identify the supporting components in the model slice data.

[0050] Optionally, the probability threshold can be a fixed threshold, a dynamic threshold, or a threshold adjusted based on model accuracy; this invention does not impose any limitations.

[0051] Optionally, the graphic similarity can be Hausdorff distance, cosine similarity, or shape matching degree, and the present invention does not limit it.

[0052] Optionally, the calculation of graphic similarity can be based on contour comparison, feature extraction, or image processing algorithms, and this invention does not limit it.

[0053] Optionally, the location difference can be Euclidean distance, Manhattan distance, or weighted distance, and this invention does not limit it.

[0054] As can be seen, through the above optional embodiments, by screening the support component graphics with a probability exceeding the threshold, the sum of the graphic similarity, the inverse of the positional difference, and the probability average of the support component graphics between adjacent model slices is calculated as the component correlation degree. The support components are identified by combining the graphics and the correlation degree, thereby achieving accurate support component identification based on multi-dimensional graphic features and correlation analysis, improving the accuracy and efficiency of 3D printing support structure analysis, and reducing the risk of structure identification error.

[0055] As an optional embodiment, the step of identifying the support components in the model slice data based on the support component graphics and component correlation includes: For each preset component in the model slice data, determine the number of all supporting component graphics corresponding to that preset component; Calculate the second probability average of the component probabilities of all supporting component graphics corresponding to the preset component; Calculate the average correlation degree of all model slices containing the preset component; The product of the quantity, the second probability average, and the correlation average is used to obtain the possible support parameters corresponding to the preset component; Select preset components whose support parameters may be greater than the preset first parameter threshold, and obtain multiple support components in the model slice data.

[0056] Optionally, the preset component can be an independent geometry, functional area, or printing unit in the model; this invention does not limit this.

[0057] As can be seen, through the above optional embodiments, by calculating the product of the number of supporting component graphics, the average probability, and the average component correlation of the preset component as the support possible parameters, preset components with parameters exceeding the threshold are selected as supporting components, thereby achieving accurate support component selection based on component distribution and correlation, improving the accuracy of 3D printing model support identification and printing stability, and reducing the risk of printing failure due to support misjudgment.

[0058] As an optional embodiment, the component operation in the above steps includes at least one of the following: component position change operation, component physical parameter change operation, component replacement operation, component addition operation, and component deletion operation.

[0059] As can be seen, the above optional embodiments limit the types of component operations to comprehensively characterize the user's operation characteristics, assist in achieving accurate prediction of support effects based on user operations and data simulation, improve the optimization accuracy and printing stability of 3D printed model support structures, and reduce the risk of support failure due to improper operation.

[0060] As an optional embodiment, the step above, determining the simulated change data corresponding to the model slice data based on the component operation and the simulation algorithm, includes: Identify the target component in the model slice data corresponding to the component operation; The component operations are input into the trained simulation prediction algorithm model to obtain the predicted change parameters corresponding to the component operations; optionally, the simulation prediction algorithm model is trained using a training dataset that includes multiple training component operations and corresponding component parameter change annotations. The predicted change parameters are applied to the target component to obtain the simulated change data corresponding to the model slice data.

[0061] Optionally, the target component can be a single component, multiple components, or a collection of regional components; the present invention does not impose any limitations.

[0062] Optionally, the process of determining the target component can be based on operation parsing, component matching, or user intent analysis, and this invention does not limit it.

[0063] Optionally, the simulation prediction algorithm model can be a regression model, a classification model, or a generative model; this invention does not impose any limitations.

[0064] Optionally, the predicted change parameters may include component size changes, position offsets, or shape adjustment parameters, which are not limited in this invention.

[0065] Optionally, the training dataset may include historical operation data, simulated operation data, or experimental data, and this invention does not impose any limitations.

[0066] Optionally, this process can be implemented based on geometric transformation, parameter mapping, or simulation, and the present invention does not limit it.

