3D printing method, printer, system, and data processing method and apparatus

Abstract

The present disclosure relates to the technical field of 3D printing, and provides a 3D printing method, a printer, a system, and a data processing method and apparatus. The 3D printing data processing method of the present disclosure comprises: on the basis of 3D data of an object to be printed, acquiring slice images comprising grayscale information; and sending the slice images to a photocuring ceramic printer, wherein a light source in the photocuring ceramic printer emits light of a fixed intensity to a light adjustment mechanism, the light adjustment mechanism adjusts, on the basis of the grayscale information in the slice images, the intensity of light projected to a printing platform, and the surface of the printing platform is located on the upper surface of a photosensitive ceramic slurry.

Description

3D printing methods, printers and systems, and data processing methods and apparatus

[0001] Cross-reference of related applications

[0002] This application is based on and claims priority to CN application No. 202411958148.6, filed on December 27, 2024, the disclosure of which is incorporated herein by reference in its entirety. Technical Field

[0003] This disclosure relates to the field of 3D printing technology, and in particular to a 3D printing method, printer and system, and data processing method and apparatus. Background Technology

[0004] Ceramic photopolymer additive manufacturing is a process that precisely controls the polymerization of multifunctional polymer monomers, transforming liquid slurries containing ceramic particles into solid three-dimensional objects. This innovation has revolutionized the traditional manufacturing of ceramic components and has wide applications and impacts in fields such as tissue engineering, dentistry, microfluidics, bioprinting, soft robotics, metamaterials, and photonics.

[0005] When printing objects using surface projection ceramic photopolymer additive manufacturing, the relevant technologies first employ 3D graphics software to design or obtain a three-dimensional model to be printed through scanning modeling. For some complex shapes, additional specific support structures are usually required to ensure printing quality and efficiency. Subsequently, the model is sliced ​​according to certain rules in slicing software and discretized into a series of ordered layer units. Finally, by setting different printing parameters, the printing paste is connected layer by layer through photopolymerization based on the previously sliced ​​layer units (printing file) to obtain the target three-dimensional ceramic part.

[0006] The general working principle of ceramic photopolymerization 3D printing technology is based on the photopolymerization reaction of liquid photosensitive resin. That is, the ceramic printing paste containing photosensitive resin can rapidly undergo a photopolymerization reaction under ultraviolet light of a specific wavelength and intensity, causing a sharp increase in molecular weight and transforming the material from a liquid to a solid state. Ideally, the energy of the light beam is concentrated, forming a uniform curing area. However, in reality, the light beam penetrates the printing paste in a Gaussian distribution, and when the light intensity reaches the critical energy (Ec), the curing process becomes more complex. c When the photopolymerization reaction begins, the cured region is C. d (curing depth) and C w (curing width). Summary of the Invention

[0007] One objective of this disclosure is to improve the accuracy of photopolymer ceramic printing.

[0008] According to one aspect of some embodiments of this disclosure, a three-dimensional printing data processing method is proposed, comprising: acquiring a slice image containing grayscale information based on three-dimensional data of an object to be printed; sending the slice image to a photocurable ceramic printer, wherein a light source in the photocurable ceramic printer emits light of a fixed intensity to a light adjustment mechanism, the light adjustment mechanism adjusts the intensity of the light projected onto the printing platform according to the grayscale information in the slice image, and the surface of the printing platform is located on the upper surface of a photosensitive ceramic paste.

[0009] In some embodiments, obtaining a slice image containing grayscale information includes: obtaining grayscale information corresponding to three-dimensional data based on a machine learning model, and obtaining a slice image including grayscale information of each pixel.

[0010] In some embodiments, the machine learning model is trained to generate a corresponding grayscale image sequence based on the ceramic photocuring sample, wherein the ceramic photocuring sample includes the correspondence between grayscale image sample data and the corresponding three-dimensional data of the cured object.

[0011] In some embodiments, the light adjustment mechanism includes a digital micromirror device (DMD) or a micromirror imager (SLM). The DMD adjusts the reflection angle of the corresponding pixel's wafer according to grayscale information to adjust the intensity of the light projected onto the printing platform.

[0012] In some embodiments, obtaining a slice image containing grayscale information includes: obtaining grayscale information corresponding to each pixel based on three-dimensional data, obtaining a first image including grayscale information; and gradually linearly shrinking or expanding the two-dimensional graphic corresponding to the boundary contour of the image to a set ratio to generate a set number of second images as slice images.

[0013] In some embodiments, obtaining slice images containing grayscale information based on the three-dimensional data of the object to be printed includes: obtaining model slice data based on the three-dimensional data; and obtaining each slice image containing grayscale information according to the depth-grayscale superposition information corresponding to the voxels of the three-dimensional data, based on the number of slice layers.

[0014] In some embodiments, obtaining slice images containing grayscale information based on the three-dimensional data in each slice data includes: for each slice data, obtaining grayscale information corresponding to each pixel based on the three-dimensional data to obtain a first image including grayscale information; for each first image, gradually linearly shrinking or expanding the two-dimensional graphic corresponding to the boundary contour of the image to a set ratio to generate a set number of second images as slice images.

[0015] In some embodiments, the light intensity of the light source used to generate the ceramic photocured sample is the same as the light intensity of the light source in the photocured ceramic printer.

[0016] In some embodiments, the total printing time for the cured object corresponding to each sample data in the ceramic photocuring sample is the same as the printing time for a single slice image by the photocuring ceramic printer.

[0017] In some embodiments, the data processing method further includes: determining a grayscale value sequence according to a predetermined gradient; generating grayscale image sample data for each grayscale value in the grayscale value sequence; using a photocurable ceramic printer to obtain the three-dimensional data of the cured object generated by printing the grayscale image sample data; determining the mapping relationship between the grayscale image sample data and the corresponding three-dimensional data of the cured object; obtaining a ceramic photocurable sample based on the correspondence between the grayscale image sample data of each grayscale value in the grayscale value sequence and the three-dimensional data of the cured object; and training the constructed machine learning neural network using the ceramic photocurable sample to obtain a machine learning model.

[0018] In some embodiments, the object to be printed includes dental veneers.

[0019] According to one aspect of some embodiments of this disclosure, a three-dimensional printing method for a photopolymer ceramic printer is proposed, comprising: acquiring a slice image, wherein the slice image is generated according to any of the three-dimensional printing data processing methods mentioned above; emitting light of a fixed intensity to a light adjustment mechanism; the light adjustment mechanism adjusting the intensity of the light projected onto a printing platform according to grayscale information in the slice image, wherein the surface of the printing platform is located on the upper surface of a photosensitive ceramic paste.

