Method and device for simplifying digital model of automatic production line

Abstract

The embodiment of the invention provides an automatic production line digital model simplification method and device, and the method comprises the steps: recognizing and storing key feature information through a significance feature extraction module, and building a four-level topological structure tree; a self-adaptive feature weight calculation mechanism is innovatively designed, and intelligent simplification of the grid is realized in combination with multi-resolution wavelet analysis. According to the system, topological structure features and local geometric features are extracted and matched under different scales through a multi-scale feature fusion module, and accurate maintenance of key features in the model simplification process is achieved. According to the method, the limitation of traditional model simplification is broken through, and an intelligent and automatic solution is provided for efficient processing of a large-scale digital model.

Description

Technical Field

[0001] This application relates to the field of data processing, and particularly to a method and device for simplifying digital models of automated production lines. Background Art

[0002] Traditional digital model simplification methods mainly use unified mesh compression algorithms, which are difficult to retain the key features of the model and have low computational efficiency when dealing with complex automated production line models. Existing technologies lack an adaptive mechanism in feature extraction and weight assignment and cannot perform differential simplification according to the importance of different regions.

[0003] At the same time, existing systems have obvious deficiencies in multi-scale feature processing and topology preservation. Traditional methods often ignore the hierarchical relationship of the model, resulting in distortion of the simplified model or loss of key structures. The system is also relatively simple in feature fusion and node optimization and fails to achieve the optimal balance between model simplification and feature preservation.

[0004] In addition, existing technologies also have limitations in mesh reconstruction and quality control. They lack an adaptive simplification strategy based on significant features and do not fully utilize the advantages of multi-resolution analysis. Solving these problems is of great significance for improving the processing efficiency and visual quality of large-scale digital models. Summary of the Invention

[0005] In view of the problems in the prior art, this application provides a method and device for simplifying digital models of automated production lines, which can break through the limitations of traditional model simplification and provide an intelligent and automated solution for the efficient processing of large-scale digital models.

[0006] To solve at least one of the above problems, this application provides the following technical solutions:

[0007] In a first aspect, this application provides a method for simplifying a digital model of an automated production line, including:

[0008] Construct a significant feature extraction module. The significant feature extraction module calculates the significance scores for the point, line, surface, and volume data of the digital model of the automated production line, extracts the feature points with significance scores exceeding a first threshold as a set of significant feature points, and at the same time stores the corresponding morphological parameters and position parameters of the set of significant feature points in a significant feature database, and establishes a four-level topology structure tree based on the set of significant feature points;

[0009] Input the four - level topological structure tree into the adaptive feature weight calculation module. The adaptive feature weight calculation module calculates the weight coefficient of each level based on the spatial distribution density of the significant feature point set, multiplies the weight coefficient by the feature data of the corresponding level to obtain weighted feature data, inputs the weighted feature data into the adaptive octree grid construction module. The adaptive octree grid construction module constructs grid nodes based on the weighted feature data, embeds multi - resolution wavelet basis functions into the grid nodes, performs hierarchical thresholding on the wavelet function coefficients, and reconstructs the thresholded wavelet coefficients into an initial simplified grid;

[0010] Input the initial simplified grid into the multi - scale feature fusion module. The multi - scale feature fusion module extracts the topological structure features of the initial simplified grid at a coarse scale, extracts the local geometric features of the initial simplified grid at a fine scale, constructs a feature correspondence graph based on the topological structure features and the local geometric features, matches the feature correspondence graph with the morphological parameters and position parameters in the significant feature database, and adjusts the nodes of the initial simplified grid according to the matching result to generate a final simplified model.

[0011] Further, construct the significant feature extraction module. The significant feature extraction module calculates the significant score for the point, line, surface, and volume data in the digital model of the automated production line, and extracts the feature points with the significant score exceeding the first threshold as the significant feature point set, including:

[0012] Perform discretization processing on the digital model of the automated production line to obtain point cloud data, establish a three - dimensional grid structure based on the point cloud data, calculate the curvature value, normal vector, and neighborhood density of each vertex in the three - dimensional grid structure, input the curvature value, normal vector, and neighborhood density into the significant calculation formula to obtain the significant coefficient, and construct the significant feature extraction module based on the significant coefficient;

[0013] Compare the significant coefficient in the significant feature extraction module with the first threshold, extract the vertices with the significant coefficient greater than the first threshold as feature points, construct a significant feature point set based on the spatial position relationship of the feature points, and store the geometric attribute parameters and topological attribute parameters of the significant feature point set as morphological parameters and position parameters respectively.

[0014] Further, store the morphological parameters and position parameters corresponding to the significant feature point set in the significant feature database, and establish a four - level topological structure tree based on the significant feature point set, including:

[0015] Input the morphological parameters and position parameters of the set of significant feature points into a feature encoder. The feature encoder performs geometric feature encoding on the morphological parameters to obtain a geometric feature vector, performs spatial position encoding on the position parameters to obtain a position feature vector, combines the geometric feature vector and the position feature vector to form feature data, and writes the feature data into a significant feature database according to a hierarchical structure;

[0016] Construct a four-level topological structure tree based on the set of significant feature points. Use the point data in the set of significant feature points as the first-level nodes, the connection relationships between adjacent points as the second-level nodes, the patches enclosed by closed line segments as the third-level nodes, and the volume elements composed of closed patches as the fourth-level nodes. Store the corresponding feature data indices in each node of the four-level topological structure tree.

[0017] Further, input the four-level topological structure tree into an adaptive feature weight calculation module. The adaptive feature weight calculation module calculates the weight coefficient for each level based on the spatial distribution density of the set of significant feature points, and multiplies the weight coefficient by the feature data of the corresponding level to obtain weighted feature data, including:

[0018] Construct a spatial density distribution map based on the four-level topological structure tree. Calculate the number of significant feature points within the neighborhood range of each node in the spatial density distribution map, divide the number by the neighborhood volume to obtain a local density value, perform normalization processing on the local density value to obtain a density distribution coefficient, and input the density distribution coefficient into a weight mapping function to generate the weight coefficient corresponding to each level;

[0019] Perform matrix multiplication on the weight coefficient and the feature data of each level node in the four-level topological structure tree to generate a weighted feature matrix, perform normalization processing on the weighted feature matrix to obtain weighted feature data, and reorganize the weighted feature data into a tree structure according to the hierarchical relationship.

[0020] Further, input the weighted feature data into an adaptive octree grid construction module. The adaptive octree grid construction module constructs grid nodes based on the weighted feature data, embeds multi-resolution wavelet basis functions into the grid nodes, performs hierarchical thresholding on the wavelet function coefficients, and reconstructs the thresholded wavelet coefficients into an initial simplified grid, including:

[0021] Determine a spatial bounding box based on the weighted feature data, recursively divide the spatial bounding box into an octree structure, calculate the local variance of the weighted feature data at each node of the octree structure, and continue to subdivide the node when the local variance of the node is greater than the second threshold until the preset subdivision level is reached or the local variance is less than the second threshold. Build an adaptive grid on the octree nodes after subdivision is completed, and map the Haar wavelet basis function to the node coordinates of the adaptive grid;

[0022] Perform coefficient decomposition on the Haar wavelet basis function on the nodes of the adaptive grid, calculate the energy value of the wavelet coefficients at each layer, set an adaptive threshold according to the energy value, set the wavelet coefficients less than the adaptive threshold to zero, perform a reconstruction operation on the remaining wavelet coefficients to generate an initial simplified grid, and establish a correspondence between the nodes of the initial simplified grid and the original weighted feature data.

[0023] Further, input the initial simplified grid into a multi-scale feature fusion module. The multi-scale feature fusion module extracts the topological structure features of the initial simplified grid at a coarse scale and extracts the local geometric features of the initial simplified grid at a fine scale, including:

[0024] Perform multi-level decomposition on the initial simplified grid to obtain grid representations at different scales. Build a grid connectivity graph at the coarse scale level, calculate the degree distribution and connectivity properties of each node in the grid connectivity graph, construct a grid skeleton based on the degree distribution and connectivity properties, and extract the branch structure and connection relationship of the grid skeleton as topological structure features;

[0025] Calculate the local curvature tensor of the nodes of the initial simplified grid at the fine scale level, calculate the principal curvature value and principal curvature direction based on the curvature tensor, combine the principal curvature value and principal curvature direction to form a curvature descriptor, perform clustering analysis on the curvature descriptor to obtain local geometric features, and establish a mapping relationship between the local geometric features and the grid nodes.

[0026] Further, construct a feature correspondence graph based on the topological structure features and the local geometric features, match the feature correspondence graph with the morphological parameters and position parameters in the significant feature database, and adjust the nodes of the initial simplified grid according to the matching result to generate a final simplified model, including:

[0027] Input the topological structure features and the local geometric features into a feature fusion network. The feature fusion network generates a feature correspondence graph based on graph convolution operations, calculates a similarity matrix between the nodes in the feature correspondence graph, matches the similarity matrix with the shape parameters and position parameters stored in the significant feature database, generates a feature matching metric value, and constructs a node optimization objective function based on the feature matching metric value;

[0028] Iteratively solve the node optimization objective function to adjust the node positions and connection relationships of the initial simplified mesh. Stop the iteration when the objective function value is less than the third threshold or the maximum number of iterations is reached. Reconstruct the final mesh node coordinates and connection relationships into a 3D model, and perform mesh smoothing on the reconstructed 3D model to obtain the final simplified model.

[0029] In a second aspect, the present application provides an automated production line digital model simplification device, including:

[0030] A feature extraction module, configured to construct a significance feature extraction module. The significance feature extraction module calculates significance scores for the point, line, surface, and volume data of the automated production line digital model, extracts the feature points with significance scores exceeding the first threshold as a significant feature point set, and at the same time stores the shape parameters and position parameters corresponding to the significant feature point set in the significant feature database, and establishes a four-level topological structure tree based on the significant feature point set;

[0031] An architecture building module, configured to input the four-level topological structure tree into an adaptive feature weight calculation module. The adaptive feature weight calculation module calculates the weight coefficient of each level based on the spatial distribution density of the significant feature point set, multiplies the weight coefficient by the feature data of the corresponding level to obtain weighted feature data, inputs the weighted feature data into an adaptive octree mesh construction module. The adaptive octree mesh construction module constructs mesh nodes based on the weighted feature data, embeds multi-resolution wavelet basis functions into the mesh nodes, performs hierarchical thresholding on the wavelet function coefficients, and reconstructs the thresholded wavelet coefficients into an initial simplified mesh;

[0032] A model simplification module, configured to input the initial simplified mesh into a multi-scale feature fusion module. The multi-scale feature fusion module extracts the topological structure features of the initial simplified mesh at a coarse scale and extracts the local geometric features of the initial simplified mesh at a fine scale, constructs a feature correspondence graph based on the topological structure features and the local geometric features, matches the feature correspondence graph with the shape parameters and position parameters in the significant feature database, and adjusts the nodes of the initial simplified mesh according to the matching result to generate a final simplified model.