[0067] As can be seen, through the above optional embodiments, by determining the target component corresponding to the component operation, combining the trained simulation prediction algorithm model to generate predicted change parameters and applying them to the target component to generate simulated change data, accurate slice data change simulation based on user operation and simulation prediction can be achieved, thereby improving the accuracy and reliability of 3D printing model operation effect evaluation and reducing the risk of printing defects caused by improper operation.

[0068] As an optional embodiment, the step above, predicting the support change effect of the model under component operation based on the simulated change data and the supporting components using a neural network algorithm, includes: Identify support members whose distance from the target component is less than a preset distance threshold, thus obtaining at least one associated component; The predicted change parameters, as well as the component parameters of the target component and all related components, are input into the trained support effect change prediction algorithm model to obtain the support change effect of the model under component operation.

[0069] Optionally, the support effect change prediction algorithm model is trained using a training dataset that includes multiple training component combination data and corresponding specific component change parameter annotations and support effect annotations.

[0070] Optionally, the component parameters include at least one of the component location, component surface area, and component volume.

[0071] Optionally, the effect of the support change is to improve the support's strength, increase its fragility, or create a risk of support collapse.

[0072] Optionally, the distance threshold can be a fixed distance, a dynamic distance, or a distance adjusted based on the model size; this invention does not impose any limitations.

[0073] Optionally, the associated component can be a neighboring component, a functionally related component, or a collaborative support component; the present invention does not limit this.

[0074] Optionally, the algorithm model for predicting changes in support effect can be a neural network model, a regression model, or a classification model; this invention does not impose any limitations.

[0075] Optionally, the component parameters may include additional parameters, such as component height, material density, or geometric complexity, which are not limited in this invention.

[0076] Optionally, the training dataset may include historical component data, simulated change data, or experimental data, and this invention does not impose any limitations.

[0077] As can be seen, through the above optional embodiments, by filtering associated components whose distance from the target component is less than a threshold, the predicted change parameters and the component parameters of the target and associated components are input into the trained support effect change prediction algorithm model, predicting that the support change effect is an increase in robustness, an increase in fragility, or a risk of collapse. This achieves accurate support effect evaluation based on component association and parameter prediction, improves the accuracy of 3D printing support optimization decisions and printing quality, and reduces the risk of support failure.

[0078] Example 2 Please see Figure 2 , Figure 2 This is a schematic diagram of the structure of a slice model support effect prediction system based on a neural network, as disclosed in an embodiment of the present invention. Figure 2 The described neural network-based slice model support effect prediction system can be applied to data processing systems / data processing equipment / data processing servers (wherein, the server includes local processing servers or cloud processing servers). For example... Figure 2 As shown, the neural network-based slice model support effect prediction system may include: The acquisition module 201 is used to acquire the slice data of the model to be 3D printed and the user's component operations on the model.

[0079] The identification module 202 is used to identify the supporting components in the model slice data based on the supporting structure identification algorithm. The determination module 203 is used to determine the simulated change data corresponding to the model slice data based on the component operation and the simulation algorithm. The prediction module 204 is used to predict the effect of the model on the support changes under component operation based on the simulated change data and the support components, using a neural network algorithm.

[0080] As can be seen, the above-described embodiments of the invention acquire slice data of the 3D printed model and user component operations, identify support components based on a support structure identification algorithm, generate simulated change data by combining simulation algorithms, and predict the support change effect under component operations through neural networks. This enables accurate prediction of support effect based on user operations and data simulation, improves the optimization accuracy and printing stability of the 3D printed model support structure, and reduces the risk of support failure due to improper operation.

[0081] As an optional embodiment, the identification module identifies the specific method by which it identifies the support components in the model slice data based on the support structure identification algorithm, including: For each model slice in the model slice data, based on the image segmentation algorithm, the cross-sectional envelope graphics of multiple components belonging to different components in the model slice are segmented out; An image containing the envelope patterns of multiple component cross sections is input into a trained support component prediction neural network to obtain the probability of each component belonging to a support component corresponding to its cross section envelope pattern. Based on the component probabilities corresponding to adjacent model slices, the supporting components in the model slice data are identified.

