Method and device for generating digital model

Abstract

The present invention discloses a method and apparatus for generating a digital model. The method comprises: obtaining overall point cloud data of the object under test based on at least one set of images of the object under test; gridding the overall point cloud data at at least two different resolutions to obtain at least two grid models of the object under test; and synthesizing the at least two grid models to generate an overall grid model with multiple resolutions. The present invention solves the technical problem in the prior art of using split-mode scanning, which reduces the efficiency of scanning the object under test and thus prevents the efficiency of generating a digital model from being improved.

Description

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on May 29, 2020, with application number 2020104767855 and application name “Dental Scanning Method, Device, System and Computer-Readable Storage Medium”, the entire contents of which are incorporated herein by reference.

[0002] At the same time, this application also claims priority to the Chinese patent application filed with the China Patent Office on July 6, 2020, with application number 2020106419086 and application name “Method and Device for Generating Digital Models”, all contents of which are incorporated by reference into this application. Technical Field

[0003] The present invention relates to the field of digital models, and in particular to a method and device for generating digital models. Background Art

[0004] In digital dental prosthesis design, there are two common methods for obtaining digital models of the dental arch: intraoral scanning within the patient's mouth, or scanning a plaster model with a desktop scanner. In these applications, the digital model is often desired to be as detailed as possible to better identify important details, such as the margins of the restored teeth and abutments, and the steps, end faces, and angles of implant stems.

[0005] Digital models are represented using triangular meshes. The minimum pitch of these triangular meshes determines the minimum level of detail that can be expressed. Therefore, the smaller the pitch, the more detailed the features of the actual tooth or restoration. However, for the same dental model, a mesh with a smaller pitch has more 3D vertices and meshes, resulting in a larger data volume. Furthermore, a mesh with a smaller pitch requires a higher-resolution scanner to obtain high-resolution raw point cloud data, which is then processed through a series of time-consuming processes to generate a highly detailed mesh.

[0006] As mentioned above, if you want to obtain a more detailed digital dental model, there will be two problems: 1) the model data volume is large and it is inconvenient to import it into the design software; 2) the processing time of generating the model during scanning increases.

[0007] To address these issues, some existing systems employ a split-mold scanning method. The dental plaster model is split, and the portion requiring restoration is scanned separately to create a fine mesh. The remaining portion is scanned separately to create a coarse mesh. The two meshes are then aligned and stitched together. This ensures detail in the critical restoration areas while maintaining an appropriate overall mesh size.

[0008] Therefore, in the prior art, a scanner is used to obtain a digital model of the object to be measured. Generally, in order to obtain a more refined digital model as a whole, the point spacing of the triangular grid used to represent the digital model is minimized. However, this method has the technical problem of low efficiency in generating the digital model. The prior art can use a split-mold scanning method to solve the above problem. However, during the scanning process, the above method reduces the efficiency of scanning the object to be measured, resulting in a technical problem that the efficiency of generating the digital model cannot be improved.

[0009] To address the above-mentioned problems, no effective solutions have been proposed so far. Summary of the Invention

[0010] The embodiments of the present invention provide a method and apparatus for generating a digital model, so as to at least solve the technical problem in the prior art that the efficiency of generating a digital model cannot be improved due to the reduced efficiency of scanning the object being measured when using split-mode scanning.

[0011] According to one aspect of an embodiment of the present invention, a method for generating a digital model is provided, comprising: obtaining overall point cloud data of the object to be measured based on at least one set of images of the object to be measured; gridding the overall point cloud data at at least two different resolutions to obtain at least two grid models of the object to be measured; and synthesizing the at least two grid models to generate an overall grid model with multiple resolutions.

[0012] Optionally, the overall point cloud data is gridded at at least two different resolutions to obtain at least two grid models of the object to be measured, including: gridding the overall point cloud data at a first resolution to obtain a first grid model of the object to be measured; determining the area to be adjusted in the first grid model; determining and segmenting the original point cloud data falling within the area to be adjusted from the overall point cloud data; gridding the original point cloud data falling within the area to be adjusted at a second resolution to obtain a second grid model of the area to be adjusted, wherein the overall point cloud data includes the point cloud data within the area to be adjusted.

[0013] Optionally, at least two mesh models are synthesized to generate an overall mesh model with multiple resolutions, including: cutting out the original mesh of the area to be adjusted from the first mesh model to generate a mesh model to be stitched; and synthesizing the second mesh model with the mesh model to be stitched to generate an overall mesh model.

[0014] Optionally, the information of the area to be adjusted includes: a range of the area to be adjusted and an area type, wherein the range of the area to be adjusted is used to filter out the area to be adjusted from the entire point cloud data, and the area type is used to determine the second resolution.

[0015] Optionally, when the first resolution is a low resolution, the first grid model is a reconstructed coarse grid model, wherein determining the area to be adjusted in the first grid model includes: using a recognition model to identify feature types from the first grid model, and determining the area to be adjusted based on the identified feature types, wherein the recognition model is a neural network model based on sample training; or, based on the received selection instruction, selecting the area to be adjusted from the first grid model; wherein the area to be adjusted is an area in the first grid model that requires fine gridding processing using high resolution.

[0016] Optionally, the method further includes: determining a non-adjustment area in the first grid model, wherein the grid model of the non-adjustment area is a portion of the first grid model excluding the second grid model.