[0020] According to one aspect of some embodiments of this disclosure, a three-dimensional printing data processing apparatus is proposed, comprising: a slice image acquisition unit configured to acquire a slice image containing grayscale information based on three-dimensional data of an object to be printed; and a data transmission unit configured to send the slice image to a photocurable ceramic printer, wherein a light source in the photocurable ceramic printer emits light of a fixed intensity to a light adjustment mechanism, the light adjustment mechanism adjusts the intensity of the light projected onto the printing platform according to the grayscale information in the slice image, and the surface of the printing platform is located on the upper surface of a photosensitive ceramic paste.

[0021] In some embodiments, the data processing apparatus further includes: a sample data acquisition unit configured to determine a grayscale value sequence according to a predetermined gradient, generate grayscale image sample data for each grayscale value in the grayscale value sequence, acquire three-dimensional data of a cured object generated by printing the grayscale image sample data using a photocurable ceramic printer, determine the correspondence between the grayscale image sample data and the corresponding three-dimensional data of the cured object, and acquire a ceramic photocurable sample based on the correspondence between the grayscale image sample data of each grayscale value in the grayscale value sequence and the three-dimensional data of the cured object; and a model training unit configured to train a machine learning network built using the ceramic photocurable sample to acquire a machine learning model, wherein the slice image acquisition unit is configured to acquire grayscale information corresponding to the three-dimensional data based on the machine learning model, and acquire a slice image including grayscale information of each pixel.

[0022] According to one aspect of some embodiments of this disclosure, a three-dimensional printing data processing apparatus is provided, comprising: a memory; and a processor coupled to the memory, the processor being configured to execute any of the three-dimensional printing data processing methods described above based on instructions stored in the memory.

[0023] According to one aspect of some embodiments of the present disclosure, a computer-readable storage medium is provided that stores computer instructions which, when executed by a processor, implement any of the three-dimensional printing data processing methods described above.

[0024] According to one aspect of some embodiments of this disclosure, a computer program product is proposed, including a computer program or instructions that, when executed by a processor, implement any of the three-dimensional printing data processing methods described above.

[0025] According to one aspect of some embodiments of this disclosure, a computer program is provided for causing a processor to execute any of the three-dimensional printing data processing methods described above.

[0026] According to one aspect of some embodiments of this disclosure, a 3D printer is provided, comprising: a communication mechanism configured to acquire a slice image, wherein the slice image is generated according to any of the 3D printing data processing methods described above; a light source configured to emit light of a fixed intensity to a light adjustment mechanism; the light adjustment mechanism configured to adjust the intensity of light projected onto a printing platform according to grayscale information in the slice image; and a printing platform, the surface of which is located on the upper surface of a photosensitive ceramic slurry and configured to carry the printed object.

[0027] According to one aspect of some embodiments of this disclosure, a three-dimensional printing system is proposed, comprising: any of the three-dimensional printing data processing devices described above; and any of the three-dimensional printers described above. Attached Figure Description

[0028] The accompanying drawings, which are included to provide a further understanding of this disclosure and form part of this disclosure, illustrate exemplary embodiments of the present disclosure and are used to explain the disclosure, but do not constitute an undue limitation of the disclosure. In the drawings:

[0029] Figure 1 is a flowchart of some embodiments of the three-dimensional printing data processing method disclosed herein.

[0030] Figure 2 is a schematic diagram of some embodiments of gamma correction of the digital light processing (DLP) optical engine in the 3D printing data processing method of this disclosure.

[0031] Figure 3 shows the relationship between the UV curing depth and exposure energy of the ceramic slurry in the 3D printing data processing method of this disclosure.

[0032] Figure 4 shows the relationship between the UV curing depth and exposure time of ceramic slurry under a fixed light intensity in the 3D printing data processing method of this disclosure.

[0033] Figure 5 is a schematic diagram of some embodiments of the 3D printing data processing method of this disclosure based on machine learning network algorithms.

[0034] Figure 6 is a flowchart of some embodiments of model training in the 3D printing data processing method of this disclosure.

[0035] Figure 7 is a flowchart of some embodiments of the three-dimensional printing method disclosed herein.

[0036] Figure 8 is a schematic diagram of some embodiments of the three-dimensional printing data processing apparatus of this disclosure.

[0037] Figure 9 is a schematic diagram of some other embodiments of the three-dimensional printing data processing apparatus of this disclosure.

[0038] Figure 10 is a schematic diagram of some further embodiments of the three-dimensional printing data processing apparatus of this disclosure.

[0039] Figure 11 is a schematic diagram of some embodiments of the 3D printer disclosed herein.

[0040] Figure 12 is a schematic diagram of some embodiments of the 3D printing system disclosed herein.

[0041] Figure 13 is a schematic diagram of an embodiment of the three-dimensional printing data processing method and three-dimensional printing method disclosed herein.

[0042] Figure 14 is a schematic diagram of another embodiment of the three-dimensional printing data processing method and three-dimensional printing method disclosed herein.

[0043] Figure 15 is a schematic diagram of yet another embodiment of the three-dimensional printing data processing method and three-dimensional printing method disclosed herein.

[0044] Figure 16 is a schematic diagram of another embodiment of the three-dimensional printing data processing method and three-dimensional printing method disclosed herein.

[0045] Figure 17 is a schematic diagram of the control experiment. Detailed Implementation

[0046] The technical solutions of this disclosure will be further described in detail below with reference to the accompanying drawings and embodiments.

[0047] The inventors discovered that during actual printing, the printing paste scatters incident light to varying degrees, causing photon energy to diffuse over a large area and disrupting the beam's focus. Light scattering increases the width of the cured region (C). w Increased), but the curing depth (C) d The size of the ceramic particles will decrease. Typically, the solid content of micro- and nano-scale ceramic particles in ceramic slurries used for photopolymer 3D printing is as high as 80 wt% (volume fraction close to or exceeding 50 vol%), exhibiting a significant light scattering effect. Due to these issues, the printed objects are rough and have low precision.

[0048] To address the problems existing in related technologies, this disclosure proposes a three-dimensional printing method, printer, system, data processing method, and apparatus, which reduces the impact of light scattering in liquids on printing accuracy and improves the precision of photocurable ceramic printing.

[0049] Flowcharts of some embodiments of the 3D printing data processing method disclosed herein are shown in Figure 1.

[0050] In step S12, a slice image containing grayscale information is obtained based on the three-dimensional data of the object to be printed.

[0051] In some embodiments, grayscale information corresponding to each pixel can be obtained first based on three-dimensional data to obtain a first image including grayscale information. Further, the two-dimensional graphics corresponding to the boundary contours of the image are gradually linearly reduced or expanded to a set ratio to generate a set number of second image combinations (e.g., 10-100 images). These second image combinations are then used as the overall slice image of the sample. In some embodiments, the printer prints the slice images continuously without raising or lowering the printing platform. The position of the printing platform and its distance from the light source remain unchanged; that is, the set number of second image combinations belong to the same layer of the printed object.