[0082] As can be seen, through the above optional embodiments, by segmenting the component cross-sectional envelope pattern in the model slice based on the image segmentation algorithm, combining the trained support component prediction neural network to obtain the component probability, and identifying the support component based on the component probability of adjacent slices, accurate support component identification based on image segmentation and probability analysis is achieved, thereby improving the accuracy and reliability of 3D printing support structure detection and reducing the risk of printing defects caused by component misjudgment.

[0083] As an optional embodiment, the support component prediction neural network is an RNN network architecture, which is trained on a training dataset that includes multiple cross-sectional images of training components and corresponding annotations indicating whether they belong to support components.

[0084] As can be seen, the model details of the predictive neural network for the support component are defined through the above optional embodiments, so as to accurately predict the shape of the support component in the future, assist in realizing accurate prediction of support effect based on user operation and data simulation, improve the optimization accuracy and printing stability of the 3D printed model support structure, and reduce the risk of support failure due to improper operation.

[0085] As an optional embodiment, the identification module identifies the specific method of supporting components in the model slice data based on the component probabilities corresponding to adjacent model slices, including: The cross-sectional envelope pattern of components with a probability greater than a preset probability threshold is identified as the support component pattern; For any two adjacent model slices, calculate the graphic similarity between the supporting component graphics of the two model slices; Calculate the reciprocal of the positional difference between the support member graphics of the two model slices; Calculate the first probability average of the component probabilities between the support component graphics of the two model slices; The component correlation degree between the two model slices is obtained by summing the graphic similarity, the reciprocal, and the first probability average. Based on the supporting component graphics and component correlation, identify the supporting components in the model slice data.

[0086] As can be seen, through the above optional embodiments, by screening the support component graphics with a probability exceeding the threshold, the sum of the graphic similarity, the inverse of the positional difference, and the probability average of the support component graphics between adjacent model slices is calculated as the component correlation degree. The support components are identified by combining the graphics and the correlation degree, thereby achieving accurate support component identification based on multi-dimensional graphic features and correlation analysis, improving the accuracy and efficiency of 3D printing support structure analysis, and reducing the risk of structure identification error.

[0087] As an optional embodiment, the identification module identifies the specific method of supporting components in the model slice data based on the supporting component graphics and component correlation, including: For each preset component in the model slice data, determine the number of all supporting component graphics corresponding to that preset component; Calculate the second probability average of the component probabilities of all supporting component graphics corresponding to the preset component; Calculate the average correlation degree of all model slices containing the preset component; The product of the quantity, the second probability average, and the correlation average is used to obtain the possible support parameters corresponding to the preset component; Select preset components whose support parameters may be greater than the preset first parameter threshold, and obtain multiple support components in the model slice data.

[0088] As can be seen, through the above optional embodiments, by calculating the product of the number of supporting component graphics, the average probability, and the average component correlation of the preset component as the support possible parameters, preset components with parameters exceeding the threshold are selected as supporting components, thereby achieving accurate support component selection based on component distribution and correlation, improving the accuracy of 3D printing model support identification and printing stability, and reducing the risk of printing failure due to support misjudgment.

[0089] As an optional embodiment, the component operation includes at least one of the following: component position change operation, component physical parameter change operation, component replacement operation, component addition operation, and component deletion operation.

[0090] As can be seen, the above optional embodiments limit the types of component operations to comprehensively characterize the user's operation characteristics, assist in achieving accurate prediction of support effects based on user operations and data simulation, improve the optimization accuracy and printing stability of 3D printed model support structures, and reduce the risk of support failure due to improper operation.

[0091] As an optional embodiment, the determining module determines the specific method by which it determines the simulated change data corresponding to the model slice data based on the component operation and the simulation algorithm, including: Identify the target component in the model slice data corresponding to the component operation; The component operations are input into the trained simulation prediction algorithm model to obtain the predicted change parameters corresponding to the component operations; optionally, the simulation prediction algorithm model is trained using a training dataset that includes multiple training component operations and corresponding component parameter change annotations. The predicted change parameters are applied to the target component to obtain the simulated change data corresponding to the model slice data.

