Exhibit touch model rapid printing method, device and equipment and storage medium

Abstract

The invention relates to the field of data processing, and discloses a rapid printing method and device for an exhibit touch model, equipment and a storage medium. A to-be-printed object is scanned through 3D structured light, and object point cloud information is aligned through a feature matching-based point cloud alignment algorithm; removing redundant details in the complete article model data by using a Gaussian filtering algorithm; based on a Canny edge detection algorithm and in combination with an adaptive threshold adjustment method, geometric feature enhancement is carried out on the filtered article model data, and the intensity value of edge pixels is adjusted according to the importance of the edge and the required sharpening degree; and according to the structural strength requirement, the topological relation of the internal structure of the model is adjusted to generate a supporting structure, and article layer-by-layer printing is carried out based on sharpened article model data and article model structure data. The problems of inaccuracy and incompleteness of scanning data are effectively solved, the geometric accuracy and completeness of the model are improved, and the finally printed article model is closer to a real article.

Description

Technical Field

[0001] The present application relates to the field of 3D printing technology, and in particular to a method, device, equipment and storage medium for rapid printing of tactile models of exhibits. Background Art

[0002] As society pays more attention to the needs of special groups and the exhibition industry develops, it is becoming increasingly important to provide tactile models for the visually impaired or other people who need to understand exhibits through touch. However, there are many problems with the existing tactile model manufacturing technology for exhibits. The traditional tactile model manufacturing process usually includes three-dimensional data acquisition, model simplification and optimization, adding auxiliary elements, material selection, CNC processing or mold die casting. The production process is complicated, the process is numerous, and the cost of correction after one-time molding is very high.

[0003] At present, conventional 3D printers usually need to be printed by high-speed computer-controlled inkjet printers. Before use, various parameters need to be set, such as material type, quantity, thickness, position, etc., and then the printing material is loaded into the printer to form a three-dimensional object. This direct 3D printing method has obvious shortcomings in the production of tactile models of exhibition scenes. On the one hand, it is difficult to obtain the original three-dimensional data of the exhibits, and a single type of 3D structured light scanner cannot quickly provide accurate three-dimensional data to the 3D printer; on the other hand, 3D prints usually produce the entire three-dimensional object, but many exhibition scenes only need to show one side, such as a tactile model of a person's facial features, without printing the back of the head, neck and other parts; in addition, the conventional 3D printing process cannot produce samples on site in a short time. It is necessary to use a special structure scanner to obtain 3D data and then go back to process the data, adjust and correct it, and then select 3D printing materials. The whole process takes a long time and cannot meet the needs of rapid updates and iterations of exhibits. Summary of the invention

[0004] The present application provides a method for quickly printing a tactile model of an exhibit, which can store asset registration in a database and audit asset change request information through a multiple audit mechanism, thereby ensuring the security of the assets.

[0005] In a first aspect, an embodiment of the present application provides a method for quickly printing a tactile model of an exhibit, the method comprising the following steps:

[0006] Scanning the object to be printed using 3D structured light to obtain object point cloud information, aligning the object point cloud information using a point cloud alignment algorithm based on feature matching to obtain printed object model data;

[0007] Processing the printed object model data using an outlier detection algorithm, deleting outliers, and obtaining initial object model data;

[0008] Intelligently fill the holes and missing data in the initial object model data during scanning according to the surrounding geometric features to obtain complete object model data;

[0009] Using a Gaussian filtering algorithm to remove redundant details in the complete object model data to obtain filtered object model data;

[0010] Based on the Canny edge detection algorithm combined with the adaptive threshold adjustment method, the geometric feature enhancement of the filtered object model data is performed, and the intensity value of the edge pixel is adjusted according to the importance of the edge and the required sharpening degree to obtain the sharpened object model data;

[0011] The topological relationship of the internal structure of the model is adjusted according to the structural strength requirements to generate a support structure, and the support position and structure are calculated according to the model shape and the physical properties of the material to obtain the object model structure data; the object is printed layer by layer based on the sharpened object model data and the object model structure data.

[0012] In a possible implementation, the method of scanning the object to be printed using 3D structured light to obtain object point cloud information, aligning the object point cloud information based on a point cloud alignment algorithm based on feature matching, and obtaining printed object model data includes:

[0013] Use 3D structured light scanning equipment to scan the object to be printed, project the coded pattern onto the surface of the object to be printed through a projector, use a binocular camera to shoot the light pattern reflected from the surface of the object from different angles, calculate the coordinates of each camera pixel in three-dimensional space, and obtain the object point cloud information P cloud ={p1,p2,···,p n}; where p i Represents the three-dimensional coordinates of the i-th point;

[0014] Obtain each point cloud data point in the object point cloud information, calculate it through the FPFH feature descriptor, and calculate all points in the point cloud data to p i The Euclidean distance of the nearest k points is selected as N(p i );

[0015] For N(p i ) for each point p i , calculate its difference with p i The relative position vector v ij =p j -p i , by N(p i ) to obtain point p i The normal vector n i , for each relative position vector vij Calculate its normal vector n i The projection length μ ij =v ij -n i , and its discovery vector n i The vertical vector t ij =v ij -μ ij n i ; will (t ij ,n i ) as a feature vector, and perform histogram statistics on the feature vectors calculated for all k nearest neighbor points to obtain point p i The FPFH feature descriptor F(p i );

[0016] By comparing the feature descriptors of two point cloud data from different viewpoints, we find corresponding points, calculate the Euclidean distance between the corresponding points, and use the least squares method to calculate the rotation and translation transformation matrix to align the two point cloud data. Let p i1 and p j2 is a pair of corresponding points, p i1 =(X i1 ,Y i1 ,Z i1 ), p j2 =(X j2 ,Y j2 ,Z j2 ), establish the error function:

[0017]

[0018] Where R is the rotation matrix, t is the translation vector, (X i1 ,Y i1 ,Z i1 ) and (X j2 ,Y j2 ,Z j2 ) is the three-dimensional coordinate of the corresponding point; the object point cloud information P cloud All points in are transformed to obtain the printed object model data.