[0052] Projection photopolymerization 3D printing, a related technology, employs layer-by-layer additive manufacturing. The basic principle involves dividing a designed 3D model into continuous thin layers (slices) of a certain thickness, and then building a complete 3D part by solidifying and stacking these layers one by one. However, this layer-by-layer manufacturing method can lead to problems such as uneven distribution of projection light or flow issues, resulting in differences in intra-layer and inter-layer bonding strength (insufficient inter-layer bonding strength) and printing internal stress. This can ultimately lead to printing failures, cracking after sintering, or poor mechanical strength of the formed part. Furthermore, layer-by-layer deposition typically requires the introduction of additional printing support rods to prevent overall deformation of the printed blank. This introduces additional pre- and post-processing steps such as the design and removal of printing support rods, increasing the complexity of the work and lengthening the printing process (usually several hours). In particular, for some extremely thin and precise complex parts, removing the support rods may cause deformation, cracking, or microcracks. In addition, the surface of layer-by-layer printed parts often exhibits printing textures, making it difficult to directly form a smooth surface.

[0053] The method described in the above embodiments of this disclosure can transform layered printing into a continuous printing process, avoiding problems such as layer texture, internal stress, anisotropy, support structure, and deformation, defects, and cracks of the part caused by removing the support, thereby improving printing efficiency and accuracy.

[0054] In some embodiments, when the object to be printed is relatively thick (e.g., the thickness exceeds a predetermined layer height), the object to be printed can be first sliced ​​to obtain slice data for each layer. Then, for each slice data, the grayscale information of each slice is obtained as the first image of the corresponding slice. Further, for each first image, according to the depth-grayscale superposition information corresponding to the three-dimensional data voxels, each slice image containing grayscale information is obtained according to the number of slice layers. For example, the two-dimensional graphic corresponding to the boundary contour of the image is linearly reduced or expanded to a set ratio to generate a set number of second images, thereby obtaining a combination of multiple second images, all of which are used as slice images. In some embodiments, the printer prints slice images in the same second image combination as a continuous operation without raising or lowering the printing platform. That is, the second images belonging to the same second image combination belong to the same layer of the printed object. After the printing of a second image combination is completed, the height of the printing platform is adjusted once, and the printing of the next set of second image combinations is executed. This reduces the number of layers required while achieving the printing of thicker objects and improves the printing accuracy of each layer.

[0055] In some embodiments, the operation of obtaining grayscale information based on three-dimensional data can be implemented using a machine learning model (e.g., a machine learning neural network). For example, by inputting three-dimensional data into a machine learning neural network, the machine learning network can obtain the grayscale information corresponding to each pixel in the three-dimensional data, and then output a slice image including the grayscale information of each pixel, thereby improving the data processing efficiency.

[0056] In some embodiments, the machine learning model used in this disclosure is a deep learning neural network model. The deep learning network is trained based on a sequence of grayscale images generated from a ceramic photocured sample to obtain a generative network model. The ceramic photocured sample includes a correspondence between grayscale image sample data and the corresponding three-dimensional data of the cured object. This method enables the generative network model to convert three-dimensional data into grayscale images. Because the correspondence between the three-dimensional data of the cured object and the grayscale image is established, printing deviations caused by light scattering and other factors can be overcome, improving the matching degree between the grayscale image and the desired printing result, and thus improving printing accuracy.

[0057] In some embodiments, the light intensity of the light source used to generate the ceramic photocurable sample is the same as the light intensity of the light source in the photocurable ceramic printer. The method described in the above embodiments avoids deviations in the printed object caused by changes in light intensity in the grayscale image generated by the deep learning-based generative model, further improving printing accuracy.

[0058] In some embodiments, the total printing time for the cured object corresponding to each sample data in the ceramic photocuring sample is the same as the printing time for a single slice image by the photocuring ceramic printer. The method described in the above embodiments avoids deviations in the printed object caused by variations in printing time based on the grayscale image generated by the generative network model, further improving printing accuracy.

[0059] In step S14, the sliced ​​image is sent to the photocurable ceramic printer.

[0060] In a photopolymer ceramic printer, a light source emits light of a fixed intensity to a light adjustment mechanism. The light adjustment mechanism adjusts the intensity of the light projected onto the printing platform based on the grayscale information in the sliced ​​image. The surface of the printing platform is located on the upper surface of the photosensitive ceramic paste. In some embodiments, the light adjustment mechanism includes a DMD (Digital Microcontroller), which adjusts the reflection angle of the corresponding pixel's wafer based on the grayscale information to adjust the intensity of the light projected onto the printing platform. In some embodiments, the height of the printing platform can be adjusted as shown in the embodiment described in step S12 above, thereby avoiding or reducing the layer-by-layer stacking structure of the printed object and improving printing efficiency and accuracy.

[0061] Based on the method described in the above embodiments, the three-dimensional data of the object to be printed can be converted into a slice image including grayscale information. The photopolymer ceramic printer adjusts the intensity of the light projected onto the printing platform according to the grayscale information. The grayscale information has multiple levels of grayscale (e.g., 256 levels of grayscale). Compared with binary images, the rich grayscale data can more smoothly transition between different areas, better represent details and surface smoothness during printing exposure, reduce the step effect of printing layers, improve surface quality, and improve printing accuracy. In addition, changing the method of layer-by-layer stacking printing to project images including grayscale information can avoid problems such as weak interlayer bonding and surface texture caused by hierarchical structures, further improving printing accuracy.

[0062] In some embodiments, the object to be printed includes a dental veneer. Using the 3D printing data processing method and 3D printing method proposed in this disclosure, the thickness of the printed dental veneer can be reduced, the fit between the veneer and the patient's teeth can be improved, the required adjustment height for the patient's teeth can be reduced, and user comfort can be improved.

[0063] DLP (Digital Light Processing) is a photopolymerization 3D printing technology based on surface projection. It consists of a high-intensity light source (such as a UV lamp or LED) and a DMD (Digital Micromirror Device). A DMD is a chip containing millions of tiny mirrors, each of which can be independently tilted to control light reflection. When the printer receives a print command, the light emitted by the light source is modulated by the DMD. Each mirror on the DMD tilts according to the slice data of the 3D model, forming a corresponding grayscale image. The image of each layer is projected through the DMD onto the surface of photosensitive resin in a resin tank. Irradiation causes the photopolymerized resin layer to solidify and form according to the predetermined design.

[0064] DLP printing relies on the ultrasensitive photoinitiation reaction of photocurable slurry to cure light, thus allowing for precise control of the light irradiation intensity and curing degree of each region using grayscale images. Given that grayscale images typically provide 256 levels of gray (as shown in Figure 2), this allows for better representation of details and surface smoothness during printing exposure. Compared to binary images, grayscale images can more smoothly transition between different regions, thereby reducing the step effect in the printed layers and improving surface quality. Therefore, the exposure intensity of different regions in the photocurable slurry can be precisely controlled, enabling precise control of the curing depth and degree of curing in different areas of three-dimensional space, laying the foundation for the subsequent fabrication of complex structures.