[0017] Optionally, after determining the area to be adjusted in the first grid model, the method further includes: gridding the original point cloud data falling within the area to be adjusted at a third resolution to obtain a third grid model of the area to be adjusted, wherein the overall point cloud data includes the point cloud data within the area to be adjusted.

[0018] Optionally, in the process of sequentially gridding the entire point cloud data using multiple resolutions, the resolution used each time increases sequentially, and the point cloud data currently being gridded is part of the point cloud data gridded last time.

[0019] According to another aspect of an embodiment of the present invention, another method for generating a digital model is provided, including: obtaining overall point cloud data of the object to be measured based on at least one group of images of the object to be measured; gridding the overall point cloud data at a first resolution to obtain a first grid model of the object to be measured, wherein the first grid model includes: a first area that needs to be gridded again; gridding the original point cloud data in the first area at a second resolution to obtain a second grid model; and synthesizing the second grid model with the grid model of the first grid model except the first area to generate an overall grid model.

[0020] Optionally, when the first resolution is low resolution and the second resolution is high resolution, the first grid model is a coarse grid model and the second grid model is a fine grid model.

[0021] According to another aspect of an embodiment of the present invention, another method for generating a digital model is provided, including: obtaining overall point cloud data of the object to be measured based on multiple frames of images of the object to be measured; gridding the overall point cloud data to obtain a first grid model of the object to be measured; identifying the area to be adjusted in the first grid model; obtaining original point cloud data falling within the area to be adjusted from the overall point cloud data; performing high-resolution fine gridding on the original point cloud data to generate a second grid model of the area to be adjusted; and replacing the original grid model of the area to be adjusted in the first grid model with the second grid model.

[0022] According to one aspect of an embodiment of the present invention, a device for generating a digital model is also provided, including: a first acquisition module, used to obtain overall point cloud data of the object to be measured based on at least one group of images of the object to be measured; a first processing module, used to grid the overall point cloud data with at least two different resolutions to obtain at least two grid models of the object to be measured; and a first generation module, used to synthesize at least two grid models to generate an overall grid model with multiple resolutions.

[0023] According to another aspect of an embodiment of the present invention, a non-volatile storage medium is provided. The non-volatile storage medium includes a stored program, wherein when the program is running, a method for controlling any device where the non-volatile storage medium is located to generate a digital model is provided.

[0024] According to another aspect of an embodiment of the present invention, a processor is further provided. The processor is configured to run a program, wherein the program executes any one of the methods for generating a digital model when the program is run.

[0025] In an embodiment of the present invention, a method of scanning the object to be measured without mold splitting is adopted, and the overall point cloud data of the object to be measured is obtained based on at least one group of images of the object to be measured; the overall point cloud data is gridded with at least two different resolutions to obtain at least two grid models of the object to be measured; the at least two grid models are synthesized to generate an overall grid model with multiple resolutions, thereby achieving the purpose of generating a grid model corresponding to the object to be measured without mold splitting scanning, thereby realizing the technical effect of quickly generating a grid model, and further solving the technical problem in the prior art that the efficiency of generating a digital model cannot be improved due to the reduced efficiency of scanning the object to be measured when using mold splitting scanning. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0027] Figure 1is a flow chart of a method for generating a digital model according to an embodiment of the present invention;

[0028] Figure 2 is a framework diagram of an optional process for generating a high-resolution fine grid of a feature area according to an embodiment of the present invention;

[0029] Figure 3 This is a flowchart of an optional process for determining feature region information in a coarse grid model according to an embodiment of the present invention;

[0030] Figure 4 This is a framework diagram of an optional process for generating a coarse grid according to an embodiment of the present invention;

[0031] Figure 5 This is a flowchart of an optional process for generating a non-feature area grid according to an embodiment of the present invention;

[0032] Figure 6 is another optional process framework diagram for generating a non-feature area grid according to an embodiment of the present invention;

[0033] Figure 7 is a flowchart of an optional process for generating a resolution grid according to an embodiment of the present invention;

[0034] Figure 8 is a flow chart of another method for generating a digital model according to an embodiment of the present invention;

[0035] Figure 9 is a flow chart of another method for generating a digital model according to an embodiment of the present invention;

[0036] Figure 10 is a schematic structural diagram of a device for generating a digital model according to an embodiment of the present invention;

[0037] Figure 11 is a schematic structural diagram of another device for generating a digital model according to an embodiment of the present invention;

[0038] Figure 12 2 is a schematic structural diagram of another device for generating a digital model according to an embodiment of the present invention. DETAILED DESCRIPTION

[0039] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0040] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes 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 devices.

[0041] According to an embodiment of the present invention, an embodiment of a method for generating a digital model is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0042] Figure 1 is a method for generating a digital model according to an embodiment of the present invention, such as Figure 1 As shown, the method includes the following steps:

[0043] Step S102, obtaining overall point cloud data of the object under test based on at least one set of images of the object under test;

[0044] Step S104, gridding the entire point cloud data at at least two different resolutions to obtain at least two grid models of the object being measured;

[0045] Step S106: synthesize at least two mesh models to generate an overall mesh model with multiple resolutions.