[0019] In a possible implementation, the using of an outlier detection algorithm to process the printed object model data, deleting outliers, and obtaining initial object model data includes:

[0020] The printed object model data is processed using an outlier detection algorithm based on statistical analysis, and the K-nearest neighbor algorithm is used to obtain the outlier detection algorithm of each point p in the printed object model data. i Neighborhood points of distance p i The nearest K points are taken as its neighborhood point set;

[0021] Calculate point p i The statistical characteristics of the distance between the points and its neighboring points are used to set an abnormal threshold through the 3σ principle, and the abnormal point is judged by using the abnormal threshold;

[0022] All points in the printed object model data are traversed, and data determined to be abnormal points are deleted from the data set to obtain initial object model data.

[0023] In a possible implementation, the intelligently filling holes and missing data in the initial object model data that appear during scanning according to surrounding geometric features to obtain complete object model data includes:

[0024] Calculating the normal vector of each point in the initial object model data based on a normal vector analysis method, and identifying holes and missing data areas in the initial object model data through the normal vector;

[0025] The holes and missing data areas are filled by using a region growing algorithm. i , initialize a growing queue Q = {p i}, and create a visited point set V = {p i}, when the growth queue Q is not empty;

[0026] Take out the point p at the head of the queue c , using the k-nearest neighbor algorithm to calculate the point p c and point p c Unvisited neighboring point p n The Euclidean distance between them is , and the points whose distance is less than the threshold are selected as neighborhood points;

[0027] Calculate the neighborhood point p n With the current point p c The normal vector angle θ cn , let p c The normal vector is n c , p n The normal vector is n n , then θ cn =arccos(n c ·n n ), if θ cn If it is smaller than the threshold, it means that the normal vector directions are close;

[0028] Calculate the neighborhood point p n The average distance to the filled area d n , let the point set of the filled area be p filled ,but:

[0029]

[0030] Where X n , X f , Y n , Y f , Z n , Z f is the three-dimensional coordinate of the neighborhood point, if d n,filled If it is less than the threshold, the surface neighborhood points are close to the filled area;

[0031] If the neighboring point p n If the above conditions of normal vector angle and distance to the filled area are met, it will be added to the growth queue Q and the visited point set V, and used as a new filling point to update the geometric feature information of the filled area, and repeated iterations will be performed to obtain the complete object model data.

[0032] In a possible implementation, removing redundant details from the complete object model data using a Gaussian filtering algorithm to obtain filtered object model data includes:

[0033] Using a Gaussian filtering algorithm to remove redundant details in the complete object model data, and determining neighborhood points in the complete object model data;

[0034] The weight of each neighborhood point is calculated based on a three-dimensional Gaussian function, the filtering degree is controlled according to the weight, and the new data value of each data point in the complete object model data is calculated according to the filtering degree; and the filtered object model data is obtained.

[0035] In a possible implementation, the geometric feature enhancement of the filtered object model data is performed based on the Canny edge detection algorithm combined with the adaptive threshold adjustment method, and the intensity value of the edge pixel is adjusted according to the importance of the edge and the required sharpening degree to obtain the sharpened object model data, including:

[0036] Performing geometric feature enhancement on the filtered object model data based on a Canny edge detection algorithm combined with an adaptive threshold adjustment method;

[0037] Calculate the partial derivatives of the data points in the filtered object model data in the X, Y, and Z directions using the central difference method to obtain the gradient amplitude of the model data;

[0038] Compare the gradient magnitude of each point with the gradient magnitude of its neighboring points along the gradient direction to obtain edge point data and non-edge point data;

[0039] Calculate the average value and standard deviation of the gradient amplitude of all points in the area, determine the statistical characteristics of the local area based on the average value and standard deviation, obtain a high threshold and a low threshold, mark the points with a gradient amplitude greater than the high threshold as edge points, and mark the points with a gradient amplitude less than the low threshold as non-edge points; connect the edge points to obtain a preliminary edge detection result;

[0040] The pixel intensity value is adjusted according to the importance of the edge and the required sharpening degree to obtain the sharpened object model data.

[0041] In a possible implementation, the topological relationship of the internal structure of the model is adjusted according to the structural strength requirement to generate a support structure, the support position and structure are calculated according to the model shape and the physical properties of the material, and the object model structure data is obtained; and the object is printed layer by layer based on the sharpened object model data and the object model structure data, including:

[0042] Divide the model into multiple small units according to its geometric shape, determine the node coordinates of each unit and the connection relationship between the nodes; input the elastic modulus, Poisson's ratio, density and thermal expansion coefficient of the material according to the material properties used;

[0043] The allowable stress is set, and the density-based topology optimization algorithm obtains the optimal topology structure that meets the structural strength requirements by iteratively adjusting the material distribution of the unit. The material density of the unit is updated according to the criteria of the topology optimization algorithm to obtain the optimized model topology structure.

[0044] Identify the cantilever structure, thin-wall structure and hole structure of the model, calculate the curvature radius, wall thickness and angle of different parts of the model, and obtain the structural data of the object model;

[0045] The object model structure data and the sharpened object model data are converted into G-CODE format data, and the model is sliced ​​reasonably according to the size of the model and the size of the printing area of ​​the printer; the printing temperature, material flow rate and printing speed are controlled to print the model layer by layer.

