Information processor and information processing method and program

Abstract

To provide a tooth segmentation mechanism with improved robustness against noise.SOLUTION: An information processor comprises a processing part that extracts a tooth area from three dimensional data by sequentially performing preprocessing that modifies the three dimensional data corresponding to an oral shape, segmentation processing that executes clustering using the area expansion method multiple times on the three dimensional data after the preprocessing, and post-processing that adjusts the clusters after the segmentation processing.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] In recent years, dental prostheses have become smarter in their production. For example, dental prostheses are designed and produced using 3D (dimensional) cameras, CT (Computed Tomography) scanners, CAD / CAM (Computer-Aided Design / Computer-Aided Manufacturing) devices, and the like.

[0003] A technology called tooth segmentation, which extracts teeth from three-dimensional data showing the shape of a patient's oral cavity, plays an important role in the production of such smart dental prostheses. As disclosed in Non-Patent Document 1, a technology for further improving the accuracy of tooth segmentation has been developed. [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] Kan Wu, Li Chen, Jing Li, Yanheng Zhou, “Tooth segmentation on dental meshes using morphologic skeleton,” Computers & Graphics, Volume 38, February 2014, Pages 199-211 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the technology disclosed in the above-mentioned Non-Patent Document 1 has only recently been developed, and there is still room for improvement in various respects.

[0006] For example, 3D data showing the shape of a patient's oral cavity contains noise such as depressions other than the boundaries of teeth. Furthermore, when 3D data is obtained from a plaster model molded from a patient's oral cavity, irregularities on the surface of the plaster model, such as chips or air bubbles, are included in the 3D data as noise. Such noise can significantly reduce segmentation accuracy.

[0007] Therefore, the present invention has been made in consideration of the above problems, and an object of the present invention is to provide a tooth segmentation mechanism with improved robustness against noise. [Means for solving the problem]

[0008] In order to solve the above problem, according to one aspect of the present invention, there is provided an information processing device including a processing unit that extracts tooth regions from three-dimensional data corresponding to the shape of the oral cavity by sequentially performing a pre-processing step of correcting the three-dimensional data, a segmentation step of performing clustering multiple times using a region growing method on the three-dimensional data after the pre-processing, and a post-processing step of adjusting the clusters after the segmentation process.

[0009] The pre-processing may include a process of setting a curvature of a vertex constituting a convex shape to a first predetermined value, and setting a curvature of a vertex constituting a steep concave shape to a second predetermined value.

[0010] The segmentation process may include a first segmentation process that performs clustering using the region growing method for the first time on the three-dimensional data after the preprocessing, extracts a region that is not shown to have a sharp concave shape based on its curvature as the tooth region or gingival region, and extracts a region other than the tooth region or gingival region as a boundary region.

[0011] The segmentation process may include a second segmentation process that performs clustering using the region growing method for a second time on the three-dimensional data after the first segmentation process to integrate a plurality of the boundary regions, and deletes the boundary regions that satisfy a predetermined condition from among the integrated plurality of boundary regions.

[0012] The segmentation process may include a third segmentation process in which clustering using the region growing method is performed a third time on the three-dimensional data after the second segmentation process, and at least a portion of the vertices belonging to the boundary region deleted in the second segmentation process are clustered into the tooth region or the gingival region.

[0013] The post-processing may include merging multiple dental regions based on phase information of the dental regions.

[0014] The post-processing may include merging the dental region that is not adjacent to a gingival region but is adjacent to one or more other dental regions with the other dental region that has the longest adjacent portion length.

[0015] The post-processing may include merging the tooth region adjacent to three or more other regions with other tooth regions whose adjacent portions have lengths equal to or greater than a predetermined value.

[0016] The post-processing may include removing the dental regions based on statistics of the dental regions.