[0033] In a third aspect, the present application provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the steps of the automated production line digital model simplification method are implemented.

[0034] In a fourth aspect, the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the automated production line digital model simplification method are implemented.

[0035] In a fifth aspect, the present application provides a computer program product, including a computer program / instructions. When the computer program / instructions are executed by a processor, the steps of the automated production line digital model simplification method are implemented.

[0036] As can be seen from the above technical solutions, the present application provides an automated production line digital model simplification method and device. The key feature information is identified and saved through the significant feature extraction module, and a four-level topology structure tree is established. An adaptive feature weight calculation mechanism is innovatively designed, and intelligent simplification of the grid is achieved by combining multi-resolution wavelet analysis. The system extracts and matches the topology structure features and local geometric features at different scales through the multi-scale feature fusion module, and accurately preserves the key features during the model simplification process. This method breaks through the limitations of traditional model simplification and provides an intelligent and automated solution for the efficient processing of large-scale digital models. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0038] Figure 1 It is one of the flow diagrams of the automated production line digital model simplification method in the embodiments of the present application;

[0039] Figure 2 It is another flow diagram of the automated production line digital model simplification method in the embodiments of the present application;

[0040] Figure 3 It is yet another flow diagram of the automated production line digital model simplification method in the embodiments of the present application;

[0041] Figure 4 It is still another flow diagram of the automated production line digital model simplification method in the embodiments of the present application;

[0042] Figure 5This is the fifth schematic flowchart of the automated production line digital model simplification method in the embodiments of the present application;

[0043] Figure 6 This is the sixth schematic flowchart of the automated production line digital model simplification method in the embodiments of the present application;

[0044] Figure 7 This is the seventh schematic flowchart of the automated production line digital model simplification method in the embodiments of the present application;

[0045] Figure 8 This is the structural diagram of the automated production line digital model simplification device in the embodiments of the present application;

[0046] Figure 9 This is the structural schematic diagram of the electronic device in the embodiments of the present application.

[0047] Reference numerals:

[0048] Electronic device 9600, central processing unit 9100, memory 9140, communication module 9110, input unit 9120, audio processor 9130, display 9160, power supply 9170, buffer memory 9141, application / function storage unit 9142, data storage unit 9143, driver program storage unit 9144, antenna 9111, speaker 9131, microphone 9132. Detailed implementation manners

[0049] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Apparently, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without making creative efforts shall fall within the protection scope of the present application.

[0050] In the technical solutions of the present application, the acquisition, storage, use, processing, etc. of data all comply with the relevant provisions of national laws and regulations.

[0051] Considering the problems existing in the prior art, the present application provides an automated production line digital model simplification method and device. The key feature information is identified and saved by the saliency feature extraction module, and a four-level topology structure tree is established. An adaptive feature weight calculation mechanism is innovatively designed, and intelligent simplification of the grid is achieved by combining multi-resolution wavelet analysis. The system extracts and matches the topology structure features and local geometric features at different scales through the multi-scale feature fusion module, and realizes the accurate preservation of key features during the model simplification process. This method breaks through the limitations of traditional model simplification and provides an intelligent and automated solution for the efficient processing of large-scale digital models.

[0052] In order to break through the limitations of traditional model simplification and provide an intelligent and automated solution for the efficient processing of large-scale digital models, this application provides an embodiment of an automated production line digital model simplification method. Refer to Figure 1 , the automated production line digital model simplification method specifically includes the following content:

[0053] Step S101: Construct a saliency feature extraction module. The saliency feature extraction module calculates saliency scores for point, line, surface, and volume data in the automated production line digital model, extracts the feature points with saliency scores exceeding the first threshold as the set of salient feature points, and at the same time stores the corresponding morphological parameters and position parameters of the set of salient feature points in the salient feature database. A four-level topological structure tree is established based on the set of salient feature points;

[0054] Optionally, when constructing the saliency feature extraction module in this embodiment, a multi-level feature extraction framework is developed for the complex spatial structure features of the automated production line. First, an improved octree space partitioning algorithm is used to divide the entire production line digital model into spatial sub-regions of unequal sizes. During the space partitioning process, an adaptive subdivision strategy based on local feature density is introduced. For the regions where important components such as key connection points, valves, and sensors in the production line are located, a more detailed partitioning is adopted, and the partitioning accuracy can reach the millimeter level; while for regions with relatively single features such as conveyor belts and pipelines, a coarser partitioning is adopted. This differential space partitioning strategy effectively balances the computational efficiency and feature extraction accuracy.

[0055] In the point cloud sampling link of this embodiment, an importance-based adaptive sampling algorithm is implemented. For each spatial sub-region, the local curvature change rate and shape complexity are first calculated, and these metrics are input into the importance evaluation function. The importance evaluation function adopts a weighted combination form, where the weight of the curvature change rate is 0.4, the weight of the shape complexity is 0.3, and the weight of the spatial position importance is 0.3. The sampling density is dynamically adjusted based on the evaluation results. A denser sampling point distribution is adopted in regions with high importance, and the sampling spacing can reach 0.1 mm; while in regions with low importance, the sampling spacing can be relaxed to 10 mm, which not only ensures the integrity of key features but also significantly reduces data redundancy.

[0056] In terms of feature point saliency calculation, this embodiment develops a multi-dimensional feature fusion calculation framework. First, local geometric features are extracted, including principal curvatures k1, k2, Gaussian curvature K, mean curvature H, as well as shape index S and curvature change rate C. At the same time, the local shape distribution features of each point are calculated. By analyzing the eigenvalues λ1, λ2, λ3 of the point cloud covariance matrix, a shape descriptor D = (λ1 - λ2) / (λ1 - λ3) is constructed to characterize the shape characteristics of the region where the point is located. For line structure features, the continuity index of the tangent vector and the cumulative value of curvature change are calculated. For surface structure features, the change gradient of the normal vector and the boundary curvature distribution of the patch are analyzed. For volume structure features, the local volume density and shape complexity index are calculated.

[0057] In the feature fusion process of this embodiment, an adaptive weight assignment mechanism is implemented. By analyzing the discriminability and stability of different types of features, a feature importance evaluation model is established. This model considers factors such as the scale invariance, noise robustness, and discriminative ability of the features. For the pipeline system, the curvature feature weight is set to 0.4, the shape feature weight is 0.3, and the position feature weight is 0.3; for the box structure, the volume feature weight is 0.4, the boundary feature weight is 0.3, and the topological feature weight is 0.3. The comprehensive saliency score S of each point is obtained through weighted fusion:

[0058] S = Σ(wi * fi), where wi is the weight of the i-th feature and fi is the normalized feature value.

[0059] In the feature point extraction step, this embodiment develops an adaptive clustering algorithm based on region growing. First, the point with the highest saliency score in each spatial sub-region is selected as the seed point, and then it is gradually expanded based on the spatial proximity relationship and feature similarity. A dynamic threshold mechanism is introduced during the expansion process. The initial threshold is set to the mean of the saliency scores and is gradually adjusted as the region expands. When the feature variance of the points in the region exceeds 1.5 times the threshold, the growth stops. This adaptive stopping criterion effectively avoids the problem of over-clustering.

[0060] In the process of morphological parameter extraction, this embodiment adopts a multi-scale feature description strategy. At the local scale (1 - 5 mm), features such as the density distribution of the point set and the normal vector consistency are calculated; at the medium scale (5 - 50 mm), features such as boundary curvature and surface roughness are extracted; at the global scale (> 50 mm), the overall shape features and topological characteristics are analyzed. Through this multi-scale feature description method, a comprehensive characterization of the geometric properties of the feature point set is achieved.

[0061] In the position parameter calculation step, this embodiment implements a position description mechanism for a multi-reference system. First, a global coordinate system for the production line is established, with the origin set at the entrance of the production line, the z-axis perpendicular to the upward direction, and the x-axis along the main transmission direction of the production line. Then, a local coordinate system is established for each main device, and the transformation relationship with the global coordinate system is recorded through a coordinate transformation matrix. The position of the feature point is represented by both the global coordinates (X, Y, Z) and the local coordinates (x, y, z) relative to the nearest device. This representation method facilitates feature matching and model simplification.

[0062] When constructing the four-level topology structure tree in this embodiment, a bottom-up hierarchical construction algorithm is developed. The first level stores the three-dimensional coordinates and significance scores of the original feature points; the second level constructs edge features by analyzing the distance relationship and feature similarity between points, and records attributes such as the length, direction, and curvature of the edges; the third level constructs surface features based on the boundary closure analysis, and stores the normal vector, area, and boundary features of the surface patches; the fourth level constructs volume features through the closure analysis of the surface patches, and records information such as volume, centroid, and main direction.

[0063] In terms of data storage, this embodiment adopts a hierarchical database structure. A feature index table is established as the main table, which contains feature ID, type, and level information. The morphological parameter table stores geometric feature data, and the position parameter table records spatial position information. Efficient data query and update are achieved through foreign key association. At the same time, a data compression mechanism based on compressive sensing is implemented to merge and store highly repetitive feature data, and the data compression rate can reach more than 60%.

[0064] Through the above implementation scheme, this embodiment establishes an accurate significant feature extraction system. This scheme is particularly suitable for the digital model simplification of complex automated production lines, and can accurately identify and retain key features in the model. Through multi-dimensional feature analysis and hierarchical data organization, it provides a reliable feature basis for subsequent model simplification. Practical applications show that this scheme can effectively extract key structural features in the production line, providing strong support for applications such as equipment monitoring and fault diagnosis.