[0046] In a second aspect, the present application provides a device for quickly printing a tactile model of an exhibit, the device comprising the following modules:

[0047] An object information scanning module is used to scan the object to be printed using 3D structured light to obtain object point cloud information, and align the object point cloud information based on a point cloud alignment algorithm based on feature matching to obtain printed object model data;

[0048] An abnormal data deletion module is used to process the printed object model data using an abnormal point detection algorithm, delete abnormal points, and obtain initial object model data;

[0049] A missing data filling module is used to intelligently fill in the holes and missing data in the initial object model data during scanning according to the surrounding geometric features to obtain complete object model data;

[0050] A print data adjustment module, used to remove redundant details in the complete object model data by using a Gaussian filtering algorithm to obtain filtered object model data;

[0051] An object internal adjustment module is used to enhance the geometric features of the filtered object model data based on the Canny edge detection algorithm combined with the adaptive threshold adjustment method, and adjust the intensity value of the edge pixel according to the importance of the edge and the required sharpening degree to obtain the sharpened object model data;

[0052] The object 3D printing module is used to adjust the topological relationship of the internal structure of the model according to the structural strength requirements to generate a support structure, calculate the support position and structure according to the model shape and the physical properties of the material, and obtain the object model structure data; based on the sharpened object model data and the object model structure data, the object is printed layer by layer.

[0053] In a third aspect, the present application provides a device for quickly printing tactile models of exhibits, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the device for quickly printing tactile models of exhibits executes each step of the above-mentioned method for quickly printing tactile models of exhibits.

[0054] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the various steps of the above-mentioned method for quickly printing a tactile model of an exhibit.

[0055] Based on the method provided in the present application, by using 3D structured light to scan the object to be printed, the object point cloud information is obtained, and the object point cloud information is aligned based on a point cloud alignment algorithm with feature matching to obtain printed object model data; the printed object model data is processed using an outlier detection algorithm, and the outliers are deleted to obtain initial object model data; the holes and missing data that appear in the initial object model data during scanning are intelligently filled according to the surrounding geometric features to obtain complete object model data; the Gaussian filtering algorithm is used to remove redundant details in the complete object model data to obtain filtered object model data; the geometric features of the filtered object model data are enhanced based on the Canny edge detection algorithm combined with the adaptive threshold adjustment method, and the intensity value of the edge pixel is adjusted according to the importance of the edge and the required sharpening degree to obtain sharpened object model data; the topological relationship of the internal structure of the model is adjusted according to the structural strength requirements to generate a support structure, and the support position and structure are calculated according to the model shape and the physical properties of the material to obtain the object model structure data; the object is printed layer by layer based on the sharpened object model data and the object model structure data. 1. Improve model accuracy: Through 3D structured light scanning and a series of data processing algorithms, such as point cloud alignment, outlier detection, hole filling, etc., the problem of inaccurate and incomplete scan data is effectively solved, the geometric accuracy and integrity of the model are improved, and the final printed object model is closer to the real object. 2. Optimize model quality: Gaussian filtering removes redundant details, Canny edge detection and adaptive threshold adjustment enhance geometric features, improve the smoothness of the model surface and edge clarity, improve the model quality, and make it more in line with actual use needs. 3. Enhance structural rationality: Adjust the topological relationship and generate support structure according to the structural strength requirements, and calculate the support position and structure considering the model shape and material properties, which improves the stability and success rate of the printing process and ensures that the printed object model has good structural strength and stability. 4. Improve printing efficiency: Accurate model data and reasonable support structure design reduce errors and rework during the printing process, improve printing efficiency, and save time and material costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 This is a flow chart of an embodiment of a first method for quickly printing a tactile model of an exhibit provided in an embodiment of the present application;

[0057] Figure 2 This is a flow chart of an embodiment of a second method for quickly printing a tactile model of an exhibit provided in an embodiment of the present application;

[0058] Figure 3 This is a flow chart of an embodiment of a third method for quickly printing a tactile model of an exhibit provided in an embodiment of the present application;

[0059] Figure 4A schematic diagram of the structure of a device for quickly printing a tactile model of an exhibit provided in an embodiment of the present application;

[0060] Figure 5 A schematic diagram of the structure of a device for quickly printing a tactile model of an exhibit provided in an embodiment of the present application. DETAILED DESCRIPTION

[0061] The embodiment of the present application provides a method for quickly printing a tactile model of an exhibit, which can store asset registration in a database and audit asset change request information through a multiple audit mechanism, thereby solving the problem of poor security in daily asset management.

[0062] The technical scheme in the embodiment of the present application will be described below in conjunction with the accompanying drawings in the embodiment of the present application. Wherein, the terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable in appropriate circumstances, so that the embodiments described here can be implemented in a sequence other than the content illustrated or described here. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, the process, method, system, product or equipment comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0063] It is understandable that any part of the data acquisition or collection involved in this application has been authorized by the user.

[0064] It is understandable that the execution subject of the present application may be a device for quickly printing a tactile model of an exhibit, or may be a mobile terminal or a server, which is not specifically limited here.