[0046] In the above-mentioned method for generating a digital model, first, the overall point cloud data of the object to be measured can be obtained based on at least one group of images of the object to be measured; then, the overall point cloud data is gridded with at least two different resolutions to obtain at least two grid models of the object to be measured; finally, the at least two grid models are synthesized to generate an overall grid model with multiple resolutions, thereby achieving the purpose of generating a grid three-dimensional model corresponding to the object to be measured without using split-mold scanning, thereby realizing the technical effect of quickly generating a grid three-dimensional model, and further solving the technical problem in the prior art that the efficiency of generating a digital model cannot be improved due to the reduced efficiency of scanning the object to be measured when using split-mold scanning.

[0047] It should be noted that multi-resolution processing is based on actual needs, using different resolutions for different areas. The two different resolutions mentioned above can be relative to the default resolution. A higher resolution than the default resolution is considered a higher resolution, and a lower resolution than the default resolution is considered a lower resolution. Generally, the default resolution is set to a lower resolution to achieve faster scanning speeds and smooth real-time imaging.

[0048] Specifically, in the above step S102, a scanner device can be used to scan the image of the object to be measured, for example, a three-dimensional scanner can be used, and then each group of measured images is three-dimensionally reconstructed to form original single-piece point cloud data. A piece of original single-piece point cloud data is formed in each group of measured images. Multiple groups of measured object images can form multiple pieces of original point cloud data. Multiple pieces of original single-piece point cloud data can be spliced ​​into overall point cloud data after being processed in a unified coordinate system, thereby obtaining the overall point cloud data of the measured object.

[0049] It is easy to notice that the above-mentioned scanner equipment needs to have a higher image resolution, that is, it can obtain high-precision original single-piece point cloud data through scanning and acquisition. For example, an intraoral scanner with a pixel greater than or equal to 300,000 or a desktop scanner with a pixel greater than or equal to 1.3 million can be used.

[0050] It should be noted that the above-mentioned original single-piece point cloud data refers to the point cloud data generated by a single acquisition from a single perspective. The scanning equipment can scan images of dentures, oral cavity, etc.

[0051] In the above step S104, in order to obtain at least two grid models of the object to be measured, in an optional embodiment of the present application, the overall point cloud data can be gridded at a first resolution to obtain a first grid model of the object to be measured; then, the area to be adjusted in the first grid model is determined; then, the original point cloud data falling within the area to be adjusted is determined and segmented from the overall point cloud data; finally, the original point cloud data falling within the area to be adjusted is gridded at a second resolution to obtain a second grid model of the area to be adjusted. It should be noted that the above-mentioned overall point cloud data includes the point cloud data within the area to be adjusted.

[0052] Taking the scanning rod area (i.e. the area where the scanning rod is located) as the first grid area as an example, the scanning rod is a standard body accessory of the 3D scanner used to scan dental models. The scanning rod can be inserted into the jaw. By scanning the standard scanning rod on the jaw, the position of the implant can be better located.

[0053] Specifically, the dental identification data is analyzed to obtain a tooth region, restoration type information, a gum region, and a base region; a scanbar region is obtained based on the tooth region, the gum region, and the base region; and first mesh data corresponding to the scanbar region in the coarse-mesh three-dimensional tooth model is separated and extracted based on the scanbar region. In one embodiment, the 3D scanner predetermines the presence of a scanbar region in the coarse-mesh three-dimensional tooth model, and the first mesh data corresponding to the scanbar region in the coarse-mesh three-dimensional tooth model is separated and extracted based on the scanbar region.

[0054] In an optional embodiment of the present application, in order to perform gridding more quickly, when the overall point cloud data is gridded at a first resolution, the gridding method adopted is incremental gridding, that is, when a part of the original point cloud is obtained by scanning, the gridding of the existing part of the original points is started, and then in the subsequent scanning process, the original point cloud generated by the new scan is gridded in an incremental manner on the basis of the previous grid to obtain a first grid model of the object to be measured, and a general gridding process, that is, non-incremental gridding process, can be used for the area to be adjusted.

[0055] In an optional embodiment of the present application, the area to be adjusted may be a feature area, such as Figure 2As shown, an embodiment of the present invention also provides a process framework diagram for generating a high-resolution fine grid of a feature area. In this process, a high-resolution fine grid of a feature area can be generated in the following manner: first, the area of ​​all original single-piece point cloud data is screened, that is, from all original single-piece point cloud data after global optimization processing, the part that falls within the spatial range of the feature area is selected and segmented; then, these segmented original point cloud data are used for high-resolution fine grid processing, the original point cloud is fused based on voxels, and the resolution size is determined by setting the voxel size. The larger the voxel, the lower the resolution. Then, based on the implicit surface of the object model, a grid is extracted to obtain a grid model of the object being measured. Then, after a series of feature protection grid processing, the final fine grid model is obtained. The above process can also be generated by stitching a single grid to generate the final fine grid model. After multiple point clouds are de-overlapped, they are meshed separately to obtain a single grid, and then the multiple grids are synthesized. Since the feature area is generally only a small part of the entire jaw, the time for high-resolution (small point pitch) processing of the local area can be effectively controlled.

[0056] It should be noted that global optimization refers to the unified determination of the optimal spatial positions of all single point clouds, and fusion processing refers to merging independent single point clouds into a unified grid model. Feature protection methods include but are not limited to: bilateral filtering, flanging processing, etc. At the same time, it is easy to notice that the smaller the point pitch of the grid model, the higher the resolution.