[0065] Step S102: Input the four-level topology structure tree into the adaptive feature weight calculation module. The adaptive feature weight calculation module calculates the weight coefficient of each level based on the spatial distribution density of the significant feature point set, multiplies the weight coefficient by the feature data of the corresponding level to obtain weighted feature data, inputs the weighted feature data into the adaptive octree grid construction module. The adaptive octree grid construction module constructs grid nodes based on the weighted feature data, embeds multi-resolution wavelet basis functions into the grid nodes, performs hierarchical thresholding processing on the wavelet function coefficients, and reconstructs the thresholded wavelet coefficients into an initial simplified grid;

[0066] Optionally, when calculating the adaptive feature weights in this embodiment, first perform a depth-first traversal of the four-level topological structure tree, and establish local feature statistical windows at each level node. For different functional areas in the automated production line, such as the conveyor belt system, the robotic arm workstation, the inspection station, etc., set different-sized feature statistical windows respectively to ensure the locality and accuracy of feature analysis.

[0067] This embodiment calculates the local density value by analyzing the distribution law of feature points in three-dimensional space. For each feature statistical window, the kernel density estimation method is adopted, and a Gaussian kernel function with an adaptive bandwidth is selected for density calculation. At key nodes of the production line, such as equipment connection points, steering devices, etc., there are often higher feature point densities, and the retention of features in these areas is crucial for model simplification.

[0068] In the process of calculating the weight coefficient in this embodiment, an adaptive weight allocation mechanism based on density gradient is realized. For the point features in the first level, the weight coefficient is positively correlated with the local density value; for the line features in the second level, the weight coefficient takes into account both the endpoint density and the line segment curvature; for the surface features in the third level, the weight coefficient combines the feature point density and the normal vector change of the patch; for the volume features in the fourth level, the weight coefficient comprehensively considers the density distribution and shape complexity of the volume element.

[0069] In the feature weighting process of this embodiment, a normalization processing mechanism for feature data is developed. First, standardize the feature data of each level to eliminate the influence of different feature dimensions. Then perform matrix multiplication on the calculated weight coefficient and the standardized feature data to obtain the weighted feature data. This weighting method ensures the reasonable retention of features at different levels during the simplification process.

[0070] When constructing the adaptive octree grid in this embodiment, a recursive subdivision strategy based on weighted features is realized. First, establish an initial octree structure within the bounding box of the production line model, and then determine whether further subdivision is required according to the weighted feature value within each node. For areas with higher weighted feature values, such as robotic arm joints, sensor installation positions, etc., perform more detailed grid division.

[0071] In the process of constructing grid nodes in this embodiment, an adaptive node arrangement strategy is adopted. In areas where features change violently, such as equipment boundaries, irregular surfaces, etc., increase the node density; in areas where features change smoothly, such as flat wall panels, straight pipelines, etc., appropriately reduce the node density. This differential node arrangement strategy not only ensures the accurate expression of key features but also reduces data redundancy.

[0072] In this embodiment, Haar wavelet basis functions are embedded into grid nodes to achieve multi - resolution feature representation. At each grid node, wavelet basis functions of appropriate scales are selected according to the complexity of local features. High - frequency wavelet basis functions are used for regions with rich details, and low - frequency wavelet basis functions are used for smooth regions, thus realizing multi - scale decomposition of features.

[0073] In the wavelet coefficient thresholding process of this embodiment, an adaptive threshold selection strategy is developed. By analyzing the energy distribution of wavelet coefficients, a threshold determination method based on signal - to - noise ratio is established. Different thresholds are set for wavelet coefficients at different levels to ensure the retention of important features and the removal of secondary features. During the thresholding process, the coefficient correlation between adjacent scales is considered to avoid sudden changes and discontinuities of features.

[0074] Finally, in this embodiment, the simplified grid is reconstructed through inverse wavelet transform. During the reconstruction process, a progressive feature recovery strategy is adopted. First, the main features are reconstructed, and then secondary features are gradually added until the preset simplification accuracy is reached. The reconstructed grid retains the key features of the original model while significantly reducing the data volume.

[0075] Through the above implementation scheme, this embodiment establishes a complete system of feature weighting and grid simplification. This scheme is particularly suitable for simplifying digital models of complex automated production lines, can adaptively retain key structures according to the importance of features, and provides an efficient data basis for subsequent model applications. Practical applications show that this scheme can effectively reduce the model complexity while maintaining the integrity of the key features of the production line, providing reliable digital model support for equipment monitoring and maintenance.

[0076] Step S103: Input the initial simplified grid into the multi - scale feature fusion module. The multi - scale feature fusion module extracts the topological structure features of the initial simplified grid at a coarse scale and extracts the local geometric features of the initial simplified grid at a fine scale. A feature correspondence graph is constructed based on the topological structure features and the local geometric features. The feature correspondence graph is matched with the morphological parameters and position parameters in the significant feature database, and the nodes of the initial simplified grid are adjusted according to the matching results to generate a final simplified model.

[0077] Optionally, in this embodiment, during the multi - scale feature fusion process, an adaptive scale decomposition strategy is adopted for the initial simplified grid. For different functional units in the automated production line, such as conveyor belt systems, robotic arms, control cabinets, etc., different feature analysis scales are used. In the coarse - scale analysis, the feature analysis window covers the entire functional unit; in the fine - scale analysis, the feature analysis window focuses on key components and connection structures.

[0078] In this embodiment, a hierarchical topological structure analysis method is implemented during the extraction of coarse-scale features. First, a grid connectivity graph is constructed, where the nodes in the graph represent grid cells and the edges represent the connection relationships between cells. By analyzing the degree distribution characteristics of the connectivity graph, key connection points and functional nodes in the production line are identified. For example, at the joints of the robotic arm, there are often high degree values, while at the straight segments of the conveyor belt, lower degree values are presented.

[0079] In this embodiment, the global structural characteristics of the connectivity graph are analyzed through the eigenvector decomposition method. The eigenvalues and eigenvectors of the Laplacian matrix are calculated, and the spectral clustering method is used to divide the grid into sub-regions with similar topological features. This spectral analysis-based method can effectively identify the boundaries of functional modules and key connection structures in the production line.

[0080] In the fine-scale feature extraction stage of this embodiment, a local geometric feature analysis framework is developed. By calculating the curvature tensor of each grid node, the principal curvature values and principal curvature directions are obtained. For the irregular surfaces in the production line, such as the groove structure of the guide rail and the mounting seat of the sensor, these local geometric features have important identification value.

[0081] In the feature fusion process of this embodiment, a feature correspondence learning mechanism based on graph neural network is implemented. The coarse-scale topological features and fine-scale geometric features are respectively encoded as feature vectors and input into the graph neural network. The network gradually learns the correlation patterns between features through multi-layer graph convolution operations. During the convolution process, the attention mechanism is adopted to highlight the influence of important features and weaken the contribution of secondary features.

[0082] In constructing the feature correspondence graph of this embodiment, a multi-level feature matching strategy is adopted. First, the correspondence relationships between functional modules are established at the coarse scale, and then the matching of local features is improved at the fine scale. By calculating the similarity matrix between feature vectors, a soft correspondence relationship between feature points is established, avoiding the limitations of traditional hard matching methods.

[0083] In the feature matching process of this embodiment, a reference matching mechanism based on a significant feature database is developed. The node features in the feature correspondence graph are compared with the morphological parameters and position parameters stored in the database. A fuzzy matching strategy is adopted during the matching, allowing feature deviations within a certain range, which improves the robustness of the matching.

[0084] In the grid node adjustment stage of this embodiment, a refined method for node positions based on optimization is implemented. An objective function for optimizing node positions is constructed, which simultaneously considers the feature preservation constraint and the grid quality constraint. The feature preservation constraint ensures that the adjusted node positions are consistent with the matched reference features, and the grid quality constraint avoids the generation of distorted cells.

[0085] In this embodiment, an iterative optimization algorithm is used to solve the node positions. In each iteration, first, the deviation between the current mesh and the target features is calculated, and then the node positions are updated based on the gradient information. To improve the optimization efficiency, an adaptive step-size strategy is adopted. A larger step size is used at the beginning of the optimization for rapid convergence, and a smaller step size is used later for fine adjustment.

[0086] Finally, in this embodiment, the optimized mesh is smoothed to eliminate local irregularities. The Laplacian smoothing algorithm that preserves features is used in the smoothing process to ensure that important features are not damaged while improving the mesh quality. The final simplified model generated in this way not only retains the key features of the production line but also has good mesh quality.

[0087] Through the above implementation solution, this embodiment establishes a complete multi-scale feature fusion framework. This solution is particularly suitable for the simplification of digital models of complex automated production lines, can accurately identify and retain the key features in the production line, and significantly reduce the model complexity at the same time. Practical applications show that this solution can effectively support the digital monitoring and optimization analysis of the production line, providing a reliable digital model basis for intelligent manufacturing.

[0088] As can be seen from the above description, the method for simplifying the digital model of an automated production line provided by the embodiment of this application can identify and save key feature information through the significant feature extraction module, and establish a four-level topology structure tree. An adaptive feature weight calculation mechanism is innovatively designed, and intelligent simplification of the mesh is achieved by combining multi-resolution wavelet analysis. The system extracts and matches the topology structure features and local geometric features at different scales through the multi-scale feature fusion module, realizing the precise preservation of key features during the model simplification process. This method breaks through the limitations of traditional model simplification and provides an intelligent and automated solution for the efficient processing of large-scale digital models.

[0089] In an embodiment of the method for simplifying the digital model of an automated production line of this application, refer to Figure 2 , and it may specifically include the following content:

[0090] Step S201: Discretize the digital model of the automated production line to obtain point cloud data, establish a three-dimensional mesh structure based on the point cloud data, calculate the curvature value, normal vector, and neighborhood density of each vertex in the three-dimensional mesh structure, input the curvature value, normal vector, and neighborhood density into the significance calculation formula to obtain a significance coefficient, and construct a significant feature extraction module based on the significance coefficient;

[0091] Step S202: Compare the significance coefficient in the significance feature extraction module with a first threshold, extract the vertices with a significance coefficient greater than the first threshold as feature points, construct a set of significant feature points based on the spatial position relationship of the feature points, and store the geometric attribute parameters and topological attribute parameters of the set of significant feature points as morphological parameters and position parameters respectively.