[0065] The specific process of the embodiment of the present application is described below. Figure 1 , Figure 1 The flowchart of an embodiment of a method for quickly printing a tactile model of an exhibit provided in an embodiment of the present application includes:

[0066] 101. Scan the object to be printed using 3D structured light to obtain object point cloud information, align the object point cloud information based on feature matching point cloud alignment algorithm, and obtain printed object model data;

[0067] It can be understood that the object to be printed is scanned by using a 3D structured light scanning device, the coded pattern is projected onto the surface of the object to be printed by a projector, the light pattern reflected from the surface of the object is photographed from different angles by a binocular camera, and the coordinates of each camera pixel in three-dimensional space are calculated to obtain the object point cloud information P cloud ={p1,p2,···,p n}; where p i Represents the three-dimensional coordinates of the i-th point;

[0068] Get each point cloud data point in the object point cloud information, calculate it through the FPFH feature descriptor, and calculate all points in the point cloud data to p i The Euclidean distance of the nearest k points is selected as N(p i );

[0069] For N(p i ) for each point p i , calculate its difference with p i The relative position vector v ij =p j -p i , by N(p i ) to obtain point p i The normal vector n i , for each relative position vector v ij Calculate its normal vector n i The projection length μ ij =v ij -n i , and its discovery vector n i The vertical vector t ij =v ij -μ ij n i ; will (t ij ,n i ) as a feature vector, and perform histogram statistics on the feature vectors calculated for all k nearest neighbor points to obtain point p i The FPFH feature descriptor F(p i );

[0070] By comparing the feature descriptors of two point cloud data from different viewpoints, we find corresponding points, calculate the Euclidean distance between the corresponding points, and use the least squares method to calculate the rotation and translation transformation matrix to align the two point cloud data. Let p i1 and p j2 is a pair of corresponding points, p i1 =(X i1 ,Y i1 ,Z i1 ), p j2 =(Xj2 ,Y j2 ,Z j2 ), establish the error function:

[0071]

[0072] Where R is the rotation matrix, t is the translation vector, (X i1 ,Y i1 ,Z i1 ) and (X j2 ,Y j2 ,Z j2 ) is the three-dimensional coordinate of the corresponding point; the object point cloud information P cloud All points in are transformed to obtain the printed object model data.

[0073] 102. Process the printed object model data using an outlier detection algorithm, delete the outliers, and obtain initial object model data;

[0074] It can be understood that the outlier detection algorithm based on statistical analysis is used to process the printed object model data, and the K-nearest neighbor algorithm is used to obtain the outlier detection algorithm of each point p in the printed object model data. i Neighborhood points of distance p i The nearest K points are taken as its neighborhood point set;

[0075] Calculate point p i The statistical characteristics of the distance to its neighboring points are used to set the abnormal threshold through the 3σ principle, and the abnormal point is judged by the abnormal threshold;

[0076] All points in the printed object model data are traversed, and the data determined as abnormal points are deleted from the data set to obtain the initial object model data.

[0077] 103. Intelligently fill in the holes and missing data in the initial object model data during scanning according to the surrounding geometric features to obtain complete object model data;

[0078] It can be understood that the normal vector of each point in the initial object model data is calculated based on the normal vector analysis method, and the holes and missing data areas in the initial object model data are identified through the normal vector;

[0079] The region growing algorithm is used to fill the holes and missing data areas. i , initialize a growing queue Q = {p i}, and create a visited point set V = {p i}, when the growth queue Q is not empty;

[0080] Take out the point p at the head of the queue c, using the k-nearest neighbor algorithm to calculate point p c and point p c Unvisited neighboring point p n The Euclidean distance between them is , and the points whose distance is less than the threshold are selected as the neighborhood points;

[0081] It is understandable that the threshold is 0.5-1;

[0082] Calculate the neighborhood point p n With the current point p c The normal vector angle θ cn , let p c The normal vector is n c , p n The normal vector is n n , then θ cn =arccos(n c ·n n ), if θ cn If it is smaller than the threshold, it means that the normal vector directions are close;

[0083] Calculate the neighborhood point p n The average distance to the filled area d n , let the point set of the filled area be p filled ,but:

[0084]

[0085] Where X n , X f , X n , Y f , Z n , Z f is the three-dimensional coordinate of the neighborhood point, if d n,filled If it is less than the threshold, the surface neighborhood points are close to the filled area;

[0086] If the neighboring point p n If the above conditions of normal vector angle and distance to the filled area are met, it will be added to the growth queue Q and the visited point set V, and used as a new filling point to update the geometric feature information of the filled area, and repeated iterations will be performed to obtain the complete object model data.

[0087] 104. Using a Gaussian filtering algorithm to remove redundant details in the complete object model data, to obtain filtered object model data;

[0088] It can be understood that the Gaussian filtering algorithm is used to remove redundant details in the complete object model data and determine the neighborhood points in the complete object model data;

[0089] The weight of each neighborhood point is calculated based on the three-dimensional Gaussian function, the filtering degree is controlled according to the weight, and the new data value of each data point in the complete object model data is calculated according to the filtering degree; and the filtered object model data is obtained.

[0090] 105. Based on the Canny edge detection algorithm combined with the adaptive threshold adjustment method, the geometric features of the filtered object model data are enhanced, and the intensity value of the edge pixel is adjusted according to the importance of the edge and the required sharpening degree to obtain the sharpened object model data;

[0091] It can be understood that the geometric features of the filtered object model data are enhanced based on the Canny edge detection algorithm combined with the adaptive threshold adjustment method;

[0092] The partial derivatives of the data points in the filter item model data in the X, Y, and Z directions are calculated using the central difference method to obtain the gradient amplitude of the model data;

[0093] Compare the gradient magnitude of each point with the gradient magnitude of its neighboring points along the gradient direction to obtain edge point data and non-edge point data;

[0094] Calculate the average value and standard deviation of the gradient amplitude of all points in the area, determine the statistical characteristics of the local area based on the average value and standard deviation, obtain the high threshold and the low threshold, mark the points with gradient amplitude greater than the high threshold as edge points, and mark the points with gradient amplitude less than the low threshold as non-edge points; connect the edge points to obtain the preliminary edge detection results;

[0095] It is understandable that the high threshold is 30-50 and the low threshold is 5-10;

[0096] The pixel intensity value is adjusted according to the importance of the edge and the required sharpening degree to obtain the sharpened object model data.