[0065] For example, using a laser power of 125mW / cm 2Taking a DLP printer as an example. Generally, the output brightness of the projector used in DLP printing is non-linear. Therefore, gamma correction is needed for the input grayscale image to ensure that the brightness of the light source output matches the desired target data of the image, thereby achieving precise control over the curing degree of the target area. By fixing the initial power of the laser, adjusting a series of grayscale values ​​of the projected image (e.g., labeled 0-255 in descending order of grayscale value), and using a light intensity meter to measure the actual light intensity values ​​corresponding to different grayscale levels in the printing area, the non-linear relationship between the grayscale values ​​in the slice data and the actual projected light intensity can be obtained (as shown in Figure 2). Gamma correction typically uses the following formula: I = C γ

[0066] In the formula: I is the corresponding optical power value; C is the known gray value; γ is the gamma value, which determines the degree of adjustment.

[0067] After gamma correction, the photosensitive properties of the photocurable paste were characterized by studying two key parameters: critical exposure and curing depth. The photocurable ceramic paste was placed in a self-made quartz groove (e.g., shown in Figure 13(c)) with a depth of 1 mm and an area of ​​50 × 50 mm, and after self-leveling, it was placed on the printing platform of a projection-type DLP printer. The projector laser power was fixed at 125 mW / cm². 2 The critical exposure of the photocurable ceramic slurry was determined by adjusting the grayscale value of the projected image to control the actual laser irradiation intensity, and by controlling the curing time (1 second). Experiments showed that the slurry could not be formed when the grayscale value was less than 5. This is because the absorption of ultraviolet light by liquid photosensitive slurry generally follows the Beer-Lambert theorem, and the absorption of ultraviolet light by ultraviolet laser onto the surface of the photosensitive slurry also conforms to the Beer-Lambert theorem, meaning that the energy of the ultraviolet laser decreases exponentially with increasing irradiation depth.

[0068] In the formula, E is the incident ultraviolet light energy density; Z is the depth, and E(z) is the laser energy density transmitted to depth Z; D p The curing depth is an inherent parameter of the photosensitive paste, representing the strength of its ability to absorb ultraviolet laser light. (D) p The smaller the value, the stronger the absorption of ultraviolet laser by the slurry.

[0069] Critical Exposure Energy (E) c The minimum laser irradiation energy (E) required for the ceramic slurry to form a solid layer is the minimum energy required for 3D printing. When the exposure amount (E) of the liquid photosensitive slurry to ultraviolet light exceeds a certain threshold, i.e., the critical exposure energy E... c After that, that is, when

[0070] The photosensitive paste will rapidly polymerize and undergo a phase transition, changing from a liquid to a solid state. At this point,

[0071] That is, the curing depth can be further expressed as C. d =D p lnE-D p lnE c

[0072] With lnE as the x-axis, C d Using lnE as the ordinate, we can see that lnE and C d The relationship is linear, and the slope of the straight line is D. p The intersection of the straight line and the lnE axis is lnE. c By setting a series of exposure light intensities and using a thickness gauge to measure the thickness of the cured slurry, the relationship between the cured layer thickness and the corresponding energy input is plotted. UV light intensity represents the amount of UV photon energy received per unit area within a specified wavelength range, representing the photon flux. Therefore, the light energy can also be controlled by increasing the exposure time of the 3D printer: E = I × t

[0073] In the formula, I represents light intensity, with units of mW / cm². 2 Where t is the exposure time in seconds. Using the curing depth formula, E can be obtained through fitting and calculation. c and D p The value is shown in Figure 3.

[0074] Under the same exposure time, the curing depth usually increases with the increase of light intensity (I); however, under a certain light intensity, the curing depth of the printed part tends to level off with the increase of exposure time (as shown in Figure 4).

[0075] Therefore, by adjusting the intensity and duration of the projected light, the penetration depth and curing position of the laser can be changed, thereby obtaining ceramic parts with controllable three-dimensional thickness.

[0076] Based on the above principles, adjusting the illumination intensity and duration of the projected light can alter the laser's penetration depth and curing position, providing a theoretical foundation for the method disclosed herein. In this disclosure, adjusting the illumination intensity of the projected light changes the laser's penetration depth and curing position; simultaneously, it avoids excessive influence of increased exposure time on the penetration depth and curing position. This facilitates continuous printing by continuously projecting different slice images containing grayscale information, thereby improving printing accuracy and efficiency.

[0077] Flowcharts of some embodiments of model training in the 3D printing data processing method disclosed herein are shown in Figure 5.

[0078] In step 511, a grayscale value sequence is determined according to a predetermined gradient. In some embodiments, grayscale value sequences can be generated according to the number of grayscale levels, for example, generating 256 sequences to improve the accuracy of model training and subsequent printing.

[0079] In some embodiments, as mentioned above, light cannot be solidified when the gray value is less than 5. Therefore, a gray value sequence can be generated for gray values ​​greater than 5, thereby reducing the amount of data.

[0080] In some embodiments, a predetermined gradient greater than 1 can be set to further reduce the amount of data that needs to be processed and printed, thereby improving the efficiency of sample acquisition and model training.

[0081] In step 512, grayscale image sample data is generated for each grayscale value in the grayscale value sequence. In some embodiments, a grayscale image of at least one shape may be generated for each grayscale value in the grayscale value sequence as grayscale image sample data.

[0082] In step 513, a photocurable ceramic printer is used to acquire the three-dimensional data of the cured object generated by printing grayscale image sample data, and the correspondence between the grayscale image sample data and the corresponding three-dimensional data of the cured object is determined. In some embodiments, after printing with the photocurable ceramic printer, the cured object can be scanned to obtain the three-dimensional data of the object. Then, the grayscale image sample data used to print the object can be associated with the three-dimensional data obtained by scanning the object to obtain a correspondence, which serves as a photocurable ceramic sample. The light intensity of the light source used in the printing process in step 513 is the same as the light intensity used in subsequent use, thereby avoiding inaccurate data conversion caused by light intensity mismatch and improving the reliability of the data processing method.

[0083] In step 514, ceramic photocuring samples are obtained based on the correspondence between the grayscale image sample data of each grayscale value in the grayscale value sequence and the three-dimensional data of the cured object. For example, by summarizing the grayscale image sample data obtained for each grayscale value in step 512 and the correspondence obtained in step 513, all ceramic photocuring samples are obtained.

[0084] In step 515, the network built by training the ceramic photocuring sample, such as a deep learning network, is used to obtain the machine learning model used in step S12 above, such as a generative network model.

[0085] Based on the method in the above embodiments, grayscale information and curing range obtained from experiments are used as sample data to train a deep learning network. The deep learning network is then used to reverse engineer the curing effect (including depth and range) into a grayscale image, generating an image with grayscale information. This allows for dynamic control of the projected light energy, enabling the one-time forming of three-dimensional parts with different thicknesses.