[0017] In addition, in order to solve the above-mentioned problems, according to another aspect of the present invention, there is provided an information processing method executed by a computer, comprising: a preprocessing step of correcting three-dimensional data corresponding to the shape of the oral cavity; a segmentation step of performing clustering multiple times using a region growing method on the three-dimensional data after the preprocessing; and a postprocessing step of adjusting the clusters after the segmentation step, thereby extracting tooth regions from the three-dimensional data.

[0018] In addition, in order to solve the above problem, according to another aspect of the present invention, a program is provided for causing a computer to function as a processing unit that extracts tooth regions from three-dimensional data by sequentially executing a preprocessing process for correcting three-dimensional data corresponding to the shape of the oral cavity, a segmentation process for performing clustering multiple times using a region growing method on the three-dimensional data after the preprocessing, and a post-processing process for adjusting the clusters after the segmentation process. Effect of the Invention

[0019] As described above, the present invention provides a tooth segmentation mechanism with improved robustness against noise. [Brief description of the drawings]

[0020] [Figure 1] FIG. 9 is a block diagram showing an example of a hardware configuration of an information processing device 900 according to an embodiment of the present invention. [Diagram 2] 11 is a flowchart showing an example of the flow of an extraction algorithm for a tooth region executed by the information processing device 900 according to the present embodiment. [Diagram 3] FIG. 13 is a diagram showing the intermediate progress of an algorithm for extracting a tooth region. [Figure 4] FIG. 13 is a diagram showing the intermediate progress of an algorithm for extracting a tooth region. [Diagram 5] FIG. 13 is a diagram showing the intermediate progress of an algorithm for extracting a tooth region. [Figure 6] FIG. 13 is a diagram showing the intermediate progress of an algorithm for extracting a tooth region. [Figure 7] FIG. 13 is a diagram showing the intermediate progress of an algorithm for extracting a tooth region. [Figure 8] FIG. 13 is a diagram showing the intermediate progress of an algorithm for extracting a tooth region. [Figure 9] FIG. 13 is a diagram showing an extraction result by an extraction algorithm for a tooth region. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0021] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In this specification and the drawings, components having substantially the same functional configurations are designated by the same reference numerals, and duplicated explanations will be omitted.

[0022] <1. Configuration example> Fig. 1 is a block diagram showing an example of a hardware configuration of an information processing device 900 according to an embodiment of the present invention. As shown in Fig. 1, the information processing device 900 includes a CPU (Central Processing Unit) 901, a ROM (Read Only Memory) 902, a RAM (Random Access Memory) 903, and a host bus 904a. The information processing device 900 also includes a bridge 904, an external bus 904b, an interface 905, an input device 906, an output device 907, a storage device 908, and a communication device 909.

[0023] The CPU 901 functions as an arithmetic processing device and control device, and controls the overall operation of the information processing device 900 in accordance with various programs. The information processing device 900 may have an electric circuit such as a microprocessor, a DSP (Digital Signal Processor), or an ASIC (Application Specific Integrated Circuit) instead of or in addition to the CPU 901. The ROM 902 stores programs and arithmetic parameters used by the CPU 901. The RAM 903 temporarily stores parameters that change as appropriate when the CPU 901 executes a program.

[0024] The CPU 901, ROM 902, and RAM 903 are connected to one another by a host bus 904a including a CPU bus, etc. The host bus 904a is connected to an external bus 904b such as a PCI (Peripheral Component Interconnect / Interface) bus via a bridge 904. Note that the host bus 904a, bridge 904, and external bus 904b do not necessarily need to be configured separately, and these functions may be implemented on a single bus.

[0025] The input device 906 is a device into which information is input by a user. Examples of such devices include a mouse, a keyboard, a touch panel, a button, a switch, and a lever. The input device 906 may also include a microphone that accepts voice input, or a camera that accepts gesture input. A user of the information processing device 900 can input various data to the information processing device 900 or instruct the information processing device 900 to execute processing by operating the input device 906.