[0097] 106. According to the structural strength requirements, the topological relationship of the internal structure of the model is adjusted to generate a support structure, and the support position and structure are calculated according to the model shape and the physical properties of the material to obtain the object model structure data; based on the sharpened object model data and the object model structure data, the object is printed layer by layer.

[0098] It can be understood that the model is divided into multiple small units according to the geometric shape of the model, and the node coordinates of each unit and the connection relationship between the nodes are determined; according to the characteristics of the material used, the elastic modulus, Poisson's ratio, density and thermal expansion coefficient of the material are input;

[0099] The allowable stress is set, and the density-based topology optimization algorithm obtains the optimal topology structure that meets the structural strength requirements by iteratively adjusting the material distribution of the unit. The material density of the unit is updated according to the criteria of the topology optimization algorithm to obtain the optimized model topology structure.

[0100] Identify the cantilever structure, thin-wall structure and hole structure of the model, calculate the curvature radius, wall thickness and angle of different parts of the model, and obtain the structural data of the object model;

[0101] The object model structure data and the sharpened object model data are converted into G-CODE format data, and the model is sliced ​​reasonably according to the size of the model and the size of the printing area of ​​the printer; the printing temperature, material flow and printing speed are controlled to print the model layer by layer.

[0102] Based on the method provided in the present application, by using 3D structured light to scan the object to be printed, the object point cloud information is obtained, and the object point cloud information is aligned based on a point cloud alignment algorithm with feature matching to obtain printed object model data; the printed object model data is processed using an outlier detection algorithm, and the outliers are deleted to obtain initial object model data; the holes and missing data that appear in the initial object model data during scanning are intelligently filled according to the surrounding geometric features to obtain complete object model data; the Gaussian filtering algorithm is used to remove redundant details in the complete object model data to obtain filtered object model data; the geometric features of the filtered object model data are enhanced based on the Canny edge detection algorithm combined with the adaptive threshold adjustment method, and the intensity value of the edge pixel is adjusted according to the importance of the edge and the required sharpening degree to obtain sharpened object model data; the topological relationship of the internal structure of the model is adjusted according to the structural strength requirements to generate a support structure, and the support position and structure are calculated according to the model shape and the physical properties of the material to obtain the object model structure data; the object is printed layer by layer based on the sharpened object model data and the object model structure data. 1. Improve model accuracy: Through 3D structured light scanning and a series of data processing algorithms, such as point cloud alignment, outlier detection, hole filling, etc., the problem of inaccurate and incomplete scan data is effectively solved, the geometric accuracy and integrity of the model are improved, and the final printed object model is closer to the real object. 2. Optimize model quality: Gaussian filtering removes redundant details, Canny edge detection and adaptive threshold adjustment enhance geometric features, improve the smoothness of the model surface and edge clarity, improve the model quality, and make it more in line with actual use needs. 3. Enhance structural rationality: Adjust the topological relationship and generate support structure according to the structural strength requirements, and calculate the support position and structure considering the model shape and material properties, which improves the stability and success rate of the printing process and ensures that the printed object model has good structural strength and stability. 4. Improve printing efficiency: Accurate model data and reasonable support structure design reduce errors and rework during the printing process, improve printing efficiency, and save time and material costs.

[0103] See also Figure 2 , Figure 2The flowchart of the third method for quickly printing a tactile model of an exhibit provided in the embodiments of the present application includes:

[0104] 201. Based on the Canny edge detection algorithm combined with the adaptive threshold adjustment method, the geometric features of the filtered object model data are enhanced;

[0105] 202. Calculate the partial derivatives of the data points in the filter item model data in the X, Y, and Z directions using the central difference method to obtain the gradient amplitude of the model data;

[0106] 203. Compare the gradient magnitude of each point with the gradient magnitude of its neighboring points along the gradient direction to obtain edge point data and non-edge point data;

[0107] 204. Calculate the average value and standard deviation of the gradient amplitude of all points in the region, determine the statistical characteristics of the local region according to the average value and the standard deviation, obtain a high threshold and a low threshold, mark the points with a gradient amplitude greater than the high threshold as edge points, and mark the points with a gradient amplitude less than the low threshold as non-edge points; connect the edge points to obtain a preliminary edge detection result;

[0108] 205. Adjust the pixel intensity value according to the importance of the edge and the required sharpening degree to obtain the sharpened object model data.

[0109] Based on the method provided in the embodiment of the present application, the geometric features of the filtered object model data are enhanced by combining the Canny edge detection algorithm with the adaptive threshold adjustment method, and the intensity value of the edge pixel is adjusted according to the importance of the edge and the required sharpening degree to obtain the sharpened object model data. It can accurately identify and strengthen the edges, making the model edges clearer and the details more prominent. By adjusting the intensity value according to the importance of the edge and the degree of sharpness, key features can be highlighted, the model recognition can be improved, and it is easy to observe and use. Geometric feature enhancement improves the overall quality of the model, making it more in line with actual needs, and can better reflect the characteristics of the object in subsequent applications.