[0057] After obtaining at least two mesh models of the object to be measured, in order to obtain an overall mesh model, the at least two mesh models are synthesized to generate an overall mesh model with multiple resolutions. In some optional embodiments of the present application, the original mesh of the area to be adjusted can be first cut out from the first mesh model to generate a mesh model to be stitched, and then the second mesh model processed with the second resolution is synthesized with the mesh model to be stitched to generate an overall mesh model. The synthesis method includes but is not limited to: using a stitching algorithm to synthesize the two mesh models to obtain an overall mesh model.

[0058] It should be noted that the information of the above-mentioned area to be adjusted includes: the range of the area to be adjusted and the area type, wherein the range of the area to be adjusted is used to filter out the area to be adjusted from the overall point cloud data, and the area type is used to determine the second resolution. Specifically, in order to determine the position information of the area to be adjusted, it can refer to using the point cloud corresponding to the area to be adjusted to represent the distribution range of its characteristic area, or it can use a regular three-dimensional bounding box to represent the distribution range corresponding to its characteristic area. The range of the above-mentioned area to be adjusted can be used to determine the range of fine grid processing, and the type of characteristic area can be further used to determine the resolution of fine grids in different areas according to application requirements.

[0059] In order to be applicable to different application scenarios, the methods used in some embodiments of the present application to generate digital models are also different. For example, when scanning the mouth, in order to make local details more prominent, the strategy adopted is to increase the resolution of the feature area. Specifically, when the handheld scanner performs intraoral scanning, the overall point cloud data is first gridded at a default resolution to obtain a first grid model, and then the feature area is gridded at a higher resolution to obtain a second grid model, that is, the first grid model is a coarse grid model with a lower resolution relative to the second grid model, and the second grid model is a fine grid model with a higher resolution relative to the first grid model. The first grid model and the second grid model are then synthesized to obtain a multi-resolution overall grid model, that is, an overall grid model with both lower and higher resolutions.

[0060] In some optional embodiments of the present application, when the first resolution is low resolution, the first grid model is a reconstructed coarse grid model, and the following method can be used to determine the area to be adjusted in the first grid model. For example, a recognition model can be used to identify feature types from the first grid model, and the area to be adjusted is determined based on the identified feature types, wherein the recognition model is a neural network model based on sample training; or, based on the received selection instruction, the area to be adjusted is selected from the first grid model to determine the area to be adjusted in the first grid model; it should be noted that the area to be adjusted is the area in the first grid model that requires fine gridding processing using high resolution (small dot pitch).

[0061] For example, for intraoral scanning, feature types include: teeth, gums, tooth preparations, scanning rods, etc. In this application scenario, in order to make local details more prominent, two-level resolution is usually used. A first-level resolution grid is formed during real-time scanning. After the scan is completed, a second-level resolution grid is generated for the points corresponding to feature areas such as tooth preparations and scanning rods. The second-level resolution is higher than the first-level resolution and is used to improve the resolution of areas such as tooth preparations and scanning rods. If three-level resolution is used, feature areas such as tooth preparations and scanning rods use high resolution, feature areas such as teeth use medium resolution, and non-feature areas such as gums use low resolution, that is, the default resolution.

[0062] For example, for denture scanning, the feature types include one of the following: teeth, gums, prepared teeth, abutments, inlays, and scanning rods. A first-level resolution grid can be generated during real-time scanning. After the scan is completed, the points corresponding to the feature areas such as prepared teeth, abutments, inlays, and scanning rods can generate a second-level resolution grid. The second-level resolution is higher than the first-level resolution. Similarly, three levels of resolution can be used to process the feature types in the denture. Specifically, high resolution is used for feature areas such as prepared teeth, abutments, inlays, and scanning rods, medium resolution is used for feature areas such as teeth, and low resolution, that is, the default resolution, is used for non-feature areas such as gums and others. It is easy to notice that more high and low resolutions can be used to process feature types according to the actual application environment.

[0063] In an optional embodiment of the present application, the area to be adjusted may be a feature area, such as Figure 3 As shown, an embodiment of the present invention provides a process framework diagram for determining feature area information in a coarse grid model. Specifically, for the coarse grid model, automatic recognition can be used first, for example, based on AI training, feature pattern matching, etc., to obtain feature areas; feature areas can also be manually selected or modified, or a combination of the two can be used, that is, first automatically identified and displayed, and then manually modified and confirmed through interaction; the generated feature area information mainly includes the range and feature type of each adjustment area, wherein the feature type includes but is not limited to: ordinary teeth, teeth that need to be restored and designed, and abutments. It should be noted that the coarse grid model can be referred to as a coarse grid model.

[0064] According to an optional embodiment of the present application, a process framework diagram for generating a coarse grid is also provided, such as Figure 4 As shown in the figure: After image acquisition and 3D reconstruction, a single frame of point cloud data is obtained, and then a fast incremental gridding method is used to obtain a coarse grid. It should be noted that incremental gridding refers to the synchronous gridding process during the scanning process, and the efficiency of the processing is improved through incremental iterative processing. Specifically, only the newly scanned single point cloud data is optimized each time to determine the optimal spatial position of the single point cloud. The independent single point clouds are merged into a unified lower-resolution grid model, and then the latest grid is obtained from the grid model. In this way, after the scan is completed, a grid model of the object under test can be obtained. By smoothing and filtering the noise of this model, a coarse grid model can be obtained.