[0086] In some embodiments, as mentioned above, taking the use of a deep learning neural network as an example, the grayscale image sample data sequence and the corresponding 3D data are used as inputs to the deep learning network. For example, PointNet is used to extract the 3D data distribution features. The calculation process is as follows: F L (P i ) = MLP(P i PointNet(P) i =FC(Concat(F) G (P i ),F L (P i )))

[0087] In the formula, F G and F L P represents the global and local features of three-dimensional data. i This represents three-dimensional data (taking point cloud data as an example).

[0088] Furthermore, a constructed 3D CNN is used to extract image features from the grayscale image sequence, which are then fused with the distribution features of the 3D data to improve the semantic consistency between the two types of features, thereby establishing a mapping relationship between the grayscale values ​​of a single-layer image and the degree of curing. Then, a recurrent neural network (RNN) is used to extract the sequence information of the grayscale image sequence, that is, the scattering effect of the exposure dose corresponding to the current layer on the curing effect of previous layers in each grayscale image. The calculation process of the RNN is shown in the formula: h t =f(Wh t－1 +Ux t )

[0089] In the formula, h t with h t-1 W and U represent the 3D data and grayscale fusion features of the current layer and the previous layer, respectively, and represent the preset weight parameters, x and y. t Let f be the input image of the current target structure, and f be a non-linear activation function.

[0090] Finally, an optimized grayscale image sequence is generated through an image sequence generator, thereby achieving intelligent control of the projection light. The network structure flow of the above processing is shown in Figure 6.

[0091] Based on the method in the above embodiments of this disclosure, the degree of photocuring reaction induced by projection light in different regions can be intelligently and programmably controlled according to a predetermined design. The relationship between light dose and curing is obtained through deep learning algorithms, and grayscale information of optimized slice files is generated. In this way, the distribution and change of projection light energy in three-dimensional space can be controlled to different degrees, which is expected to form ceramic components with complex three-dimensional structures in a very short time. No additional pre- and post-processing steps of designing and removing printing support rods are required, simplifying the process and significantly improving printing efficiency.

[0092] Flowcharts of some embodiments of the 3D printing method disclosed herein are shown in Figure 7.

[0093] In step S71, the slice image is obtained. The slice image is generated according to any of the 3D printing data processing methods mentioned above.

[0094] In step S72, light of a fixed intensity is emitted to the light adjustment mechanism. In some embodiments, the light adjustment mechanism is a DMD.

[0095] In step S73, the light adjustment mechanism adjusts the intensity of the light projected onto the printing platform based on the grayscale information in the slice image. The surface of the printing platform is located on the upper surface of the photosensitive ceramic paste. In some embodiments, the position between the printing platform and the light adjustment mechanism remains unchanged during the printing of the object corresponding to the slice image.

[0096] Based on the method in the above embodiments, the intensity of light projected onto the printing platform can be adjusted using slice images containing grayscale information converted from the three-dimensional data of the object to be printed. Compared with binary images, the rich grayscale data allows for smoother transitions between different areas, better representation of details and surface smoothness during printing exposure, reduced step effects in printing layers, improved surface quality, and increased printing accuracy. Furthermore, changing the method of layer-by-layer stacking printing to project images containing grayscale information can avoid problems such as weak interlayer bonding and surface texture caused by hierarchical structures, further improving printing accuracy.

[0097] Schematic diagrams of some embodiments of the 3D printing data processing apparatus disclosed herein are shown in Figure 8.

[0098] The slice image acquisition unit 813 can acquire a slice image containing grayscale information based on the three-dimensional data of the object to be printed. In some embodiments, the slice image acquisition unit 812 can perform the method in any embodiment of step S12 above.

[0099] The data sending unit 814 can send the sliced ​​image to the photopolymer ceramic printer, wherein the light source in the photopolymer ceramic printer emits light of a fixed intensity to the light adjustment mechanism 3, and the light adjustment mechanism 3 adjusts the light intensity projected onto the printing platform according to the grayscale information in the sliced ​​image, and the surface of the printing platform is located on the upper surface of the photosensitive ceramic paste. In some embodiments, the data sending unit 814 can perform the method in any embodiment of step S14 above.

[0100] Based on the apparatus in the above embodiments, the intensity of light projected onto the printing platform can be adjusted using a slice image containing grayscale information, which is converted from the three-dimensional data of the object to be printed. Compared with binary images, the rich grayscale data allows for a smoother transition between different areas, better representation of details and surface smoothness during printing exposure, reduced step effects in the printing layers, improved surface quality, and increased printing accuracy. Furthermore, by changing the method of layer-by-layer stacking printing to project images containing grayscale information, problems such as weak interlayer bonding and surface texture caused by hierarchical structures can be avoided, further improving printing accuracy.

[0101] In some embodiments, as shown in FIG8, the 3D printing data processing device further includes a sample data acquisition unit 811 and a model training unit 812.

[0102] The sample data acquisition unit 811 can determine a grayscale value sequence according to a predetermined gradient, generate grayscale image sample data for each grayscale value in the grayscale value sequence, acquire the three-dimensional data of the cured object generated by printing the grayscale image sample data using a photocurable ceramic printer, determine the correspondence between the grayscale image sample data and the corresponding three-dimensional data of the cured object, and acquire a photocurable ceramic sample based on the correspondence between the grayscale image sample data of each grayscale value in the grayscale value sequence and the three-dimensional data of the cured object. In some embodiments, the sample data acquisition unit 811 can execute the method in any of the embodiments described in steps 511-514 above.

[0103] The model training unit 812 can train the deep learning network built using ceramic photocuring samples to obtain a machine learning model, such as a generative network model. The slice image acquisition unit 813 can acquire the grayscale information corresponding to the three-dimensional data based on the machine learning model, and acquire a slice image including the grayscale information of each pixel.

[0104] Based on the device in the above embodiment, grayscale information and curing range can be obtained through experiments, and these can be used as sample data to train a deep learning network. This enables the reverse engineering of the curing effect (including depth and range) into a grayscale image using a deep learning network model, generating an image with grayscale information. This allows for dynamic control of the projected light energy, and one-time forming of three-dimensional parts with different thicknesses.

[0105] A schematic diagram of one embodiment of the 3D printing data processing apparatus disclosed herein is shown in Figure 9. The 3D printing data processing apparatus includes a memory 901 and a processor 902. The memory 901 can be a disk, flash memory, or any other non-volatile storage medium. The memory is used to store instructions in the corresponding embodiments of the 3D printing data processing method described above. The processor 902 is coupled to the memory 901 and can be implemented as one or more integrated circuits, such as a microprocessor or microcontroller. The processor 902 is used to execute the instructions stored in the memory, which can improve the accuracy of photopolymer ceramic printing.

[0106] In one embodiment, as shown in FIG10, the 3D printing data processing device 1000 includes a memory 1001 and a processor 1002. The processor 1002 is coupled to the memory 1001 via a BUS bus 1003. The 3D printing data processing device 1000 can also be connected to an external storage device 1005 via a storage interface 1004 to access external data, and can also be connected to a network or another computer system (not shown) via a network interface 1006. Further details are omitted here.