[0110] See also Figure 3 , Figure 3 The flowchart of the third method for quickly printing a tactile model of an exhibit provided in the embodiments of the present application includes:

[0111] 301. Divide the model into multiple small units according to the geometric shape of the model, determine the node coordinates of each unit and the connection relationship between the nodes; input the elastic modulus, Poisson's ratio, density and thermal expansion coefficient of the material according to the material properties used;

[0112] 302. Set the allowable stress, and the density-based topology optimization algorithm obtains the optimal topology structure that meets the structural strength requirements by iteratively adjusting the material distribution of the unit, and updates the material density of the unit according to the criteria of the topology optimization algorithm to obtain the optimized model topology structure;

[0113] 303. Identify the cantilever structure, thin-wall structure and hole structure of the model, calculate the curvature radius, wall thickness and angle of different parts of the model, and obtain the object model structure data;

[0114] 304. Convert the object model structure data and the sharpened object model data into G-CODE format data, and slice the model reasonably according to the size of the model and the size of the printing area of ​​the printer; control the printing temperature, material flow rate and printing speed to print the model layer by layer.

[0115] Based on the method provided in the embodiment of the present application, the topological relationship of the internal structure of the model is adjusted according to the structural strength requirements to generate a support structure, the support position and structure are calculated according to the physical properties of the model shape and the material, and the object model structure data is obtained; the object is printed layer by layer based on the sharpened object model data and the object model structure data. Adjusting the topological relationship and generating the support structure according to the structural strength requirements can optimize the stress distribution, make the model better able to withstand stress, and reduce the risk of deformation and damage. The support position structure is calculated based on the shape and material properties to ensure reasonable support and improve structural stability. The support structure ensures printing continuity, improves the success rate, and reduces material and time waste.

[0116] The above describes the method for quickly printing the tactile model of an exhibit in the embodiment of the present application. The following describes the device for quickly printing the tactile model of an exhibit in the embodiment of the present application. Figure 4 , Figure 4 A schematic diagram of a device for quickly printing a tactile model of an exhibit provided in an embodiment of the present application includes:

[0117] The object information scanning module 401 is used to scan the object to be printed using 3D structured light to obtain object point cloud information, and align the object point cloud information based on a point cloud alignment algorithm based on feature matching to obtain printed object model data;

[0118] The abnormal data deletion module 402 is used to process the printed object model data using an abnormal point detection algorithm, delete the abnormal points, and obtain the initial object model data;

[0119] A missing data filling module 403 is used to intelligently fill in the holes and missing data in the initial object model data during scanning according to the surrounding geometric features to obtain complete object model data;

[0120] The print data adjustment module 404 is used to remove redundant details in the complete object model data using a Gaussian filtering algorithm to obtain filtered object model data;

[0121] The object internal adjustment module 405 is used to enhance the geometric features of the filtered object model data based on the Canny edge detection algorithm combined with the adaptive threshold adjustment method, and adjust the intensity value of the edge pixel according to the importance of the edge and the required sharpening degree to obtain the sharpened object model data;

[0122] The object 3D printing module 406 is used to adjust the topological relationship of the internal structure of the model according to the structural strength requirements to generate a support structure, calculate the support position and structure according to the model shape and the physical properties of the material, and obtain the object model structure data; and print the object layer by layer based on the sharpened object model data and the object model structure data.

[0123] Based on the device provided by the present application, by using 3D structured light to scan the object to be printed, the object point cloud information is obtained, and the object point cloud information is aligned based on the point cloud alignment algorithm of feature matching to obtain the printed object model data; the printed object model data is processed by using the outlier detection algorithm, and the outliers are deleted to obtain the initial object model data; the holes and missing data that appear in the initial object model data during the scan are intelligently filled according to the surrounding geometric features to obtain the complete object model data; the Gaussian filtering algorithm is used to remove the redundant details in the complete object model data to obtain the filtered object model data; the geometric features of the filtered object model data are enhanced based on the Canny edge detection algorithm combined with the adaptive threshold adjustment method, and the intensity value of the edge pixel is adjusted according to the importance of the edge and the required sharpening degree to obtain the sharpened object model data; the topological relationship of the internal structure of the model is adjusted according to the structural strength requirements to generate a support structure, and the support position and structure are calculated according to the model shape and the physical properties of the material to obtain the object model structure data; the object is printed layer by layer based on the sharpened object model data and the object model structure data. 1. Improve model accuracy: Through 3D structured light scanning and a series of data processing algorithms, such as point cloud alignment, outlier detection, hole filling, etc., the problem of inaccurate and incomplete scan data is effectively solved, the geometric accuracy and integrity of the model are improved, and the final printed object model is closer to the real object. 2. Optimize model quality: Gaussian filtering removes redundant details, Canny edge detection and adaptive threshold adjustment enhance geometric features, improve the smoothness of the model surface and edge clarity, improve the model quality, and make it more in line with actual use needs. 3. Enhance structural rationality: Adjust the topological relationship and generate support structure according to the structural strength requirements, and calculate the support position and structure considering the model shape and material properties, which improves the stability and success rate of the printing process and ensures that the printed object model has good structural strength and stability. 4. Improve printing efficiency: Accurate model data and reasonable support structure design reduce errors and rework during the printing process, improve printing efficiency, and save time and material costs.

[0124] See also Figure 5 , Figure 51 is a schematic diagram of the structure of a device for quickly printing a tactile model of an exhibit provided by an embodiment of the present application. The device 500 for quickly printing a tactile model of an exhibit may have relatively large differences due to different configurations or performances, and may include one or more processors 510 (for example, one or more processors) and a memory 520, and one or more storage media 530 for storing application programs 533 or data 532. Among them, the memory 520 and the storage medium 530 may be temporary storage or permanent storage. The program stored in the storage medium 530 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the device 500 for quickly printing a tactile model of an exhibit. Furthermore, the processor 510 may be configured to communicate with the storage medium 5630 to execute a series of instruction operations in the storage medium 530 on the device 500 for quickly printing a tactile model of an exhibit.