[0092] As can be seen, through the above optional embodiments, by determining the target component corresponding to the component operation, combining the trained simulation prediction algorithm model to generate predicted change parameters and applying them to the target component to generate simulated change data, accurate slice data change simulation based on user operation and simulation prediction can be achieved, thereby improving the accuracy and reliability of 3D printing model operation effect evaluation and reducing the risk of printing defects caused by improper operation.

[0093] As an optional embodiment, the prediction module, based on simulated change data and supporting components, uses a neural network algorithm to predict the specific method by which the model predicts the support change effect under component operation, including: Identify support members whose distance from the target component is less than a preset distance threshold, thus obtaining at least one associated component; The predicted change parameters, as well as the component parameters of the target component and all associated components, are input into the trained support effect change prediction algorithm model to obtain the support change effect of the model under component operation; optionally, the support effect change prediction algorithm model is trained through a training dataset that includes multiple training component combination data and corresponding specific component change parameter annotations and support effect annotations; the component parameters include at least one of component position, component surface area and component volume; the support change effect is an increase in support strength, an increase in support fragility or a risk of support collapse.

[0094] As can be seen, through the above optional embodiments, by filtering associated components whose distance from the target component is less than a threshold, the predicted change parameters and the component parameters of the target and associated components are input into the trained support effect change prediction algorithm model, predicting that the support change effect is an increase in robustness, an increase in fragility, or a risk of collapse. This achieves accurate support effect evaluation based on component association and parameter prediction, improves the accuracy of 3D printing support optimization decisions and printing quality, and reduces the risk of support failure.

[0095] Example 3 Please see Figure 3 , Figure 3 This is another neural network-based slice model support effect prediction system disclosed in the embodiments of the present invention. Figure 3 The described neural network-based slice model support effect prediction system is applied in data processing systems / data processing equipment / data processing servers (wherein, the server includes local processing servers or cloud processing servers). For example... Figure 3 As shown, the neural network-based slice model support effect prediction system may include: Memory 301 storing executable program code; Processor 302 coupled to memory 301; The processor 302 calls the executable program code stored in the memory 301 to execute the steps of the neural network-based slice model support effect prediction method described in Embodiment 1.

[0096] Example 4 This invention discloses a computer read storage medium that stores a computer program for electronic data interchange, wherein the computer program causes a computer to execute the steps of the neural network-based slice model support effect prediction method described in Embodiment 1.

[0097] Example 5 This invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps of the neural network-based slice model support effect prediction method described in Embodiment 1.

[0098] The foregoing has described specific embodiments of this specification; other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims may be performed in a different order than those shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily have to follow the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0099] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0100] For ease of description, the above devices are described in terms of function, divided into various units. Of course, in implementing this specification, the functions of each unit can be implemented in one or more software and / or hardware components.

[0101] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the embodiments of this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0102] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0103] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0104] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0105] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0106] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0107] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0108] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0109] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0110] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0111] Finally, it should be noted that the method and system for predicting the support effect of a slice model based on a neural network disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting the performance of a slice model based on a neural network, characterized in that, The method includes: Acquire slice data of the model to be 3D printed and user operations on the components of the model; Based on the support structure identification algorithm, the support components in the model slice data are identified; Based on the component operation and the simulation algorithm, the simulated change data corresponding to the model slice data is determined; Based on the simulated change data and the supporting components, a neural network algorithm is used to predict the support change effect of the model under the operation of the components.

2. The method for predicting the support effect of a slice model based on a neural network according to claim 1, characterized in that, The support structure identification algorithm identifies the support components in the model slice data, including: For each model slice in the model slice data, based on the image segmentation algorithm, multiple component cross-sectional envelope patterns belonging to different components in the model slice are segmented out; An image containing multiple cross-sectional envelope patterns of the components is input into a trained support component prediction neural network to obtain the probability of each cross-sectional envelope pattern of the components belonging to the support component. Based on the component probabilities corresponding to adjacent model slices, the supporting components in the model slice data are identified.