[0107] In this embodiment, storing data instructions in a memory and then processing the instructions with a processor can improve the accuracy of photopolymer ceramic printing.

[0108] In another embodiment, a computer-readable storage medium stores computer program instructions that, when executed by a processor, implement the steps of the method in the corresponding embodiment of the 3D printing data processing method. Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, apparatus, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more computer-usable non-transitory storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0109] Schematic diagrams of some embodiments of the 3D printer disclosed herein are shown in Figure 11.

[0110] The communication mechanism 1121 is capable of acquiring slice images. These slice images are generated according to any of the 3D printing data processing methods mentioned above, such as those generated by any of the 3D printing data processing devices mentioned above. The slice images are transmitted to the communication mechanism 1121 via wired or wireless means. After processing by the communication mechanism 1121 or the controller of the 3D printer, control commands are sent to other parts of the 3D printer. The direction of command transmission can be indicated by the thin arrow in Figure 11.

[0111] The light source 1122 can emit light of a fixed intensity to the light adjustment mechanism.

[0112] The light adjustment mechanism 1123 can adjust the intensity of the light projected onto the printing platform according to the grayscale information in the slice image. In some embodiments, the light adjustment mechanism 1123 can continuously project light according to the slice image, without interruption during the projection of different slice images.

[0113] The surface of the printing platform 1124 is located on the upper surface of the photosensitive ceramic paste, and the printing platform is capable of carrying the printed object. In some embodiments, the printing platform 1124 maintains its position.

[0114] The 3D printer in the above embodiment can adjust the light intensity projected onto the printing platform by converting the 3D data of the object to be printed into a slice image including grayscale information. Compared with binary images, the rich grayscale data can more smoothly transition between different areas, better represent details and surface smoothness during printing exposure, reduce the step effect of printing layers, improve surface quality, and improve printing accuracy. In addition, changing the method of layer-by-layer stacking printing to project images including grayscale information can avoid problems such as weak interlayer bonding and surface texture caused by hierarchical structure, further improving printing accuracy.

[0115] Figure 12 shows schematic diagrams of some embodiments of the 3D printing system disclosed herein.

[0116] The 3D printing data processing device 1210 can be any of the 3D printing data processing devices mentioned above. The 3D printer 1220 can be any of the 3D printers mentioned above.

[0117] The 3D printing system disclosed herein can adjust the intensity of light projected onto the printing platform by converting the 3D data of the object to be printed into a slice image including grayscale information. Compared with binary images, the rich grayscale data can more smoothly transition between different areas, better represent details and surface smoothness during printing exposure, reduce the step effect of printing layers, improve surface quality, and improve printing accuracy. In addition, by changing the method of layer-by-layer stacking printing to the method of projecting images including grayscale information, problems such as weak interlayer bonding and surface texture caused by hierarchical structure can be avoided, further improving printing accuracy.

[0118] By establishing a mapping relationship between the intensity of surface projection light and the curing effect through a deep learning-based intelligent programming control method, the forming area is fully irradiated. The linearly and dynamically varying projection light influences the initiation sequence and degree of the photocurable slurry in different areas, further optimizing the local photocuring reaction kinetics. This allows for the suppression of over-curing caused by light scattering by dynamically controlling the projection light pattern in the photocurable ceramic slurry, eliminating rough boundaries and achieving three-dimensional forming of the target geometry, without the need for layer-by-layer stacking or additional supports. This method significantly improves the printing efficiency of ceramic parts (reducing printing time by more than 1000 times) and fundamentally eliminates problems caused by traditional layer-by-layer photocuring stacking, such as layer textures, internal stress, anisotropy, support structures, and deformation, defects, and cracks caused by support removal.

[0119] The following examples demonstrate the effects of the 3D printing method, printer, system, data processing method, and apparatus disclosed herein.

[0120] Example 1:

[0121] To achieve integrated molding of complex irregularly shaped ceramic parts, a 3D modeling software was first used to construct a printing model with a maximum height of 0.2 cm. The model was then arranged in a specific orientation without the need for additional support structures (as shown in Figure 13(a)). Subsequently, the 3D information of the model was decoupled into a series of 2D slice files containing pixel-level grayscale information using Python programming software (as shown in Figure 13(b)). Specifically, the 2D horizontal dimension of the ceramic part model in the initial image was set to 10.00 × 10.00 mm, with the grayscale value maintained at 255. Based on a deep learning network algorithm, the 2D graphic corresponding to the boundary contour of the extremely thin part was linearly reduced (divided into 10 steps for smooth display of the transition process) to 7 / 8 of the original area size.

[0122] The power of the fixed light source is 125mW / cm. 2 A commercially available photosensitive ceramic paste was placed in a self-made quartz plate with a 1mm deep and 50×50mm groove (as shown in Figure 13(c)). After self-leveling, it was placed on the printing platform of a top-projection DLP printer (as shown in Figure 13(d)). The exposure time for a single grayscale image was set to 0.1s, and the above-mentioned slice dataset was continuously exposed until the end, thus forming a complex free-form thin-walled ceramic part in a single printing process in 40s. This stereolithography method has extremely high forming efficiency, and ceramic parts can usually be completed within a few minutes.

[0123] As shown in Figure 13(e), the printed object exhibits an overall appearance as a 2.5D curved ceramic part with a complex freeform surface. The actual object matches the preset model well, and its bending depth changes along the longitudinal direction, conforming to the preset target. The method eliminates the need for additional pre- and post-processing steps to design and remove the printing support rods, greatly simplifying the overall process. Furthermore, the method utilizes light scattering generated by incident light and highly particle-filled ceramic slurry to prepare the ceramic part, demonstrating its programmable design advantage.

[0124] [Example 2] Printing petal-shaped ceramic parts

[0125] To achieve integrated molding of complex irregularly shaped ceramic parts, a 3D modeling software was first used to construct a printing model with a maximum height of 0.5 cm. The model was then arranged in a specific orientation without the need for additional support structures (as shown in Figure 14(a)). Subsequently, the 3D information of the model was decoupled into a series of 2D slice files containing pixel-level grayscale information using Python programming software (as shown in Figure 14(b)). Specifically, the maximum lateral dimension of the ceramic part model in the initial image was set to 10.00 × 10.00 mm, with the grayscale value maintained at 255. Based on a deep learning network algorithm, the 2D graphic corresponding to the boundary contour of the extremely thin part was linearly reduced (divided into 100 steps for smooth display of the transition process) to 1 / 4 of the original area (as shown in Figure 14(b)).