[0092] Optionally, in this embodiment, when discretizing the digital model of the automated production line, a multi-view scanning strategy is adopted. For different types of equipment in the production line, such as robotic arms, conveyor belts, inspection equipment, etc., the optimal scanning angles and sampling densities are set respectively. A higher sampling density is used at the equipment connection points and functional interaction areas; a lower sampling density is used in the planar areas and simple structures to ensure the integrity and pertinence of the collected data.

[0093] In this embodiment, an adaptive noise reduction algorithm is implemented during the acquisition of point cloud data. First, noise points are identified through statistical outlier analysis, and the distance distribution between each point and its neighborhood points is calculated. A smaller noise threshold is used in the precision component areas of the production line, such as the sensor installation positions and precision fixtures; a larger noise threshold is used in the large structural component areas, such as the support frames and protective covers.

[0094] In this embodiment, a grid division strategy based on local features is developed when constructing the three-dimensional grid structure. First, an initial triangular grid is generated using an improved Poisson reconstruction algorithm, and then the grid is optimized according to the local geometric features. The grid density is increased in the areas with drastic curvature changes, such as equipment corners and pipe elbows; the grid density is appropriately reduced in the flat areas, realizing an adaptive distribution of the grid density.

[0095] In this embodiment, a multi-scale feature extraction framework is implemented in the vertex feature calculation link. For each grid vertex, first determine the neighborhood radius for feature calculation. A smaller neighborhood radius is used in the areas with rich details, and a larger neighborhood radius is used in the areas with simple features. Within this neighborhood, the principal curvature and Gaussian curvature are calculated by fitting a local quadratic surface, and the normal vector direction is estimated simultaneously.

[0096] In this embodiment, an adaptive kernel function method is adopted during the neighborhood density calculation. The kernel function bandwidth is dynamically adjusted according to the local point cloud distribution characteristics, with a larger bandwidth used in the sparse point cloud areas and a smaller bandwidth used in the dense areas. This adaptive strategy ensures the accuracy of density estimation, especially in the special-shaped structure and precision component areas of the production line.

[0097] In this embodiment, a comprehensive feature fusion model was developed for saliency calculation. This model combines curvature features, normal vector changes, and neighborhood density into a saliency coefficient through non-linear mapping. During the calculation process, dynamic weights are assigned to different features. For example, the curvature weight is increased in the edge feature region, the normal vector weight is increased in the surface feature region, and the density weight is increased in the connection structure region.

[0098] In the construction process of the saliency feature extraction module in this embodiment, a feature scale adaptive mechanism was implemented. By analyzing the spatial distribution law of local features, the scale parameter of feature extraction is dynamically adjusted. For precision components in the production line, a smaller feature extraction scale is adopted; for large structural components, a larger feature extraction scale is used.

[0099] In the process of feature point selection in this embodiment, a multi-level threshold screening strategy is adopted. First, a global threshold is used for preliminary screening, and then local adaptive thresholds are used for fine screening in different functional regions. This hierarchical screening method ensures the representativeness and integrity of feature points, especially in the key functional regions of the production line.

[0100] In this embodiment, when constructing a set of significant feature points, an organization method based on spatial clustering was developed. By analyzing the spatial distance and feature similarity between feature points, the related feature points are organized into a set of feature points. During the clustering process, the functional module division of the production line is considered to ensure a good correspondence between the set of feature points and the actual functional units.

[0101] In the parameter storage section of this embodiment, a hierarchical data organizational structure was implemented. The morphological parameters include local geometric features, curvature distribution, surface properties, etc.; the position parameters include spatial coordinates, relative position relationships, topological connectivity, etc. By establishing a parameter index table, efficient access and management of feature data are achieved.

[0102] Through the above implementation solutions, this embodiment established a complete feature extraction and storage system. This solution is particularly suitable for feature recognition in complex automated production lines and can accurately extract and save the key feature information in the production line. Practical applications show that this solution can effectively support the digital modeling and optimization analysis of the production line, providing a reliable data basis for equipment monitoring and maintenance.

[0103] In an embodiment of the method for simplifying the digital model of an automated production line in this application, refer to Figure 3 , and it may specifically include the following content:

[0104] Step S301: Input the morphological parameters and position parameters of the significant feature point set into the feature encoder. The feature encoder performs geometric feature encoding on the morphological parameters to obtain a geometric feature vector, performs spatial position encoding on the position parameters to obtain a position feature vector, combines the geometric feature vector and the position feature vector to form feature data, and writes the feature data into the significant feature database according to the hierarchical structure.

[0105] Step S302: Construct a four-level topological structure tree based on the significant feature point set. Use the point data in the significant feature point set as the first-level nodes, the connection relationships between adjacent points as the second-level nodes, the patches enclosed by closed line segments as the third-level nodes, and the volume elements composed of closed patches as the fourth-level nodes. Store the corresponding feature data index in each node of the four-level topological structure tree.

[0106] Optionally, in the feature encoding process of this embodiment, a two-stream feature encoding architecture is developed. For the morphological parameters of the production line, such as the surface curvature of the equipment, edge features, material properties, etc., a deep convolutional network is used for geometric feature encoding; for the position parameters, such as spatial coordinates, relative position relationships, assembly constraints, etc., a graph attention network is used for spatial position encoding. This two-stream encoding architecture can fully extract the geometric and spatial feature information in the production line.

[0107] In the geometric feature encoding link of this embodiment, a multi-scale feature extraction mechanism is implemented. Through the pyramid pooling structure, morphological features are extracted at different spatial scales. For precision components in the production line, such as sensor mounts, positioning pins, etc., detailed features are extracted at a small scale; for large structural components, such as machine frames, protective covers, etc., overall morphological features are extracted at a large scale. Residual connections are set in each layer of the encoding network to ensure the effective transmission of feature information.

[0108] In the position feature encoding process of this embodiment, a hierarchical spatial relationship modeling method is adopted. First, a local reference coordinate system is established to map each functional unit in the production line to a unified feature space. Then, the spatial dependence relationships between different units are captured through the graph attention mechanism, and the attention weights are dynamically adjusted according to the distance and functional association degree between the units. This encoding method is particularly suitable for expressing the complex spatial layout and connection relationships in the production line.

[0109] In the feature vector combination stage of this embodiment, an adaptive feature fusion strategy is developed. The fusion weights of the geometric feature vector and the position feature vector are dynamically adjusted through a gating mechanism, increasing the weight of the position feature at key positions such as equipment joints, and increasing the weight of the geometric feature in areas with complex shapes. The fused feature data retains both local geometric details and global spatial information.

[0110] In this embodiment, a multi-level indexing mechanism is implemented during the database writing process. A hierarchical storage structure for feature data is established, including a geometric feature layer, a spatial relationship layer, and a semantic attribute layer. Each layer of data is equipped with an efficient retrieval index, supporting multi-dimensional queries such as by feature type, spatial location, and functional attributes. This structural design significantly improves the access efficiency of feature data.

[0111] In this embodiment, when constructing a four-level topology structure tree, a bottom-up construction strategy is adopted. First, the set of significant feature points is mapped to the first-level nodes, and each node contains the coordinate information and feature attributes of the points. Then, the spatial adjacency relationships between the points are analyzed, and the second-level connection relationship nodes are constructed through the minimum spanning tree algorithm. These connections reflect the physical and logical associations in the production line.

[0112] In this embodiment, during the patch construction phase, an adaptive triangulation algorithm is developed. By analyzing the closure and coplanarity features of the line segments, the third-level patch nodes are identified and constructed. During the construction process, geometric constraints unique to the production line, such as parallelism and perpendicularity, are considered to ensure that the generated patches conform to the actual structural features.

[0113] In this embodiment, during the voxel construction process, a recognition method based on patch closure is implemented. By analyzing the connection relationships and normal vector distributions of the patches, the closed voxel structures are identified as the fourth-level nodes. For complex components in the production line, such as valve assemblies and transmission devices, a hierarchical combination method is used to construct their complete voxel representations.

[0114] In this embodiment, during the construction of the node feature index, a distributed storage strategy is adopted. In each level of nodes, in addition to storing the feature data index of this level, an association index with other level nodes is also established. This multi-level index structure supports fast feature retrieval and association analysis.

[0115] Through the above implementation solution, this embodiment establishes a complete feature coding and topology structure expression system. This solution is particularly suitable for the digital expression of complex automated production lines and can effectively capture and organize multi-dimensional feature information in the production line. Practical applications show that this solution reliably supports the feature analysis and model simplification of the production line, providing a solid data foundation for the digital twin and intelligent operation and maintenance of the production line.

[0116] In an embodiment of the method for simplifying the digital model of an automated production line in this application, refer to Figure 4 , and it may specifically include the following content:

[0117] Step S401: Construct a spatial density distribution map based on the four - level topological structure tree. Calculate the number of significant feature points within the neighborhood range of each node in the spatial density distribution map. Divide the number by the neighborhood volume to obtain the local density value. Perform normalization processing on the local density value to obtain the density distribution coefficient. Input the density distribution coefficient into the weight mapping function to generate the weight coefficient corresponding to each level.

[0118] Step S402: Perform matrix multiplication on the weight coefficient and the feature data of each level node in the four - level topological structure tree to generate a weighted feature matrix. Perform standardization processing on the weighted feature matrix to obtain weighted feature data. Re - organize the weighted feature data into a tree structure according to the hierarchical relationship.

[0119] Optionally, in this embodiment, when constructing the spatial density distribution map, an adaptive grid division strategy is adopted. According to the feature distribution characteristics of different functional areas in the production line, the grid size is dynamically adjusted. A smaller grid size is used in feature - dense areas such as equipment connection points and control panels, and a larger grid size is used in feature - sparse areas such as conveyor belts and pipelines to ensure that the density distribution map can accurately reflect the spatial distribution law of production line features.

[0120] In this embodiment, when calculating the node neighborhood range, a multi - scale neighborhood analysis method is implemented. First, determine the basic neighborhood radius of each node, which is inversely proportional to the feature density of the area where it is located. Then, use the spherical search algorithm to count the number of significant feature points within the neighborhood. For key functional components in the production line, such as sensor components and actuators, a smaller neighborhood radius is used to ensure the accurate expression of local features.