[0125] The exhibit tactile model rapid printing device 500 may also include one or more power supplies 540, one or more wired or wireless network interfaces 550, one or more input and output interfaces 560, and / or one or more operating systems 531, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will appreciate that Figure 5 The structure of the rapid printing device for exhibit tactile models shown does not constitute a limitation on the rapid printing device for exhibit tactile models, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0126] The present application also provides a device for quickly printing tactile models of exhibits. The computer device includes a memory and a processor. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor executes each step of the method for quickly printing tactile models of exhibits in the above-mentioned embodiments.

[0127] The present application also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the various steps of the method for quickly printing a tactile model of an exhibit.

[0128] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0129] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., and other media that can store program codes.

[0130] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0131] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for quickly printing a tactile model of an exhibit, characterized in that: The method for quickly printing an exhibit tactile model comprises the following steps: Scanning the object to be printed using 3D structured light to obtain object point cloud information, aligning the object point cloud information using a point cloud alignment algorithm based on feature matching to obtain printed object model data; Processing the printed object model data using an outlier detection algorithm, deleting outliers, and obtaining initial object model data; Intelligently fill the holes and missing data in the initial object model data during scanning according to the surrounding geometric features to obtain complete object model data; Using a Gaussian filtering algorithm to remove redundant details in the complete object model data to obtain filtered object model data; Based on the Canny edge detection algorithm combined with the adaptive threshold adjustment method, the geometric feature enhancement of the filtered object model data is performed, and the intensity value of the edge pixel is adjusted according to the importance of the edge and the required sharpening degree to obtain the sharpened object model data; The topological relationship of the internal structure of the model is adjusted according to the structural strength requirements to generate a support structure, and the support position and structure are calculated according to the model shape and the physical properties of the material to obtain the object model structure data; the object is printed layer by layer based on the sharpened object model data and the object model structure data.

2. A method for quickly printing a tactile model of an exhibit as claimed in claim 1, characterized in that: The method of scanning the object to be printed by 3D structured light to obtain object point cloud information, aligning the object point cloud information based on a feature matching point cloud alignment algorithm to obtain printed object model data includes: Use 3D structured light scanning equipment to scan the object to be printed, project the coded pattern onto the surface of the object to be printed through a projector, use a binocular camera to shoot the light pattern reflected from the surface of the object from different angles, calculate the coordinates of each camera pixel in three-dimensional space, and obtain the object point cloud information P cloud ={p1, p2, ..., p n }; where p i Represents the three-dimensional coordinates of the i-th point; Obtain each point cloud data point in the object point cloud information, calculate it through the FPFH feature descriptor, and calculate all points in the point cloud data to p i The Euclidean distance of the nearest k points is selected as N(p i ); For N(p i ) for each point p i , calculate its difference with p i The relative position vector v ij =p j -p i , by N(p i ) to obtain point p i The normal vector n i , for each relative position vector v ij Calculate its normal vector n i The projection length μ ij =v ij -n i , and its discovery vector n i The vertical vector t ij =v ij -μ ij n i ; will (t ij , n i ) as a feature vector, and perform histogram statistics on the feature vectors calculated for all k nearest neighbor points to obtain point p i The FPFH feature descriptor F(p i ); By comparing the feature descriptors of two point cloud data from different viewpoints, we find corresponding points, calculate the Euclidean distance between the corresponding points, and use the least squares method to calculate the rotation and translation transformation matrix to align the two point cloud data. Let p i1 and p j2 is a pair of corresponding points, p i1 =(X i1 , Y i1 , Z i1 ), p j2 =(X j2 , Y j2 , Z j2 ), establish the error function: Where R is the rotation matrix, t is the translation vector, (X i1 , Y i1 , Z i1 ) and (X j2 , Y j2 , Z j2 ) is the three-dimensional coordinate of the corresponding point; the object point cloud information P cloud All points in are transformed to obtain the printed object model data.

3. The method for quickly printing a tactile model of an exhibit according to claim 1, characterized in that: The method of processing the printed object model data by using an outlier detection algorithm, deleting outliers, and obtaining initial object model data includes: The printed object model data is processed using an outlier detection algorithm based on statistical analysis, and the K-nearest neighbor algorithm is used to obtain the outlier detection algorithm of each point p in the printed object model data. i Neighborhood points of distance p i The nearest K points are taken as its neighborhood point set; Calculate point p i The statistical characteristics of the distance between the points and its neighboring points are used to set an abnormal threshold through the 3σ principle, and the abnormal point is judged by using the abnormal threshold; All points in the printed object model data are traversed, and data determined to be abnormal points are deleted from the data set to obtain initial object model data.