[0126] The power of the fixed light source is 125mW / cm. 2 A commercially available photosensitive ceramic paste was placed in a self-made quartz plate with a 1mm deep groove and a 50×50mm area (as shown in Figure 13(c)). After self-leveling, it was placed on the printing platform of an up-projection DLP printer (as shown in Figure 14(c)). The exposure time of a single grayscale image was set to 0.1s, and the above-mentioned slice dataset was continuously exposed until the end, thus forming a complex free-form thin-walled ceramic part in one go. The printing time was 5 minutes and 30 seconds. The stereolithography method has extremely high forming efficiency, and ceramic parts can usually be completed within a few minutes.

[0127] Figure 14(d) shows a physical image of the printed part obtained in Example 2. The overall appearance is a petal-shaped ceramic part with a large curvature. The physical part matches the preset model well, and the bending depth changes along the longitudinal direction, consistent with the preset target. This method eliminates the need for additional pre- and post-processing steps to design and remove the printing support rods, greatly simplifying the overall process. Furthermore, the method utilizes light scattering generated by incident light and highly particle-filled ceramic slurry to prepare the ceramic part, demonstrating its programmable design advantage.

[0128] [Example 3] Printing of Ceramic Parts for Dental Veneers

[0129] To achieve integrated molding of complex irregularly shaped ceramic parts, a 3D modeling software was first used to construct the printing model, which was then arranged in a specific orientation without the need for additional support structures (as shown in Figure 15(a)). Subsequently, the 3D information of the model was decoupled into a series of 2D slice files containing pixel-level grayscale information using Python programming software (as shown in Figure 15(b)). Specifically, the 2D horizontal dimension of the thin-walled ceramic part model in the initial image was set to 11.86 × 9.8 mm, with the grayscale value linearly varying from 64 from left to right to 255. Keeping this setting unchanged, based on a deep learning network algorithm, the 2D graphic corresponding to the boundary contour of the extremely thin part was linearly shrunk (divided into 100 steps for smooth display of the transition) to 3 / 4 of its original area (as shown in Figures 15(b)(c), where (c) illustrates the linear transformation and grayscale gradient diffusion process of the projected image).

[0130] The power of the fixed light source is 200mW / cm. 2 A commercially available photosensitive ceramic paste was placed in a self-made quartz plate with a 1mm deep and 50×50mm groove (as shown in Figure 13(c)). After self-leveling, it was placed on the printing platform of a top-projection DLP printer (as shown in Figure 22). The exposure time for a single grayscale image was set to 0.1s, and the above-mentioned slice dataset was continuously exposed until the end. Thus, the printing time for forming a complex free-form thin-walled ceramic part in one step was 5 minutes and 30 seconds. This stereolithography method has extremely high forming efficiency, and thin-walled ceramic parts can usually be completed within a few minutes.

[0131] Based on previous experimental results, the three-dimensional information of the model is decoupled. By coupling the relationship between the critical exposure amount and curing depth of the photocurable ceramic slurry, the required optical information is calculated in reverse, thereby obtaining information such as grayscale, light intensity, time and exposure superposition of the target 3D structure. This enables pixel-level light energy control and integrated construction of the three-dimensional morphology of the target part.

[0132] Figure 15(d) shows the physical image of the printed part obtained in Example 3. The overall appearance is a thin-walled ceramic part with a complex free-form surface. The physical part matches the preset model well (as shown in Figure 15(e), which shows the fit between the shaped thin-walled ceramic part and the preset model). It can fit well with the model, and its thickness changes along the longitudinal direction, which is consistent with the preset target (as shown in Figure 15(f), which shows the thickness change of the three-dimensional scanning and forming of the thin-walled ceramic part). Observation of its microstructure revealed that the 3D-forming method produces a relatively uniform grain size distribution in the printed parts. Unlike layer-by-layer deposition methods, this method completely eliminates layer cracks, reduces interlayer defects and weak bonding interfaces, effectively improving manufacturing success rate and part performance (as shown in Figure 15(g), which displays the physical image and 3D morphology of the printed part; the two images in the upper left are 1000x magnified views of the outer and inner sides of the printed part; the image in the upper right is a 5000x magnified view of the cross-section; and the three images at the bottom are 10000x magnified views of the outer, inner, and cross-section of the printed part). Furthermore, this method eliminates the need for additional pre- and post-processing steps to design and remove the printing support rods, greatly simplifying the overall process. In addition, the ceramic parts prepared by this method exhibit gradient thickness, demonstrating its programmable design advantage.

[0133]

Example 4

[0134] Example 4 follows the same steps as Example 3, the only difference being the setting method for the two-dimensional slice image of the printed model. Specifically, the two-dimensional horizontal dimension of the model in the initial image is set to 11.86 × 9.8 mm, with fixed grayscale values ​​(set sequentially to 64, 128, and 160) to obtain a thin-walled ceramic part with uniform thickness. Keeping this setting unchanged, the two-dimensional graphic outline of the thin-walled ceramic part model is linearly reduced to 3 / 4 of its original size and broken down into 100 steps for smoothly displaying the transition process (as shown in Figures 16(a) and 16(b)). Figure 16(a) shows the slice file of the printed model containing pixel-level linear change information of graphic size, and Figure 16(b) shows the linear transformation process of the projected image.

[0135] Figure 16(c) shows a physical image of the printed part obtained in Example 4. Figure 16(d) shows a comparison of the thickness variations of the thin-walled ceramic parts obtained by the 3D scanning stereolithography method in Example 3 (left) and Example 4 (right). The results show that, compared with Example 3, the thin-walled ceramic parts obtained by Example 4 through uniform linear exposure of the projected image have a more uniform thickness, and thin-walled ceramic parts with different uniform thicknesses can be obtained by adjusting the grayscale value.

[0136]

Example 5

[0137] In the control example, the two-dimensional grayscale image of the thin-walled ceramic part model contour was directly exposed without any processing (grayscale value 255) (as shown in Figure 17(a)), and the exposure time was set to 10s. The results showed that the obtained sample differed greatly from the preset model, did not have obvious curing boundaries, and had a very poor matching degree with the preset model (as shown in Figure 17(b)).

[0138] The above results demonstrate that, compared with the method of direct transmission using a binarized image of the shape and structure of the object to be printed, the method of generating grayscale information in this disclosure can better control the printing range, improve printing accuracy, reduce the workload of subsequent model repair, simplify the overall printing process, and improve printing efficiency.

[0139] The different gray levels in the accompanying drawings of this disclosure are for illustrative purposes only and do not provide any additional information.

[0140] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more flowchart illustrations and / or one or more block diagrams.

[0141] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.

[0142] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more flowcharts and / or one or more block diagrams.

[0143] This concludes the detailed description of the present disclosure. To avoid obscuring the concept of the disclosure, some details known in the art have not been described. Those skilled in the art will fully understand how to implement the technical solutions disclosed herein based on the above description.

[0144] The methods and apparatus of this disclosure may be implemented in many ways. For example, they may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order of steps for the methods is for illustrative purposes only, and the steps of the methods of this disclosure are not limited to the order specifically described above unless otherwise specifically stated. Furthermore, in some embodiments, this disclosure may also be implemented as a program recorded on a recording medium, the program including machine-readable instructions for implementing the methods according to this disclosure. Thus, this disclosure also covers recording media storing programs for performing the methods according to this disclosure.