[0121] In this embodiment, during the local density calculation process, an adaptive kernel density estimation algorithm is developed. By analyzing the spatial distribution pattern of feature points, the bandwidth parameter of the kernel function is dynamically adjusted. In areas with uneven feature distribution, such as robotic arm joints, a variable bandwidth strategy is adopted to ensure the accuracy of density estimation. The calculated local density value not only reflects the aggregation degree of features but also retains the continuity of spatial distribution.

[0122] In this embodiment, when performing normalization processing on the density distribution coefficient, a hierarchical normalization strategy is adopted. According to the feature distribution characteristics of different levels, corresponding normalization functions are designed. For the point level, maximum - minimum normalization is used; for the line level, the influence of line segment length is considered; for the surface level, area weights are introduced; for the volume level, normalization is combined with volume factors. This hierarchical normalization method ensures the comparability of features at different levels.

[0123] In the design of the weight mapping function in this embodiment, an adaptive mapping mechanism based on feature importance is implemented. For different functional areas of the production line, such as workstations, logistics channels, control areas, etc., a feature importance evaluation model is established. This model comprehensively considers factors such as the importance of regional functions, operation frequencies, and maintenance requirements, and maps the density distribution coefficient to a reasonable weight value.

[0124] In the process of generating the weight coefficient in this embodiment, a weight balance mechanism between levels is developed. By analyzing the interdependent relationships of features at each level, a weight adjustment rule is established. For example, when a certain voxel node has important functional attributes, the weights of the patches, line segments, and points it contains are correspondingly increased to ensure the hierarchical consistency of features.

[0125] In the matrix multiplication operation step of this embodiment, a sparse matrix optimization algorithm is implemented. Considering the sparsity characteristics of the production line feature data, a compressed storage format and an efficient sparse matrix operation method are adopted. Through a block calculation strategy, the processing efficiency of large-scale feature data is significantly improved.

[0126] When performing the normalization process on the weighted feature matrix in this embodiment, a feature correlation preservation strategy is adopted. By analyzing the covariance structure between features, a normalization method that preserves feature correlations is designed. This method not only achieves the unification of feature scales but also retains the important correlation information between features.

[0127] In the process of reorganizing the tree structure in this embodiment, a feature clustering optimization algorithm is developed. Based on the similarity of weighted features, hierarchical clustering is performed on feature nodes. During the clustering process, the functional module division of the production line is considered to ensure that the reorganized tree structure reflects the actual functional hierarchical relationship.

[0128] In this embodiment, through a dynamic feature indexing mechanism, an efficient organization of the tree structure is achieved. Multidimensional indexes are established for each node, including spatial position indexes, feature type indexes, and functional attribute indexes. This indexing mechanism supports fast feature retrieval and update operations.

[0129] Through the above implementation solutions in this embodiment, a complete feature weighting and organization system is established. This solution is particularly suitable for the feature importance analysis of complex automated production lines, can reasonably allocate feature weights according to actual application requirements, and provides a reliable weight basis for subsequent model simplification. Practice shows that this solution can effectively support the feature optimization and simplification of the production line, and significantly improves the practicality of the digital model in actual applications.

[0130] In an embodiment of the method for simplifying the digital model of an automated production line in this application, refer to Figure 5 , and it may also specifically include the following content:

[0131] Step S501: Determine a spatial bounding box based on the weighted feature data, recursively divide the spatial bounding box into an octree structure, calculate the local variance of the weighted feature data at each node of the octree structure, and continue to subdivide the node when the local variance of the node is greater than the second threshold until the preset subdivision level is reached or the local variance is less than the second threshold. Build an adaptive grid on the octree nodes after subdivision is completed, and map the Haar wavelet basis function to the node coordinates of the adaptive grid;

[0132] Step S502: Perform coefficient decomposition on the Haar wavelet basis function on the adaptive grid nodes, calculate the energy value of the wavelet coefficients at each layer, set an adaptive threshold according to the energy value, set the wavelet coefficients less than the adaptive threshold to zero, perform a reconstruction operation on the remaining wavelet coefficients to generate an initial simplified grid, and establish a corresponding relationship between the nodes of the initial simplified grid and the original weighted feature data.

[0133] Optionally, in this embodiment, when determining the spatial bounding box, a feature distribution adaptive strategy is adopted. According to the spatial distribution characteristics of the weighted feature data, the size and direction of the bounding box are dynamically adjusted. For the movable part areas such as the robotic arm in the production line, the bounding box range is appropriately enlarged to accommodate the movement space; for the fixed equipment areas, the bounding box is set closely along the equipment contour to improve the space utilization efficiency.

[0134] In the process of recursively dividing the octree structure in this embodiment, an adaptive subdivision mechanism based on feature distribution is realized. By analyzing the distribution characteristics of the weighted feature data in each node, a variance calculation model is established. This model not only considers the dispersion degree of the feature values, but also combines the weight information of the features to ensure more detailed division in the key functional areas.

[0135] In the local variance calculation link of this embodiment, a multi-scale feature aggregation algorithm is developed. For feature distributions at different scales, a weighted statistical method is used to calculate the local variance. In the precision control areas of the production line, such as the sensor installation positions and precision positioning structures, a smaller statistical window is used; in the large structural part areas, a larger statistical window is used.

[0136] In the octree subdivision control of this embodiment, a dual-threshold adaptive mechanism is realized. In addition to using the local variance threshold, a depth control threshold is also introduced. In areas where the features change violently, such as the equipment interfaces and function conversion areas, deeper subdivision is allowed; in areas where the features change gently, the subdivision depth is appropriately controlled to avoid the computational burden caused by excessive subdivision.

[0137] In this embodiment, during the adaptive mesh construction process, a mesh generation strategy that preserves features is adopted. Based on the distribution characteristics of octree nodes, a tetrahedral mesh with variable distribution density is constructed. In the key functional areas of the production line, such as mechanical interfaces, control panels, etc., the mesh density is increased; in simple structure areas, the mesh density is appropriately reduced.

[0138] In this embodiment, when mapping the Haar wavelet basis function, a mapping mechanism based on feature importance is developed. According to the feature weights of the regions where the mesh nodes are located, wavelet basis functions of appropriate scales are selected. For high-weight regions, more basis functions are used to improve the expression accuracy; for low-weight regions, fewer basis functions are used to reduce the data volume.

[0139] In this embodiment, during the wavelet coefficient decomposition process, a multi-level energy analysis method is implemented. The feature information is decomposed into different frequency levels through wavelet transform, and the energy distribution of each level is calculated. In the energy calculation, the weight information of the features is considered to ensure that important features are fully reflected in the energy distribution.

[0140] In this embodiment, in the adaptive threshold setting step, a dynamic threshold strategy based on energy distribution is adopted. By analyzing the energy distribution law of wavelet coefficients, an adaptive threshold determination model is established. This model effectively removes minor features with low energy while preserving important features, achieving efficient data compression.

[0141] In this embodiment, during the wavelet coefficient reconstruction process, a feature-preserving reconstruction algorithm is developed. The simplified mesh is reconstructed by layer-by-layer stacking of the retained wavelet coefficients. During the reconstruction process, a progressive strategy is adopted. First, the main features are reconstructed, and then the minor features are gradually added until the preset accuracy requirements are met.

[0142] In this embodiment, during the establishment of node correspondence relationships, a mapping mechanism based on feature similarity is implemented. By calculating the geometric distance and feature similarity between the nodes of the simplified mesh and the original feature data, an optimal matching relationship is established. This mapping mechanism ensures the feature consistency between the simplified model and the original model.

[0143] Through the above implementation solutions, this embodiment establishes a complete mesh simplification and feature preservation system. This solution is particularly suitable for the simplification of digital models of complex automated production lines, and can significantly reduce the data volume while preserving key features. Practice shows that this solution can effectively support the digital monitoring and optimization analysis of production lines, providing an efficient digital model basis for intelligent manufacturing.

[0144] In an embodiment of the digital model simplification method for an automated production line in this application, refer to Figure 6 , and it may also specifically include the following content:

[0145] Step S601: Perform multi-level decomposition on the initial simplified mesh to obtain mesh representations at different scales. Construct a mesh connectivity graph at the coarse scale level, calculate the degree distribution and connectivity properties of each node in the mesh connectivity graph, construct a mesh skeleton based on the degree distribution and connectivity properties, and extract the branch structure and connection relationship of the mesh skeleton as topological structure features;

[0146] Step S602: Calculate the local curvature tensor of the nodes of the initial simplified mesh at the fine scale level, calculate the principal curvature values and principal curvature directions based on the curvature tensor, combine the principal curvature values and principal curvature directions to form a curvature descriptor, perform clustering analysis on the curvature descriptor to obtain local geometric features, and establish a mapping relationship between the local geometric features and the mesh nodes.

[0147] Optionally, in the multi-level decomposition process of this embodiment, an adaptive mesh simplification strategy is adopted. Different simplification parameters are set for different functional modules in the production line, such as conveyor devices, robotic arms, control units, etc. A higher mesh accuracy is maintained in the key functional areas, and the accuracy is appropriately reduced in the secondary areas to ensure that the simplified mesh can accurately represent the geometric features of the production line.

[0148] In the mesh connectivity graph construction link of this embodiment, a connection weight mechanism based on functional association is implemented. By analyzing the spatial relationship and functional dependence between mesh nodes, different weights are assigned to the connection edges. For example, a higher weight is assigned to the connection at the device interface, and a lower weight is assigned to the connection at the ordinary structure, so as to highlight the key connection structures in the production line.

[0149] In the process of calculating the degree distribution of this embodiment, a multi-scale statistical analysis method is developed. First, calculate the degree characteristics of the nodes in the local neighborhood, and then obtain the degree distribution at different scales through hierarchical aggregation. This method can effectively identify the key connection points in the production line, such as the positions of multi-device intersections, functional module interfaces, etc.

[0150] When constructing the mesh skeleton in this embodiment, a feature-preserving skeleton extraction algorithm is adopted. By analyzing the degree distribution and connectivity properties, the main structure lines of the mesh are identified. During the extraction process, special attention is paid to the functional flow lines of the production line, such as the material transmission path, signal transmission channel, etc., to ensure that the skeleton structure can reflect the functional layout of the production line.