[0065] In some optional embodiments of the present application, if the first grid model can meet actual application requirements, the non-adjustment area in the first grid model can be directly determined based on the area to be adjusted. It is easy to notice that the grid model of the non-adjustment area is the part of the first grid model other than the second grid model.

[0066] In an optional embodiment of the present application, the area to be adjusted may be a characteristic area, and the non-adjustment area may be a non-characteristic area, such as Figure 5 As shown, a process framework diagram for generating a non-feature area grid is also provided. In this process, first, the feature area in the coarse grid can be directly cut off according to the area information of the feature area, and then the non-feature area grid is obtained.

[0067] In some optional embodiments of the present application, when there are multiple areas to be adjusted, after the original point cloud data within the areas to be adjusted are gridded at a second resolution, for other areas to be adjusted except the areas to be adjusted processed at the second resolution, the original point cloud data falling within the other areas to be adjusted can be gridded at a third resolution to obtain a third grid model of the other areas to be adjusted, wherein the overall point cloud data includes the point cloud data within the other areas to be adjusted. It is easy to note that the above-mentioned other areas to be adjusted can also be processed at different resolutions.

[0068] In some embodiments of the present application, if the first grid model cannot meet the actual application requirements, for example, the resolution is too low, then after determining the area to be adjusted in the first grid model, the non-adjustment area in the first grid model can be generated by the following steps: reading the overall point cloud data, fusing the overall point cloud data, and generating an overall grid with a fourth resolution; based on the information of the area to be adjusted, cutting out the determined area to be adjusted from the low-resolution overall grid to generate the non-adjustment area in the first grid model, wherein the information of the area to be adjusted includes: the range and area type of the area to be adjusted, and the grid model of the non-adjustment area is the part of the first grid model other than the second grid model.

[0069] For example, in the application of generating digital dental models based on high and low resolution processing, it can be further extended to a scheme of processing three or more resolutions of high, medium and low. In an optional embodiment of the present application, in the process of gridding the overall point cloud data in sequence using multiple resolutions, the resolution used each time can be increased in sequence, and the point cloud data currently being gridded is part of the point cloud data of the last gridding process. Through this multi-resolution processing method, the details of important feature areas can be best reflected, and the data size of the overall dental model can be better controlled; and the speed of processing and generating the overall dental mesh model is effectively improved, and there is no need to increase the number and steps of scanning, thereby greatly saving data processing time.

[0070] In an optional embodiment of the present application, the area to be adjusted may be a characteristic area, and the area not to be adjusted may be a non-characteristic area, such as Figure 6As shown, another process framework diagram for generating a mesh for a non-featured area is also provided. In this process, the original single-piece point cloud data is first used to regenerate a mesh model. Specifically, a relatively low-resolution mesh model is used to fuse all the single-piece point cloud data. A low-resolution mesh is then extracted from the mesh model. After simple smoothing, a low-resolution mesh model is obtained. Then, the feature area is removed, and finally, the mesh for the non-featured area is obtained. Although this method regenerates the mesh model for the entire area, due to the use of lower-resolution meshing and simpler processing, it still saves a lot of processing time compared to the overall high-resolution (small point pitch) fine meshing process.

[0071] In an optional embodiment of the present application, Figure 7 As shown, a process framework diagram for generating a resolution grid is also provided. In this process, first, the non-feature areas and feature areas can be determined by the above-mentioned optional embodiments, and then the multi-resolution grids are synthesized to finally obtain a multi-resolution overall grid model.

[0072] Figure 8 is another method for generating a digital model according to an embodiment of the present invention, such as Figure 2 As shown, the method includes the following steps:

[0073] S202, acquiring overall point cloud data of the object under test based on at least one set of images of the object under test;

[0074] S204, gridding the entire point cloud data at a first resolution to obtain a first grid model of the object to be measured, wherein the first grid model includes: a first area that needs to be gridded again;

[0075] S206, performing gridding processing on the original point cloud data in the first area at a second resolution to obtain a second grid model;

[0076] S208 : Synthesize the second grid model with the grid model of the first grid model except for the first region to generate an overall grid model.

[0077] In the above-mentioned method for generating a digital model, first, the overall point cloud data of the object to be measured can be obtained based on at least one group of images of the object to be measured; then, the overall point cloud data is gridded at a first resolution to obtain a first grid model of the object to be measured, wherein the first grid model includes: a first area that needs to be gridded again; secondly, the original point cloud data in the first area is gridded at a second resolution to obtain a second grid model; finally, the second grid model is synthesized with the grid model of the first grid model except the first area to generate an overall grid model, thereby achieving the purpose of generating a grid three-dimensional model corresponding to the object to be measured without using split-mold scanning, thereby realizing the technical effect of quickly generating a grid three-dimensional model, and further solving the technical problem in the prior art that the efficiency of generating a digital model cannot be improved due to the reduced efficiency of scanning the object to be measured when using split-mold scanning.

[0078] It should be noted that, when the first resolution is low resolution, the second resolution is high resolution (small dot pitch), the first grid model is a coarse grid model, and the second grid model is a fine grid model.