[0026] The output device 907 is a device that outputs information to a user. Examples of such devices include devices that output visual information such as a display and a projector, devices that output auditory information such as a speaker, and devices that output tactile information such as an eccentric motor. The output device 907 outputs, for example, results obtained by various processes performed by the information processing device 900.

[0027] The storage device 908 is a device that stores data. Examples of such devices include a magnetic storage device such as a hard disk drive (HDD), a semiconductor storage device, an optical storage device, and a magneto-optical storage device. The storage device 908 may include a storage medium, a recording device that records data in the storage medium, a reading device that reads data from the storage medium, and a deleting device that deletes data recorded in the storage medium.

[0028] The communication device 909 is a device that communicates with other devices in a wired or wireless manner. The communication device 909 performs communication in compliance with any communication standard, such as Wi-Fi (registered trademark), Bluetooth (registered trademark), LTE (Long Term Evolution), LPWA (Low Power Wide Area), or USB (Universal Serial Bus).

[0029] An example of a hardware configuration capable of realizing the functions of the information processing device 900 according to this embodiment has been described above. Each of the above components may be realized using general-purpose components, or may be realized by hardware specialized for the function of each component. Therefore, the hardware configuration to be used can be changed as appropriate depending on the technical level at the time of implementing this embodiment.

[0030] <2. Operation processing> The information processing device 900 according to this embodiment performs a process of extracting a region constituting teeth from three-dimensional data corresponding to the shape of an oral cavity. For example, when three-dimensional data corresponding to the shape of an oral cavity is input, the information processing device 900 extracts a region constituting teeth from the input three-dimensional data and outputs information indicating the extracted region. The CPU 901 is an example of a processing unit that performs such a process.

[0031] The input three-dimensional data may be mesh data. Mesh data is information indicating a set of surfaces that constitute a three-dimensional object. The mesh data includes three-dimensional coordinates of a plurality of vertices, information indicating a plurality of edges connecting two vertices, and information indicating a plurality of polygons surrounded by the plurality of edges. The mesh data may be, for example, 3D CAD (three-dimensional Computer-Aided Design) data.

[0032] The three-dimensional data corresponding to the shape of the oral cavity may be obtained by 3D scanning the oral cavity of the patient. Alternatively, the three-dimensional data corresponding to the shape of the oral cavity may be obtained by 3D scanning a plaster model of the oral cavity of the patient.

[0033] A region is a set of vertices. A region may also be referred to as a cluster. A region that constitutes a tooth is hereinafter also referred to as a tooth region. A region that constitutes a gum is hereinafter also referred to as a gum region. A region that corresponds to a tooth boundary is hereinafter also referred to as a boundary region.

[0034] To briefly explain the algorithm for extracting the tooth region by the information processing device 900, the information processing device 900 detects a portion having a steep concave shape in the three-dimensional data as a boundary region. Then, the information processing device 900 extracts an area surrounded by the detected boundary region as the tooth region. Hereinafter, the algorithm for extracting the tooth region will be described in detail with reference to Figs. 2 to 9.

[0035] Fig. 2 is a flowchart showing an example of the flow of a dental region extraction algorithm executed by the information processing device 900 according to this embodiment. Figs. 3 to 8 are diagrams showing intermediate steps of the dental region extraction algorithm. Fig. 9 is a diagram showing an extraction result by the dental region extraction algorithm.

[0036] (Step S100: Pre-processing) 2, first, the information processing device 900 executes pre-processing, which is processing for correcting three-dimensional data, and includes steps S102 to S106 in that order.

[0037] (Step S102: Creating point cloud data) The information processing device 900 creates point cloud data from the input mesh data. Point cloud data is an example of three-dimensional data, and is information indicating a set of vertices that form a three-dimensional object. The point cloud data includes three-dimensional coordinates of a plurality of vertices. The information processing device 900 creates the point cloud data by, for example, deleting information indicating edges and information indicating polygons from the mesh data.

[0038] (Step S104: Creating attribute information) Next, the information processing device 900 creates attribute information for each of the multiple vertices included in the point cloud data.