[0151] In the branch structure extraction link of this embodiment, a hierarchical extraction mechanism based on functional modules is implemented. First, identify the main functional lines, such as the main conveyor belt path, main control channel, etc., and then gradually extract the secondary branches, such as auxiliary conveyor devices, monitoring equipment connections, etc. This hierarchical extraction method effectively expresses the functional hierarchical relationship of the production line.

[0152] In this embodiment, during the calculation of the curvature tensor, an adaptive neighborhood selection strategy is developed. The size of the calculation neighborhood is dynamically adjusted according to the complexity of the local surface. At precise structures on the production line, such as sensor mounting seats and positioning reference planes, a smaller calculation neighborhood is adopted; while at large planar structures, a larger calculation neighborhood is used.

[0153] In the principal curvature analysis step of this embodiment, a robust eigen-direction estimation method is adopted. By performing eigenvalue decomposition on the curvature tensor, the principal curvature values and principal curvature directions are obtained. During the calculation process, the influence of noise and local perturbations is considered to ensure that the extracted curvature features have good stability.

[0154] In the process of constructing the curvature descriptor in this embodiment, a multi-scale feature fusion mechanism is implemented. The principal curvature information at different scales is weighted and combined to form a comprehensive curvature descriptor. The weight coefficients are dynamically adjusted according to the importance of the features to ensure that the descriptor can accurately represent the local geometric features.

[0155] In the clustering analysis stage of this embodiment, an adaptive clustering algorithm based on geometric similarity is developed. By analyzing the distribution characteristics of the curvature descriptors, the number of clusters and the cluster centers are dynamically determined. During the clustering process, the specific geometric feature types on the production line, such as cylindrical surfaces, planar surfaces, and irregular surfaces, are considered.

[0156] In the process of establishing the feature mapping relationship in this embodiment, a two-way association mechanism is adopted. Not only is a forward mapping from geometric features to grid nodes established, but also a reverse index from grid nodes to geometric features is constructed. This two-way mapping mechanism supports fast feature query and local update operations.

[0157] Through the above implementation solutions, this embodiment establishes a complete multi-scale feature analysis system. This solution is particularly suitable for feature extraction and expression of complex automated production lines and can accurately capture the geometric and topological features of the production line at different scales. Practical applications show that this solution can effectively support the feature analysis and optimal design of the production line, providing reliable technical support for digital transformation.

[0158] In an embodiment of the method for simplifying the digital model of an automated production line in this application, refer to Figure 7 , and it may specifically include the following content:

[0159] Step S701: Input the topological structure features and the local geometric features into a feature fusion network. The feature fusion network generates a feature correspondence graph based on graph convolution operations, calculates the similarity matrix between the nodes in the feature correspondence graph, performs a matching operation between the similarity matrix and the shape parameters and position parameters stored in the significant feature database to generate a feature matching metric value, and constructs a node optimization objective function based on the feature matching metric value;

[0160] Step S702: Iteratively solve the node optimization objective function, adjust the node positions and connection relationships of the initial simplified mesh. Stop the iteration when the objective function value is less than the third threshold or the maximum number of iterations is reached. Reconstruct the final mesh node coordinates and connection relationships into a 3D model, and perform mesh smoothing on the reconstructed 3D model to obtain the final simplified model.

[0161] Optionally, in the design of the feature fusion network in this embodiment, a multi-layer graph convolution structure is adopted. This network first receives topological structure features and local geometric features as inputs, and gradually fuses these feature information through multi-layer graph convolution operations. In key functional areas of the production line, such as equipment interfaces, control units, etc., the network uses deeper convolution layers to capture more complex feature correlations.

[0162] In the process of graph convolution operation in this embodiment, an adaptive receptive field mechanism is implemented. For different types of functional modules in the production line, the receptive field range of graph convolution is dynamically adjusted. For example, for complex moving components such as robotic arms, a larger receptive field is used to capture overall motion features; for precision positioning structures, a smaller receptive field is adopted to retain local details.

[0163] In the similarity matrix calculation step of this embodiment, a multi-dimensional feature similarity measurement method is developed. By combining geometric similarity, topological similarity, and functional similarity, a comprehensive similarity evaluation model is constructed. This model emphasizes functional consistency when calculating the similarity at equipment connections, and pays more attention to the matching degree of geometric shapes when evaluating the similarity of structural components.

[0164] In the process of feature matching operation in this embodiment, a hierarchical matching strategy is adopted. First, perform fast matching at a coarse-grained level to screen out candidate feature pairs; then perform precise matching at a fine-grained level to determine the optimal feature correspondence. This hierarchical matching method significantly improves the processing efficiency of large-scale feature data.

[0165] When constructing the objective function in this embodiment, a multi-objective balance mechanism is implemented. The objective function not only considers the feature matching metric value, but also introduces a geometric preservation term and a topological preservation term. During the optimization process, the weights of each item are dynamically adjusted according to the functional importance of different regions to ensure that the key features are retained in the simplified model.

[0166] In the process of iterative solution in this embodiment, an adaptive step size control strategy is developed. By analyzing the gradient information of the objective function, the optimization step size is dynamically adjusted. A smaller step size is used in regions where the features change violently, and a larger step size is used in regions where the features are smooth, improving the optimization efficiency while ensuring convergence stability.

[0167] In the node position adjustment stage of this embodiment, a constraint-based optimization method is adopted. By establishing geometric constraints and topological constraints, the functional characteristics of the production line are ensured during the node adjustment process. For example, flatness constraints are maintained on the equipment installation surface, cylindricity constraints are maintained on the bearing seat, and relative position constraints are maintained on the precision positioning structure.

[0168] In the process of optimizing the connection relationship of this embodiment, a topology-preserving reconstruction mechanism is realized. By analyzing the importance of connections between nodes, key connections are preferentially maintained, and secondary connections are appropriately simplified. This differential processing strategy ensures that while the simplified model maintains functional integrity, the data volume is significantly reduced.

[0169] In the three-dimensional model reconstruction stage of this embodiment, a feature-aware reconstruction algorithm is developed. Based on the optimized mesh nodes and connection relationships, a progressive reconstruction strategy is adopted. First, the main functional structures are reconstructed, and then detailed features are gradually added to ensure the quality and efficiency of the reconstructed model.

[0170] In the mesh smoothing process of this embodiment, a smoothing algorithm that preserves local features is adopted. By analyzing local curvature and feature line information, the sharpness of key features is maintained during the smoothing process. For important features such as precision mating surfaces and positioning references in the production line, a smaller smoothing intensity is used; for ordinary transition surfaces, a larger smoothing intensity can be used.

[0171] Through the above implementation solutions, this embodiment establishes a complete model simplification and optimization system. This solution is particularly suitable for simplifying the digital models of complex automated production lines, and can significantly reduce the data volume while maintaining key functional characteristics. Practical applications show that this solution reliably supports the virtual simulation and digital twin applications of the production line, providing an efficient digital foundation for intelligent manufacturing.

[0172] In order to break through the limitations of traditional model simplification and provide an intelligent and automated solution for the efficient processing of large-scale digital models, this application provides an embodiment of an automated production line digital model simplification device for implementing all or part of the content of the above-mentioned automated production line digital model simplification method. See Figure 8 , the automated production line digital model simplification device specifically includes the following contents:

[0173] The feature extraction module 10 is used to construct a saliency feature extraction module. The saliency feature extraction module calculates saliency scores for point, line, surface, and volume data in the automated production line digital model, extracts feature points with saliency scores exceeding the first threshold as a set of salient feature points, and at the same time stores the morphological parameters and position parameters corresponding to the set of salient feature points in the salient feature database, and establishes a four-level topological structure tree based on the set of salient feature points;

[0174] The architecture construction module 20 is configured to input the four - level topological structure tree into the adaptive feature weight calculation module. The adaptive feature weight calculation module calculates the weight coefficient of each level based on the spatial distribution density of the significant feature point set, multiplies the weight coefficient by the feature data of the corresponding level to obtain weighted feature data, inputs the weighted feature data into the adaptive octree grid construction module. The adaptive octree grid construction module constructs grid nodes based on the weighted feature data, embeds multi - resolution wavelet basis functions into the grid nodes, performs hierarchical thresholding processing on the wavelet function coefficients, and reconstructs the thresholded wavelet coefficients into an initial simplified grid;

[0175] The model simplification module 30 is configured to input the initial simplified grid into the multi - scale feature fusion module. The multi - scale feature fusion module extracts the topological structure features of the initial simplified grid at a coarse scale, extracts the local geometric features of the initial simplified grid at a fine scale, constructs a feature correspondence graph based on the topological structure features and the local geometric features, matches the feature correspondence graph with the morphological parameters and position parameters in the significant feature database, and adjusts the nodes of the initial simplified grid according to the matching result to generate a final simplified model.

[0176] As can be seen from the above description, the automated production line digital model simplification device provided by the embodiments of the present application can identify and save key feature information through the significant feature extraction module and establish a four - level topological structure tree. An adaptive feature weight calculation mechanism is innovatively designed, and intelligent grid simplification is achieved by combining multi - resolution wavelet analysis. The system extracts and matches topological structure features and local geometric features at different scales through the multi - scale feature fusion module, and precisely preserves key features during the model simplification process. This method breaks through the limitations of traditional model simplification and provides an intelligent and automated solution for the efficient processing of large - scale digital models.

[0177] At the hardware level, in order to break through the limitations of traditional model simplification and provide an intelligent and automated solution for the efficient processing of large - scale digital models, the present application provides an embodiment of an electronic device for implementing all or part of the content in the automated production line digital model simplification method. The electronic device specifically includes the following content:

[0178] A processor, a memory, a communications interface, and a bus; wherein, the processor, the memory, and the communications interface complete communication with each other through the bus; the communications interface is used to implement information transmission between the automated production line digital model simplification device and related devices such as a core business system, a user terminal, and a related database, etc.; the logic controller can be a desktop computer, a tablet computer, a mobile terminal, etc., and this embodiment is not limited thereto. In this embodiment, the logic controller can be implemented with reference to the embodiments of the automated production line digital model simplification method and the embodiments of the automated production line digital model simplification device in the embodiments, the content of which is incorporated herein, and the repeated parts will not be elaborated again.

[0179] It can be understood that the user terminal may include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. Among them, the smart wearable device may include smart glasses, a smart watch, a smart bracelet, etc.