[0145] It should be noted that the terms "first," "second," etc., used in the specification, claims, and drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0146] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this disclosure and not to limit them; although this disclosure has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications can still be made to the specific implementation of this disclosure or equivalent substitutions can be made to some technical features without departing from the spirit of the technical solutions of this disclosure, and all such modifications and substitutions should be covered within the scope of the technical solutions claimed in this disclosure.

Claims

1. A method for processing 3D printing data, comprising: Based on the 3D data of the object to be printed, obtain slice images containing grayscale information; The sliced ​​image is sent to a photocurable ceramic printer, wherein the light source in the photocurable ceramic printer emits light of a fixed intensity to a light adjustment mechanism, and the light adjustment mechanism adjusts the intensity of the light projected onto the printing platform according to the grayscale information in the sliced ​​image, and the surface of the printing platform is located on the upper surface of the photosensitive ceramic paste.

2. The data processing method according to claim 1, wherein, The step of obtaining a slice image containing grayscale information includes: obtaining the grayscale information corresponding to the three-dimensional data based on a machine learning model, and obtaining the slice image including the grayscale information of each pixel.

3. The data processing method according to claim 2, wherein, The machine learning model is generated based on ceramic photocuring samples, wherein the ceramic photocuring samples include the correspondence between grayscale image sample data and the corresponding three-dimensional data of the cured object.

4. The data processing method according to any one of claims 1-3, wherein, The light adjustment mechanism includes a digital micromirror device (DMD), which adjusts the reflection angle of the corresponding pixel's wafer according to the grayscale information to adjust the intensity of the light projected onto the printing platform.

5. The data processing method according to any one of claims 1-4, wherein, The process of obtaining the slice image containing grayscale information includes: Based on the three-dimensional data, obtain the grayscale information corresponding to each pixel, and obtain a first image including the grayscale information; Gradually shrink or expand the two-dimensional graphic corresponding to the boundary contour of the image to a set ratio to generate a set number of second images, which are used as the slice images.

6. The data processing method according to any one of claims 1-5, wherein, The step of obtaining a slice image containing grayscale information based on the three-dimensional data of the object to be printed includes: Obtain model slice data based on the three-dimensional data; Based on the three-dimensional data in each slice data, obtain the slice image containing grayscale information.

7. The data processing method according to claim 6, wherein, The step of obtaining the slice image containing grayscale information based on the three-dimensional data in each slice data includes: For each slice of data, the grayscale information corresponding to each pixel is obtained based on the three-dimensional data, and a first image including the grayscale information is obtained. For each of the first images, the two-dimensional graphics corresponding to the boundary contour of the image are gradually reduced or expanded to a set ratio to generate a set number of second images, which are used as the slice images.

8. The data processing method according to claim 2 or 3, wherein, The light intensity of the light source used to generate the ceramic photocurable sample is the same as the light intensity of the light source in the photocurable ceramic printer; and / or The total printing time for generating the cured object corresponding to each sample data in the ceramic photocuring sample is the same as the printing time for the photocuring ceramic printer to print one slice image.

9. The data processing method according to claim 3 or 8, further comprising: A grayscale value sequence is determined according to a predetermined gradient. For each grayscale value in the grayscale value sequence, grayscale image sample data is generated. Using the photopolymer ceramic printer, the three-dimensional data of the cured object generated by printing the grayscale image sample data is obtained. The correspondence between the grayscale image sample data and the corresponding three-dimensional data of the cured object is determined. The ceramic photocuring sample is obtained based on the mapping relationship between the grayscale image sample data of each grayscale value in the grayscale value sequence and the three-dimensional data of the cured object. A machine learning neural network was built using the ceramic photocured sample to obtain a machine learning model.

10. The data processing method according to any one of claims 1-9, wherein, The object to be printed includes dental veneers.

11. A three-dimensional printing method for a photopolymer ceramic printer, comprising: Obtain slice images, wherein the slice images are generated by the three-dimensional printing data processing method according to any one of claims 1-8; A fixed intensity of light is emitted to the light adjustment mechanism; The light adjustment mechanism adjusts the intensity of the light projected onto the printing platform according to the grayscale information in the sliced ​​image, and the surface of the printing platform is located on the upper surface of the photosensitive ceramic paste.

12. A three-dimensional printing data processing device, comprising: The slice image acquisition unit is configured to acquire a slice image containing grayscale information based on the three-dimensional data of the object to be printed; A data sending unit is configured to send the sliced ​​image to a photocurable ceramic printer, wherein a light source in the photocurable ceramic printer emits light of a fixed intensity to a light adjustment mechanism, and the light adjustment mechanism adjusts the intensity of the light projected onto the printing platform according to the grayscale information in the sliced ​​image, and the surface of the printing platform is located on the upper surface of the photosensitive ceramic paste.

13. The data processing apparatus according to claim 12, further comprising: The sample data acquisition unit is configured to determine a grayscale value sequence according to a predetermined gradient, generate grayscale image sample data for each grayscale value in the grayscale value sequence, acquire the three-dimensional data of the cured object generated by printing the grayscale image sample data using the photocurable ceramic printer, and determine the correspondence between the grayscale image sample data and the corresponding three-dimensional data of the cured object. Based on the correspondence between the grayscale image sample data of each grayscale value and the three-dimensional data of the cured object, the ceramic photocured sample is obtained; The model training unit is configured to train a machine learning neural network using the ceramic photocured samples to obtain the machine learning model. The slice image acquisition unit is configured to acquire the grayscale information corresponding to the three-dimensional data based on a machine learning neural network model, and acquire the slice image including the grayscale information of each pixel.

14. A three-dimensional printing data processing device, comprising: Memory; as well as A processor coupled to the memory, the processor being configured to perform the method as described in any one of claims 1 to 10 based on instructions stored in the memory.

15. A computer-readable storage medium having stored thereon computer instructions that, when executed by a processor, implement the method of any one of claims 1 to 10.

16. A computer program product comprising a computer program or instructions which, when executed by a processor, implement the method of any one of claims 1 to 10.

17. A three-dimensional printer, comprising: A communication mechanism is configured to acquire slice images, wherein the slice images are generated by the three-dimensional printing data processing method according to any one of claims 1-8; The light source is configured to emit light of a fixed intensity to the light adjustment mechanism; A light adjustment mechanism is configured to adjust the intensity of light projected onto the printing platform based on grayscale information in the sliced ​​image; and The printing platform, with its surface located on the upper surface of the photosensitive ceramic paste, is configured to carry the printed object.

18. A three-dimensional printing system, comprising: The three-dimensional printing data processing apparatus according to any one of claims 12-14; And the 3D printer as described in claim 17.

19. A computer program for causing a processor to perform the method according to any one of claims 1 to 10.

Images