[0079] Figure 9 is another method for generating a digital model according to an embodiment of the present invention, such as Figure 3 As shown, the method includes the following steps

[0080] S302, acquiring overall point cloud data of the object under test based on the collected multiple frames of images of the object under test;

[0081] S304, performing grid processing on the entire point cloud data to obtain a first grid model of the object to be measured;

[0082] S306, identifying the area to be adjusted in the first grid model;

[0083] S308, obtaining original point cloud data falling within the area to be adjusted from the entire point cloud data;

[0084] S310, performing high-resolution fine meshing processing on the original point cloud data to generate a second mesh model of the area to be adjusted;

[0085] S312: Use the second grid model to replace the original grid model of the area to be adjusted in the first grid model.

[0086] In the above-mentioned method for generating a digital model, first, the overall point cloud data of the object to be measured can be obtained based on the collected multi-frame images of the object to be measured; secondly, the overall point cloud data can be gridded to obtain a first grid model of the object to be measured; then, the area to be adjusted in the first grid model is identified; then, the original point cloud data falling in the area to be adjusted is obtained from the overall point cloud data, and the original point cloud data is subjected to high-resolution fine gridding processing to generate a second grid model of the area to be adjusted; finally, the second grid model is used to replace the original grid model of the area to be adjusted in the first grid model, thereby achieving the purpose of generating a grid three-dimensional model corresponding to the object to be measured without using split-mold scanning, thereby realizing the technical effect of quickly generating a grid three-dimensional model, and further solving the technical problem in the prior art that the efficiency of generating a digital model cannot be improved due to the reduced efficiency of scanning the object to be measured when using split-mold scanning.

[0087] Figure 10 According to another aspect of an embodiment of the present invention, a device for generating a digital model is provided, such as Figure 10 As shown, the device includes:

[0088] A first acquisition module 10 is configured to acquire overall point cloud data of the object under test based on at least one set of images of the object under test;

[0089] A first processing module 12 is configured to perform gridding processing on the entire point cloud data at at least two different resolutions to obtain at least two grid models of the object being measured;

[0090] The first generating module 14 is configured to synthesize at least two mesh models to generate an overall mesh model with multiple resolutions.

[0091] The above-mentioned device for generating a digital model includes: a first acquisition module 10, a first processing module 12, and a first generation module 14; wherein the first acquisition module 10 is used to obtain the overall point cloud data of the object to be measured based on at least one group of images of the object to be measured; the first processing module 12 is used to grid the overall point cloud data with at least two different resolutions to obtain at least two grid models of the object to be measured; the first generation module 14 is used to synthesize at least two grid models to generate an overall grid model with multiple resolutions, thereby achieving the purpose of generating a grid three-dimensional model corresponding to the object to be measured without using split-mold scanning, thereby realizing the technical effect of quickly generating a grid three-dimensional model, and further solving the technical problem in the prior art that the efficiency of generating a digital model cannot be improved due to the reduced efficiency of scanning the object to be measured when using split-mold scanning.

[0092] Figure 11 According to another aspect of an embodiment of the present invention, there is provided another device for generating a digital model, such as Figure 11 The device comprises:

[0093] A second acquisition module 20 is configured to acquire overall point cloud data of the object under test based on at least one set of images of the object under test;

[0094] The second processing module 22 is configured to perform gridding processing on the entire point cloud data at a first resolution to obtain a first grid model of the object to be measured, wherein the first grid model includes: a first area that needs to be gridded again;

[0095] A third processing module 24 is configured to perform gridding processing on the original point cloud data in the first area at a second resolution to obtain a second grid model;

[0096] The second generating module 26 is configured to synthesize the second grid model with the grid model of the first grid model except the first region to generate an overall grid model.

[0097] The above-mentioned device for generating a digital model includes: a second acquisition module 20, a second processing module 22, a third processing module 24, and a second generation module 26; wherein the second acquisition module 20 is used to obtain the overall point cloud data of the object to be measured based on at least one group of images of the object to be measured; the second processing module 22 is used to grid the overall point cloud data at a first resolution to obtain a first grid model of the object to be measured, wherein the first grid model includes: a first area that needs to be gridded again; the third processing module 24 is used to grid the original point cloud data in the first area at a second resolution to obtain a second grid model; the second generation module 26 is used to synthesize the second grid model with the grid model of the first grid model except the first area to generate an overall grid model, thereby achieving the purpose of generating a grid three-dimensional model corresponding to the object to be measured without using split-mold scanning, thereby achieving the technical effect of quickly generating a grid three-dimensional model, and thus solving the technical problem in the prior art that the efficiency of generating a digital model cannot be improved due to the reduced efficiency of scanning the object to be measured when using split-mold scanning.

[0098] Figure 12 According to another aspect of an embodiment of the present invention, a device for generating a digital model is provided, such as Figure 12 As shown, the device includes:

[0099] A third acquisition module 30 is used to acquire overall point cloud data of the object under test based on the collected multiple frames of images of the object under test;

[0100] The fourth processing module 32 is used to perform gridding processing on the entire point cloud data to obtain a first grid model of the object under test;

[0101] an identification module 34 for identifying a region to be adjusted in the first grid model;

[0102] A fourth acquisition module 36 is configured to acquire original point cloud data falling within the area to be adjusted from the entire point cloud data;

[0103] The fifth processing module 38 is used to perform high-resolution fine meshing processing on the original point cloud data to generate a second mesh model of the area to be adjusted;

[0104] The replacement module 40 is configured to replace the original grid model of the area to be adjusted in the first grid model with the second grid model.