[0180] In practical applications, part of the automated production line digital model simplification method can be executed on the electronic device side as described above, or all operations can be completed in the client device. Specifically, it can be selected according to the processing capacity of the client device and the limitations of the user usage scenario, etc. This application does not make any limitations in this regard. If all operations are completed in the client device, the client device may further include a processor.

[0181] The above-mentioned client device may have a communication module (i.e., a communication unit), and can communicate with a remote server to realize data transmission with the server. The server may include a server on the task scheduling center side, and in other implementation scenarios, it may also include a server on an intermediate platform, such as a server on a third-party server platform having a communication link with the task scheduling center server. The server may include a single computer device, or may include a server cluster composed of multiple servers, or a server structure of a distributed device.

[0182] Figure 9 This is a schematic block diagram of the system composition of the electronic device 9600 according to an embodiment of the present application. As Figure 9 shown, the electronic device 9600 may include a central processing unit 9100 and a memory 9140; the memory 9140 is coupled to the central processing unit 9100. It should be noted that this Figure 9 is exemplary; other types of structures can also be used to supplement or replace this structure to implement telecommunications functions or other functions.

[0183] In one embodiment, the function of the digital model simplification method for the automated production line can be integrated into the central processing unit 9100. Among them, the central processing unit 9100 can be configured to perform the following controls:

[0184] Step S101: Construct a saliency feature extraction module. The saliency feature extraction module calculates the saliency scores for the point, line, surface, and volume data in the digital model of the automated production line, extracts the feature points with saliency scores exceeding the first threshold as the set of salient feature points, and at the same time stores the morphological parameters and position parameters corresponding to the set of salient feature points in the salient feature database. A four-level topological structure tree is established based on the set of salient feature points;

[0185] Step S102: Input the four-level topological structure tree into the adaptive feature weight calculation module. The adaptive feature weight calculation module calculates the weight coefficient for each level based on the spatial distribution density of the set of salient feature points, multiplies the weight coefficient by the feature data of the corresponding level to obtain the weighted feature data, inputs the weighted feature data into the adaptive octree grid construction module. The adaptive octree grid construction module constructs grid nodes based on the weighted feature data, embeds the multi-resolution wavelet basis function into the grid nodes, performs hierarchical thresholding on the wavelet function coefficients, and reconstructs the thresholded wavelet coefficients into the initial simplified grid;

[0186] Step S103: Input the initial simplified grid into the multi-scale feature fusion module. The multi-scale feature fusion module extracts the topological structure features of the initial simplified grid at a coarse scale and extracts the local geometric features of the initial simplified grid at a fine scale, constructs a feature correspondence graph based on the topological structure features and the local geometric features, matches the feature correspondence graph with the morphological parameters and position parameters in the salient feature database, and adjusts the nodes of the initial simplified grid according to the matching result to generate the final simplified model.

[0187] As can be seen from the above description, the electronic device provided in the embodiment of the present application identifies and saves key feature information through the saliency feature extraction module and establishes a four-level topological structure tree. An adaptive feature weight calculation mechanism is innovatively designed, and intelligent grid simplification is achieved by combining multi-resolution wavelet analysis. The system extracts and matches topological structure features and local geometric features at different scales through the multi-scale feature fusion module, and accurately preserves key features during the model simplification process. This method breaks through the limitations of traditional model simplification and provides an intelligent and automated solution for the efficient processing of large-scale digital models.

[0188] In another embodiment, the automated production line digital model simplification device can be separately configured from the central processing unit 9100. For example, the automated production line digital model simplification device can be configured as a chip connected to the central processing unit 9100, and the functions of the automated production line digital model simplification method can be realized through the control of the central processing unit.

[0189] As Figure 9 shown, the electronic device 9600 may further include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It should be noted that the electronic device 9600 does not necessarily have to include Figure 9 all the components shown in Figure 9 ; in addition, the electronic device 9600 may further include

[0190] As Figure 9 shown, the central processing unit 9100, sometimes also referred to as a controller or operation control, may include a microprocessor or other processor devices and / or logic devices. The central processing unit 9100 receives inputs and controls the operations of the various components of the electronic device 9600.

[0191] Among them, the memory 9140, for example, can be one or more of a buffer, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory, or other suitable devices. It can store the above information related to failures, and can also store programs for executing relevant information. And the central processing unit 9100 can execute the programs stored in the memory 9140 to implement information storage or processing, etc.

[0192] The input unit 9120 provides inputs to the central processing unit 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to supply power to the electronic device 9600. The display 9160 is used to display display objects such as images and texts. The display can be, for example, an LCD display, but is not limited thereto.

[0193] The memory 9140 can be a solid-state memory. For example, a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It can also be such a memory that stores information even when powered off, can be selectively erased and has more data. Examples of such a memory are sometimes referred to as EPROMs, etc. The memory 9140 can also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 can include an application / function storage unit 9142, and the application / function storage unit 9142 is used to store application programs and function programs or the processes for operating the electronic device 9600 through the central processing unit 9100.

[0194] The memory 9140 may further include a data storage unit 9143 for storing data such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for the communication functions of the electronic device and / or for performing other functions of the electronic device (such as a messaging application, an address book application, etc.).

[0195] The communication module 9110 is a transmitter / receiver that transmits and receives signals via the antenna 9111. The communication module 9110 (transmitter / receiver) is coupled to the central processor 9100 to provide input signals and receive output signals, which may be the same as in the case of a conventional mobile communication terminal.

[0196] Based on different communication technologies, multiple communication modules 9110 may be provided in the same electronic device, such as a cellular network module, a Bluetooth module, and / or a wireless local area network module, etc. The communication module 9110 (transmitter / receiver) is also coupled to the speaker 9131 and the microphone 9132 via the audio processor 9130 to provide an audio output via the speaker 9131 and receive an audio input from the microphone 9132, thereby implementing normal telecommunication functions. The audio processor 9130 may include any suitable buffers, decoders, amplifiers, etc. Additionally, the audio processor 9130 is also coupled to the central processor 9100, so that recording can be performed on the local machine through the microphone 9132, and the sound stored on the local machine can be played through the speaker 9131.

[0197] Embodiments of the present application also provide a computer-readable storage medium capable of implementing all steps of the automated production line digital model simplification method with the execution subject being a server or a client in the above embodiments. A computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, all steps of the automated production line digital model simplification method with the execution subject being a server or a client in the above embodiments are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0198] Step S101: Construct a saliency feature extraction module. The saliency feature extraction module calculates saliency scores for point, line, surface, and volume data in the automated production line digital model, extracts the feature points with saliency scores exceeding a first threshold as a set of salient feature points, and at the same time stores the morphological parameters and position parameters corresponding to the set of salient feature points in a salient feature database, and establishes a four-level topological structure tree based on the set of salient feature points;

[0199] Step S102: Input the four - level topological structure tree into the adaptive feature weight calculation module. The adaptive feature weight calculation module calculates the weight coefficient of each level based on the spatial distribution density of the significant feature point set, multiplies the weight coefficient by the feature data of the corresponding level to obtain weighted feature data, inputs the weighted feature data into the adaptive octree grid construction module. The adaptive octree grid construction module constructs grid nodes based on the weighted feature data, embeds multi - resolution wavelet basis functions into the grid nodes, performs hierarchical thresholding on the wavelet function coefficients, and reconstructs the thresholded wavelet coefficients into an initial simplified grid;

[0200] Step S103: Input the initial simplified grid into the multi - scale feature fusion module. The multi - scale feature fusion module extracts the topological structure features of the initial simplified grid at a coarse scale, extracts the local geometric features of the initial simplified grid at a fine scale, constructs a feature correspondence graph based on the topological structure features and the local geometric features, matches the feature correspondence graph with the morphological parameters and position parameters in the significant feature database, and adjusts the nodes of the initial simplified grid according to the matching result to generate a final simplified model.

[0201] As can be seen from the above description, the computer - readable storage medium provided by the embodiments of the present application identifies and saves key feature information through the significant feature extraction module, and establishes a four - level topological structure tree. An adaptive feature weight calculation mechanism is innovatively designed, and intelligent grid simplification is achieved by combining multi - resolution wavelet analysis. The system extracts and matches topological structure features and local geometric features at different scales through the multi - scale feature fusion module, and accurately preserves key features during the model simplification process. This method breaks through the limitations of traditional model simplification and provides an intelligent and automated solution for the efficient processing of large - scale digital models.

[0202] The embodiments of the present application also provide a computer program product that can implement all the steps of the automated production line digital model simplification method with the execution subject being a server or a client in the above embodiments. When the computer program / instructions are executed by a processor, the steps of the automated production line digital model simplification method are implemented. For example, the computer program / instructions implement the following steps:

[0203] Step S101: Construct a significant feature extraction module. The significant feature extraction module calculates the significant scores for the point, line, surface, and volume data of the automated production line digital model, extracts the feature points with the significant scores exceeding the first threshold as the significant feature point set, and at the same time stores the morphological parameters and position parameters corresponding to the significant feature point set into the significant feature database, and establishes a four - level topological structure tree based on the significant feature point set;

[0204] Step S102: Input the four-level topological structure tree into the adaptive feature weight calculation module. The adaptive feature weight calculation module calculates the weight coefficient of each level based on the spatial distribution density of the significant feature point set, multiplies the weight coefficient by the feature data of the corresponding level to obtain weighted feature data, inputs the weighted feature data into the adaptive octree grid construction module. The adaptive octree grid construction module constructs grid nodes based on the weighted feature data, embeds multi-resolution wavelet basis functions into the grid nodes, performs hierarchical thresholding on the wavelet function coefficients, and reconstructs the thresholded wavelet coefficients into an initial simplified grid;

[0205] Step S103: Input the initial simplified grid into the multi-scale feature fusion module. The multi-scale feature fusion module extracts the topological structure features of the initial simplified grid at a coarse scale, extracts the local geometric features of the initial simplified grid at a fine scale, constructs a feature correspondence graph based on the topological structure features and the local geometric features, matches the feature correspondence graph with the morphological parameters and position parameters in the significant feature database, and adjusts the nodes of the initial simplified grid according to the matching result to generate a final simplified model.