3. The method for predicting the support effect of a slice model based on a neural network according to claim 2, characterized in that, The supporting component prediction neural network is an RNN network architecture, which is trained using a training dataset that includes multiple cross-sectional images of training components and corresponding annotations indicating whether they belong to supporting components.

4. The method for predicting the support effect of a slice model based on a neural network according to claim 2, characterized in that, The step of identifying supporting components in the model slice data based on the component probabilities corresponding to adjacent model slices includes: The cross-sectional envelope pattern of the component whose probability is greater than a preset probability threshold is determined as the support component pattern; For any two adjacent model slices, calculate the graphic similarity between the supporting component graphics of the two model slices; Calculate the reciprocal of the positional difference between the support member graphics of the two model slices; Calculate the first probability average of the component probabilities between the support component graphics of the two model slices; The component correlation degree between the two model slices is obtained by summing the graphic similarity, the reciprocal, and the first probability average. Based on the supporting component graphic and the component correlation, the supporting components in the model slice data are identified.

5. The method for predicting the support effect of a slice model based on a neural network according to claim 4, characterized in that, The step of identifying the support components in the model slice data based on the support component graphic and the component correlation degree includes: For each preset component in the model slice data, determine the number of all the supporting component graphics corresponding to that preset component; Calculate the second probability average of the component probabilities of all the supporting component graphics corresponding to the preset component; Calculate the average correlation degree of the component correlation degree corresponding to all model slices containing the preset component; Calculate the product of the quantity, the second probability average, and the correlation average to obtain the possible support parameters corresponding to the preset component; The preset components whose possible support parameters are greater than a preset first parameter threshold are selected to obtain multiple support components in the model slice data.

6. The method for predicting the support effect of a slice model based on a neural network according to claim 1, characterized in that, The component operations include at least one of the following: component position change operation, component physical parameter change operation, component replacement operation, component addition operation, and component deletion operation.

7. The method for predicting the support effect of a slice model based on a neural network according to claim 1, characterized in that, The step of determining the simulated change data corresponding to the model slice data based on the component operation and a simulation algorithm includes: Determine the target active component in the model slice data corresponding to the component operation; The component operation is input into the trained simulation prediction algorithm model to obtain the predicted change parameters corresponding to the component operation; the simulation prediction algorithm model is trained using a training dataset that includes multiple training component operations and corresponding component parameter change annotations. The predicted change parameters are applied to the target component to obtain the simulated change data corresponding to the model slice data.

8. The method for predicting the support effect of a slice model based on a neural network according to claim 7, characterized in that, The step of predicting the support change effect of the model under the operation of the component based on the simulated change data and the support component, using a neural network algorithm, includes: Determine at least one associated component by identifying the support component whose positional distance from the target component is less than a preset distance threshold; The predicted change parameters, along with the component parameters of the target component and all associated components, are input into a trained support effect change prediction algorithm model to obtain the support change effect of the model under the component operation. The support effect change prediction algorithm model is trained using a training dataset that includes multiple training component combination data and corresponding specific component change parameter annotations and support effect annotations. The component parameters include at least one of component position, component surface area, and component volume. The support change effect is an increase in support strength, an increase in support fragility, or a risk of support collapse.

9. A slice model support effect prediction system based on neural networks, characterized in that, The system includes: The acquisition module is used to acquire slice data of the model to be 3D printed and user operations on the components of the model; The identification module is used to identify the supporting components in the model slice data based on the supporting structure identification algorithm; The determination module is used to determine the simulated change data corresponding to the model slice data based on the component operation and a simulation algorithm. The prediction module is used to predict the support change effect of the model under the operation of the component based on the simulated change data and the support component, using a neural network algorithm.

10. A slice model support effect prediction system based on neural networks, characterized in that, The system includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the neural network-based slice model support effect prediction method as described in any one of claims 1-8.

Images