[0105] The device for generating a digital model includes: a third acquisition module 30, a fourth processing module 32, a recognition module 34, a fourth acquisition module 36, a fifth processing module 38, and a replacement module 40; wherein the third acquisition module 30 is used to acquire the overall point cloud data of the object under test based on the collected multi-frame images of the object under test; the fourth processing module 32 is used to grid the overall point cloud data to obtain a first grid model of the object under test; the recognition module 34 is used to identify the area to be adjusted in the first grid model; the fourth acquisition module 36 is used to acquire the area to be adjusted from the overall point cloud data. The original point cloud data in the area; the fifth processing module 38 is used to perform high-resolution fine gridding processing on the original point cloud data to generate a second grid model of the area to be adjusted; the replacement module 40 is used to replace the original grid model of the area to be adjusted in the first grid model with the second grid model, so as to achieve the purpose of generating a grid three-dimensional model corresponding to the measured object without using split-mold scanning, thereby realizing the technical effect of quickly generating a grid three-dimensional model, and further solving the technical problem in the prior art that the efficiency of generating a digital model cannot be improved due to the reduced efficiency of scanning the measured object when using split-mold scanning.

[0106] According to another aspect of an embodiment of the present invention, a non-volatile storage medium is provided. The non-volatile storage medium includes a stored program, wherein when the program is running, a method for controlling any device where the non-volatile storage medium is located to generate a digital model is provided.

[0107] Specifically, the above-mentioned storage medium is used to store program instructions for executing the following functions to achieve the following functions:

[0108] Based on at least one set of images of the object to be measured, overall point cloud data of the object to be measured is obtained; the overall point cloud data is gridded at at least two different resolutions to obtain at least two grid models of the object to be measured; and the at least two grid models are synthesized to generate an overall grid model with multiple resolutions.

[0109] According to another aspect of an embodiment of the present invention, a processor is further provided. The processor is configured to run a program, wherein the program executes any one of the methods for generating a digital model when the program is run.

[0110] Specifically, the processor is used to call program instructions in the memory to implement the following functions:

[0111] Based on at least one set of images of the object to be measured, overall point cloud data of the object to be measured is obtained; the overall point cloud data is gridded at at least two different resolutions to obtain at least two grid models of the object to be measured; and the at least two grid models are synthesized to generate an overall grid model with multiple resolutions.

[0112] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0113] In the above embodiments of the present invention, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0114] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of the units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0115] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0116] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0117] 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 invention, 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. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk, etc. Various media that can store program codes.

[0118] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for generating a digital model, characterized in that: include: Based on at least one set of images of the object to be measured, obtaining overall point cloud data of the object to be measured; The overall point cloud data is gridded at at least two different resolutions to obtain at least two grid models of the object to be measured, including: gridding the overall point cloud data at a first resolution to obtain a first grid model of the object to be measured; determining an area to be adjusted in the first grid model; and generating a second grid model of a second resolution corresponding to the area to be adjusted based on information about the area to be adjusted; wherein the grid model is a digital model of the dental jaw; the information about the area to be adjusted includes: a range and an area type of the area to be adjusted, the range of the area to be adjusted being used to filter out the area to be adjusted from the overall point cloud data, and the area type of the area to be adjusted being used to filter out the area to be adjusted from the overall point cloud data. The domain type is used to determine the second resolution; determining the area to be adjusted in the first grid model includes: using a recognition model to identify a feature type from the first grid model, and determining the area to be adjusted based on the identified feature type, wherein the recognition model is a neural network model based on sample training; or, based on a received selection instruction, selecting the area to be adjusted from the first grid model; wherein the area to be adjusted is an area in the first grid model that requires fine grid processing using high resolution; when the entire point cloud data is gridded for the first time, the gridding process is performed using incremental gridding; The at least two grid models are synthesized to generate an overall grid model with multiple resolutions.

2. The method according to claim 1, characterized in that Gridding the entire point cloud data at at least two different resolutions to obtain at least two grid models of the object under test, including: Determining and segmenting original point cloud data falling within the area to be adjusted from the entire point cloud data; The gridding process is performed on the original point cloud data falling within the area to be adjusted at a second resolution to obtain a second grid model of the area to be adjusted, wherein the overall point cloud data includes the point cloud data within the area to be adjusted.

3. The method according to claim 2, characterized in that Combining the at least two grid models to generate an overall grid model with multiple resolutions includes: Cutting out the original mesh of the area to be adjusted from the first mesh model to generate a mesh model to be synthesized; The second grid model is synthesized with the grid model to be synthesized to generate the overall grid model.

4. The method according to claim 2, characterized in that The non-adjustment area of ​​the first grid model is determined by: Reading the overall point cloud data; performing fusion processing on the overall point cloud data to generate an overall grid with a fourth resolution; and cutting out the determined area to be adjusted from the overall grid with a low resolution based on information of the area to be adjusted; A non-adjusted area in the first grid model is generated.

5. The method according to claim 2, characterized in that When the first resolution is low resolution, the first grid model is a reconstructed coarse grid model.

6. The method according to claim 2, characterized in that The method further comprises: A non-adjustment area in the first grid model is determined, wherein the grid model of the non-adjustment area is a portion of the first grid model excluding the second grid model.