4. The method for quickly printing a tactile model of an exhibit according to claim 1, characterized in that: The method of intelligently filling holes and missing data in the initial object model data during scanning according to surrounding geometric features to obtain complete object model data includes: Calculating the normal vector of each point in the initial object model data based on a normal vector analysis method, and identifying holes and missing data areas in the initial object model data through the normal vector; The holes and missing data areas are filled by using a region growing algorithm. i , initialize a growing queue Q = {p i }, and create a visited point set V = {p i }, when the growth queue Q is not empty; Take out the point p at the head of the queue c , using the k-nearest neighbor algorithm to calculate the point p c and point p c Unvisited neighboring point p n The Euclidean distance between them is , and the points whose distance is less than the threshold are selected as the neighborhood points; Calculate the neighborhood point p n With the current point p c The normal vector angle θ cn , let p c The normal vector is n c , p n The normal vector is n n , then θ cn =arccos(n c ·n n ), if θ cn If it is smaller than the threshold, it means that the normal vector directions are close; Calculate the neighborhood point p n The average distance to the filled area d n , let the point set of the filled area be p filled ,but: Where X n , X f , Y n , Y f , Z n , Z f is the three-dimensional coordinate of the neighborhood point, if d n，filled If it is less than the threshold, the surface neighborhood points are close to the filled area; If the neighboring point p n If the above conditions of normal vector angle and distance to the filled area are met, it will be added to the growth queue Q and the visited point set V, and used as a new filling point to update the geometric feature information of the filled area, and repeated iterations will be performed to obtain the complete object model data.

5. The method for quickly printing a tactile model of an exhibit according to claim 1, characterized in that: The step of removing redundant details from the complete object model data by using a Gaussian filtering algorithm to obtain filtered object model data includes: Using a Gaussian filtering algorithm to remove redundant details in the complete object model data, and determining neighborhood points in the complete object model data; The weight of each neighborhood point is calculated based on a three-dimensional Gaussian function, the filtering degree is controlled according to the weight, and the new data value of each data point in the complete object model data is calculated according to the filtering degree; and the filtered object model data is obtained.

6. The method for quickly printing a tactile model of an exhibit as claimed in claim 1, characterized in that: The method of performing geometric feature enhancement on the filtered object model data based on the Canny edge detection algorithm combined with the adaptive threshold adjustment method, adjusting the intensity value of the edge pixel according to the importance of the edge and the required sharpening degree, and obtaining the sharpened object model data includes: Performing geometric feature enhancement on the filtered object model data based on a Canny edge detection algorithm combined with an adaptive threshold adjustment method; Calculate the partial derivatives of the data points in the filtered object model data in the X, Y, and Z directions using the central difference method to obtain the gradient amplitude of the model data; Compare the gradient magnitude of each point with the gradient magnitude of its neighboring points along the gradient direction to obtain edge point data and non-edge point data; Calculate the average value and standard deviation of the gradient amplitude of all points in the area, determine the statistical characteristics of the local area based on the average value and standard deviation, obtain a high threshold and a low threshold, mark the points with a gradient amplitude greater than the high threshold as edge points, and mark the points with a gradient amplitude less than the low threshold as non-edge points; connect the edge points to obtain a preliminary edge detection result; The pixel intensity value is adjusted according to the importance of the edge and the required sharpening degree to obtain the sharpened object model data.

7. The method for quickly printing a tactile model of an exhibit according to claim 1, characterized in that: The topological relationship of the internal structure of the model is adjusted according to the structural strength requirement to generate a support structure, and the support position and structure are calculated according to the physical properties of the model shape and the material to obtain the object model structure data; and the object is printed layer by layer based on the sharpened object model data and the object model structure data, including: Divide the model into multiple small units according to its geometric shape, determine the node coordinates of each unit and the connection relationship between the nodes; input the elastic modulus, Poisson's ratio, density and thermal expansion coefficient of the material according to the material properties used; The allowable stress is set, and the density-based topology optimization algorithm obtains the optimal topology structure that meets the structural strength requirements by iteratively adjusting the material distribution of the unit. The material density of the unit is updated according to the criteria of the topology optimization algorithm to obtain the optimized model topology structure. Identify the cantilever structure, thin-wall structure and hole structure of the model, calculate the curvature radius, wall thickness and angle of different parts of the model, and obtain the structural data of the object model; The object model structure data and the sharpened object model data are converted into G-CODE format data, and the model is sliced ​​reasonably according to the size of the model and the size of the printing area of ​​the printer; the printing temperature, material flow rate and printing speed are controlled to print the model layer by layer.

8. A device for quickly printing tactile models of exhibits, characterized in that: The exhibit tactile model rapid printing device includes the following modules: An object information scanning module is used to scan the object to be printed using 3D structured light to obtain object point cloud information, and align the object point cloud information based on a point cloud alignment algorithm based on feature matching to obtain printed object model data; An abnormal data deletion module is used to process the printed object model data using an abnormal point detection algorithm, delete abnormal points, and obtain initial object model data; A missing data filling module is used to intelligently fill in the holes and missing data in the initial object model data during scanning according to the surrounding geometric features to obtain complete object model data; A print data adjustment module, used to remove redundant details in the complete object model data by using a Gaussian filtering algorithm to obtain filtered object model data; An object internal adjustment module is used to enhance the geometric features of the filtered object model data based on the Canny edge detection algorithm combined with the adaptive threshold adjustment method, and adjust the intensity value of the edge pixel according to the importance of the edge and the required sharpening degree to obtain the sharpened object model data; The object 3D printing module is used to adjust the topological relationship of the internal structure of the model according to the structural strength requirements to generate a support structure, calculate the support position and structure according to the model shape and the physical properties of the material, and obtain the object model structure data; based on the sharpened object model data and the object model structure data, the object is printed layer by layer.

9. A rapid printing device for tactile models of exhibits, characterized in that: The exhibit tactile model rapid printing device comprises: a memory and at least one processor, wherein the memory stores instructions; The at least one processor calls the instructions in the memory to enable the device for quickly printing the tactile model of an exhibit to execute each step of the method for quickly printing the tactile model of an exhibit according to any one of claims 1 to 7.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the steps of the method for quickly printing the tactile model of an exhibit are implemented.

Images