[0206] As can be seen from the above description, the computer program product provided by the embodiments of the present application identifies and saves key feature information through the significant feature extraction module, and establishes a four-level topological structure tree. An adaptive feature weight calculation mechanism is innovatively designed, and intelligent grid simplification is achieved by combining multi-resolution wavelet analysis. The system extracts and matches topological structure features and local geometric features at different scales through the multi-scale feature fusion module, and realizes the precise preservation of key features during the model simplification process. This method breaks through the limitations of traditional model simplification and provides an intelligent and automated solution for the efficient processing of large-scale digital models.

[0207] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, apparatus, or computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention 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.

[0208] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (devices), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices produce a means for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or a means for implementing the functions specified in multiple blocks.

[0209] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including an instruction means that implements the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or a means for implementing the functions specified in multiple blocks.

[0210] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or a means for implementing the functions specified in multiple blocks.

[0211] Specific embodiments are applied in the present invention to elaborate on the principles and implementation manners of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, based on the idea of the present invention, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A method for simplifying a digital model of an automated production line, characterized in that: The method comprises: Constructing a significant feature extraction module, the significant feature extraction module calculates the significance scores of the point, line, surface and volume data in the digital model of the automated production line, extracts the feature points whose significance scores exceed a first threshold as a significant feature point set, and stores the morphological parameters and position parameters corresponding to the significant feature point set in a significant feature database, and establishes a four-level topological structure tree based on the significant feature point set; Input the four-level topological structure tree into an adaptive feature weight calculation module, the adaptive feature weight calculation module calculates the weight coefficient of each level based on the spatial distribution density of the significant feature point set, multiplies the weight coefficient by the feature data of the corresponding level to obtain weighted feature data, inputs the weighted feature data into an adaptive octree grid construction module, the adaptive octree grid construction module constructs grid nodes based on the weighted feature data, embeds multi-resolution wavelet basis functions into the grid nodes, performs hierarchical thresholding on the wavelet basis function coefficients, and reconstructs the thresholded wavelet coefficients into an initial simplified grid; The initial simplified mesh is input into a multi-scale feature fusion module, which extracts the topological structure features of the initial simplified mesh at a coarse scale, extracts the local geometric features of the initial simplified mesh at a fine scale, constructs a feature correspondence graph based on the topological structure features and the local geometric features, matches the feature correspondence graph with the morphological parameters and position parameters in the significant feature database, adjusts the nodes of the initial simplified mesh according to the matching results, and generates a final simplified model.

2. The method for simplifying the digital model of an automated production line according to claim 1, characterized in that: The significant feature extraction module is constructed, and the significant feature extraction module calculates the significance scores of the points, lines, surfaces, and volumes in the digital model of the automated production line, and extracts the feature points whose significance scores exceed a first threshold as a significant feature point set, including: Discretize the digital model of the automated production line to obtain point cloud data, establish a three-dimensional grid structure based on the point cloud data, calculate the curvature value, normal vector, and neighborhood density of each vertex in the three-dimensional grid structure, input the curvature value, normal vector, and neighborhood density into a significance calculation formula to obtain a significance coefficient, and construct a significance feature extraction module based on the significance coefficient; The significance coefficient in the significant feature extraction module is compared with a first threshold value, vertices whose significance coefficient is greater than the first threshold value are extracted as feature points, a significant feature point set is constructed based on the spatial position relationship of the feature points, and the geometric attribute parameters and topological attribute parameters of the significant feature point set are stored as morphological parameters and position parameters, respectively.

3. The method for simplifying the digital model of an automated production line according to claim 1, characterized in that: The step of storing the morphological parameters and position parameters corresponding to the salient feature point set in a salient feature database and establishing a four-level topological structure tree based on the salient feature point set includes: Inputting the morphological parameters and position parameters of the salient feature point set into a feature encoder, the feature encoder performs geometric feature encoding on the morphological parameters to obtain a geometric feature vector, performs spatial position encoding on the position parameters to obtain a position feature vector, combines the geometric feature vector and the position feature vector to form feature data, and writes the feature data into a salient feature database according to a hierarchical structure; A four-level topological structure tree is constructed based on the set of significant feature points, with the point data in the set of significant feature points as first-level nodes, the connection relationship between adjacent points as second-level nodes, the faces surrounded by closed line segments as third-level nodes, and the volume elements composed of closed faces as fourth-level nodes, and the corresponding feature data index is stored in each node of the four-level topological structure tree.

4. The method for simplifying the digital model of an automated production line according to claim 1, characterized in that: The four-level topological structure tree is input into an adaptive feature weight calculation module, the adaptive feature weight calculation module calculates a weight coefficient of each level based on the spatial distribution density of the significant feature point set, and multiplies the weight coefficient by the feature data of the corresponding level to obtain weighted feature data, including: Constructing a spatial density distribution map based on the four-level topological structure tree, calculating the number of significant feature points within the neighborhood of each node in the spatial density distribution map, dividing the number by the neighborhood volume to obtain a local density value, normalizing the local density value to obtain a density distribution coefficient, and inputting the density distribution coefficient into a weight mapping function to generate a weight coefficient corresponding to each level; The weight coefficient is subjected to matrix multiplication operation with the feature data of each level node in the four-level topological structure tree to generate a weighted feature matrix, the weighted feature matrix is ​​standardized to obtain weighted feature data, and the weighted feature data is reorganized into a tree structure according to the hierarchical relationship.

5. The method for simplifying the digital model of an automated production line according to claim 1, characterized in that: The step of inputting the weighted feature data into an adaptive octree grid construction module, constructing grid nodes based on the weighted feature data, embedding multi-resolution wavelet basis functions into the grid nodes, performing hierarchical thresholding on the wavelet basis function coefficients, and reconstructing the thresholded wavelet coefficients into an initial simplified grid comprises: Determine a spatial bounding box based on the weighted feature data, recursively divide the spatial bounding box into an octree structure, calculate the local variance of the weighted feature data at each node of the octree structure, continue to subdivide the node when the local variance of the node is greater than a second threshold, until a preset number of subdivision levels is reached or the local variance is less than the second threshold, construct an adaptive grid on the subdivided octree node, and map the Haar wavelet basis function to the node coordinates of the adaptive grid; The Haar wavelet basis functions on the adaptive grid nodes are coefficient decomposed, the energy value of each layer of wavelet coefficients is calculated, an adaptive threshold is set according to the energy value, the wavelet coefficients smaller than the adaptive threshold are set to zero, the retained wavelet coefficients are reconstructed to generate an initial simplified grid, and a corresponding relationship is established between the nodes of the initial simplified grid and the original weighted feature data.

6. The method for simplifying the digital model of an automated production line according to claim 1, characterized in that: The step of inputting the initial simplified grid into a multi-scale feature fusion module, wherein the multi-scale feature fusion module extracts the topological structure features of the initial simplified grid at a coarse scale and extracts the local geometric features of the initial simplified grid at a fine scale, comprises: Performing multi-level decomposition on the initial simplified grid to obtain grid representations of different scales, constructing a grid connectivity graph at a coarse-scale level, calculating the degree distribution and connectivity properties of each node in the grid connectivity graph, constructing a grid skeleton based on the degree distribution and connectivity properties, and extracting the branch structure and connection relationship of the grid skeleton as topological structure features; The local curvature tensor of the initial simplified grid node is calculated at a fine-scale level, the principal curvature value and the principal curvature direction are calculated based on the curvature tensor, the principal curvature value and the principal curvature direction are combined to form a curvature descriptor, a cluster analysis is performed on the curvature descriptor to obtain local geometric features, and a mapping relationship is established between the local geometric features and the grid nodes.

7. The method for simplifying the digital model of an automated production line according to claim 1, characterized in that: The step of constructing a feature correspondence diagram based on the topological structure features and the local geometric features, matching the feature correspondence diagram with the morphological parameters and position parameters in the significant feature database, and adjusting the nodes of the initial simplified mesh according to the matching results to generate a final simplified model includes: The topological structure features and the local geometric features are input into a feature fusion network, the feature fusion network generates a feature correspondence graph based on a graph convolution operation, calculates a similarity matrix between nodes in the feature correspondence graph, performs a matching operation on the similarity matrix with morphological parameters and position parameters stored in a significant feature database, generates a feature matching metric value, and constructs a node optimization objective function based on the feature matching metric value; The node optimization objective function is iteratively solved, the node positions and connection relationships of the initial simplified mesh are adjusted, and the iteration is stopped when the objective function value is less than a third threshold or the maximum number of iterations is reached, the final mesh node coordinates and connection relationships are reconstructed into a three-dimensional model, and the reconstructed three-dimensional model is meshed and smoothed to obtain the final simplified model.

8. A digital model simplification device for an automated production line, characterized in that: The device comprises: A feature extraction module is used to construct a significant feature extraction module, wherein the significant feature extraction module calculates the significance scores of the points, lines, surfaces and volumes in the digital model of the automated production line, extracts the feature points whose significance scores exceed a first threshold as a significant feature point set, and stores the morphological parameters and position parameters corresponding to the significant feature point set in a significant feature database, and establishes a four-level topological structure tree based on the significant feature point set; An architecture building module, used for inputting the four-level topological structure tree into an adaptive feature weight calculation module, the adaptive feature weight calculation module calculates the weight coefficient of each level based on the spatial distribution density of the significant feature point set, multiplies the weight coefficient by the feature data of the corresponding level to obtain weighted feature data, inputs the weighted feature data into an adaptive octree grid construction module, the adaptive octree grid construction module constructs grid nodes based on the weighted feature data, embeds multi-resolution wavelet basis functions into the grid nodes, performs hierarchical thresholding on the wavelet basis function coefficients, and reconstructs the thresholded wavelet coefficients into an initial simplified grid; A model simplification module is used to input the initial simplified grid into a multi-scale feature fusion module, the multi-scale feature fusion module extracts the topological structure features of the initial simplified grid at a coarse scale, extracts the local geometric features of the initial simplified grid at a fine scale, constructs a feature correspondence diagram based on the topological structure features and the local geometric features, matches the feature correspondence diagram with the morphological parameters and position parameters in the significant feature database, adjusts the nodes of the initial simplified grid according to the matching results, and generates a final simplified model.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the method for simplifying the digital model of an automated production line described in any one of claims 1 to 7 are implemented.

10. 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 method for simplifying the digital model of an automated production line described in any one of claims 1 to 7 are implemented.

Images