7. The method according to claim 2, characterized in that After determining the area to be adjusted in the first grid model, the method further includes: The gridding process is performed on the original point cloud data falling within the area to be adjusted at a third resolution to obtain a third grid model of the area to be adjusted, wherein the overall point cloud data includes the point cloud data within the area to be adjusted.

8. The method according to claim 1, characterized in that In the process of sequentially gridding the entire point cloud data using multiple resolutions, the resolution used each time increases sequentially, and the point cloud data currently being gridded is part of the point cloud data gridded last time.

9. The method according to claim 1, characterized in that Acquiring overall point cloud data of the object under test based on at least one set of images of the object under test, including: Scanning the object to be measured by a scanner device; Reconstruct each set of scanned object images to form original point cloud data through three-dimensional reconstruction, wherein each set of object images forms an original single piece of point cloud data, and multiple sets of object images form multiple pieces of original point cloud data; The plurality of original point cloud data are processed in a unified coordinate system and then spliced ​​together to obtain the overall point cloud data.

10. The method according to claim 9, characterized in that The original point cloud data includes: point cloud data generated by a single acquisition from a single perspective.

11. The method according to claim 2, characterized in that The second resolution is greater than the first resolution.

12. The method according to claim 2, characterized in that Gridding the entire point cloud data at a first resolution to obtain a first grid model of the object under test includes: When part of the original point cloud is obtained by scanning, the existing part of the original point cloud is meshed; In the subsequent scanning process, the original point cloud newly generated by the scanning is meshed on the basis of the previous mesh to obtain a first mesh model of the object to be measured.

13. The method according to claim 2, characterized in that When the first resolution is a low resolution, the first grid model is a reconstructed coarse grid model; The step of determining the area to be adjusted in the first grid model includes: Using a recognition model to identify feature types from the first grid model, and determining the area to be adjusted based on the identified feature types, wherein the recognition model is a neural network model based on sample training; or Based on the received selection instruction, the area to be adjusted is selected from the first grid model, wherein the area to be adjusted is an area in the first grid model that requires fine gridding processing using high resolution.

14. A method for generating a digital model, characterized in that: include: Based on at least one set of images of the object to be measured, obtaining overall point cloud data of the object to be measured; Meshing the entire point cloud data at a first resolution to obtain a first mesh model of the object under test, wherein the first mesh model includes a first region that needs to be meshed again, wherein when meshing the entire point cloud data for the first time is performed, the meshing is performed using incremental meshing; Performing the gridding process on the original point cloud data within the first area at a second resolution to obtain a second grid model; synthesizing the second grid model with the grid model of the first grid model except the first region to generate an overall grid model; Wherein, the overall mesh model is a digital model of the dental jaw; the information of the first area includes: the range and area type of the first area, wherein the range of the first area is used to filter out the first area from the overall point cloud data, and the area type is used to determine the second resolution; determining the first area in the first mesh model includes: using a recognition model to identify the feature type from the first mesh model, and determining the first area based on the identified feature type, wherein the recognition model is a neural network model based on sample training; or, based on the received selection instruction, selecting the first area from the first mesh model; wherein the first area is an area in the first mesh model that requires fine mesh processing with high resolution;.

15. The method according to claim 14, characterized in that In the case where the first resolution is a low resolution, the second resolution is a high resolution, the first grid model is a coarse grid model, and the second grid model is a fine grid model.

16. The method according to claim 14, characterized in that The first area includes: an area to be adjusted; integrating the second grid model with the grid model of the first grid model except the first area to generate an overall network model includes: The second grid model is used to replace the original grid model of the area to be adjusted in the first grid model.

17. A device for generating a digital model, characterized in that: include: A first acquisition module is configured to acquire overall point cloud data of the object under test based on at least one set of images of the object under test; The first processing module is used to grid the overall point cloud data at at least two different resolutions to obtain at least two grid models of the object to be measured, including: gridding the overall point cloud data at a first resolution to obtain a first grid model of the object to be measured; determining the area to be adjusted in the first grid model; generating a second grid model of a second resolution corresponding to the area to be adjusted based on the information of the area to be adjusted; wherein the grid model is a digital model of the dental jaw; the information of the area to be adjusted includes: the range and area type of the area to be adjusted, wherein the range of the area to be adjusted is used to filter out the area to be adjusted from the overall point cloud data. domain, the region type is used to determine the second resolution; the determining of the region to be adjusted in the first grid model includes: using a recognition model to identify a feature type from the first grid model, and determining the region to be adjusted based on the identified feature type, wherein the recognition model is a neural network model based on sample training; or, based on a received selection instruction, selecting the region to be adjusted from the first grid model; wherein the region to be adjusted is a region in the first grid model that requires fine grid processing using high resolution; wherein, when the entire point cloud data is gridded for the first time, the gridding process adopts incremental gridding; The first generation module is used to synthesize the at least two grid models to generate an overall grid model with multiple resolutions.

18. A non-volatile storage medium, characterized in that: The non-volatile storage medium includes a stored program, wherein when the program is executed, the device where the non-volatile storage medium is located is controlled to execute the method for generating a digital model according to any one of claims 1 to 16.

19. A processor, characterized in that: The processor is configured to run a program, wherein the program, when running, executes the method for generating a digital model according to any one of claims 1 to 16.

Images