[0039] The attribute information includes an adjacent list of the target vertex. The adjacent list is information indicating one or more other vertices (hereinafter also referred to as adjacent vertices) adjacent to the target vertex. Adjacent here means located within a specified distance from the target vertex. It is desirable that the information indicating adjacent vertices included in the adjacent list is sorted in order starting from the adjacent vertex located closest to the target vertex, and then in the order of adjacent vertices located counterclockwise from the target vertex. In subsequent processing, the adjacent vertices are identified by referring to the adjacent list.

[0040] The attribute information includes information indicating the normal and curvature of the target vertex. In this specification, the curvature of a convex shape is positive, and the curvature of a concave shape is negative. The information processing device 900 calculates the curvature of the vertex based on the normal of the vertex. For example, the information processing device 900 calculates the curvature of the target vertex by performing principal component analysis on the normal vectors of a group of vertices around the target vertex.

[0041] The information processing device 900 sets the curvature of the vertices constituting the convex shape to a first predetermined value (for example, 0). Considering that the segmentation process in this algorithm is performed based on the concave shape and the convex shape does not contribute at all, this configuration makes it possible to reduce the processing load. In addition, the information processing device 900 sets the curvature of the vertices constituting the concave shape to the principal curvature of the concave shape.

[0042] (Step S106: Correction of curvature) Next, the information processing device 900 modifies the curvature included in the attribute information.

[0043] In detail, the information processing device 900 sets the curvature of the vertices constituting a sharp edge (i.e., a steep concave shape) to a second predetermined value. The second predetermined value is a negative value, and is an arbitrary value whose absolute value is greater than an expansion threshold value described later. The information processing device 900 may determine, as a sharp edge, a vertex in which the angle formed by the target vertex in a triangle formed by the target vertex and two adjacent vertices is smaller than a predetermined angle (i.e., an acute angle). When creating attribute information, the principal curvature of the sharp edge of the concave shape may be calculated as a positive value. In this regard, according to this configuration, the principal curvature of the sharp edge of the concave shape can be made negative. As a result, it is possible to improve the accuracy of the segmentation process described later.

[0044] 3 shows the state of the three-dimensional data D100 after preprocessing. In particular, vertices whose curvature is set to the second predetermined value, that is, vertices belonging to the sharp edge S, are colored.

[0045] (Step S200: Segmentation process) The information processing device 900 executes a segmentation process to extract a tooth region from the preprocessed three-dimensional data. The segmentation process includes performing clustering using a region growing method multiple times on the preprocessed three-dimensional data. The region growing method is a technique for recognizing spatially continuous regions that satisfy set conditions as a single region. By performing clustering using the region growing method multiple times, noise contained in the three-dimensional data can be removed, and the tooth region can be appropriately extracted. The segmentation process includes steps S202 to S208 in order.

[0046] (Step S202: Provisional extraction of tooth region) First, the information processing device 900 provisionally extracts a tooth region.

[0047] In detail, the information processing device 900 performs clustering using the first region expansion method on a plurality of vertices included in the three-dimensional data. At that time, the information processing device 900 clusters adjacent vertices whose absolute value of curvature is equal to or less than a predetermined threshold (hereinafter also referred to as the expansion threshold) as vertices included in one region. The information processing device 900 repeats this process on vertices that are not clustered. Then, the information processing device 900 identifies the maximum region among the regions formed by vertices whose absolute value of curvature is equal to or less than the expansion threshold (i.e., the region whose curvature indicates that the shape is not a steep concave shape) as a gingival region. Furthermore, the information processing device 900 identifies the non-maximum region among the regions formed by vertices whose absolute value of curvature is equal to or less than the expansion threshold as a tooth region. On the other hand, the information processing device 900 identifies the vertices that are not clustered in either the tooth region or the gingival region, i.e., the vertices whose absolute value of curvature is greater than the expansion threshold, as a boundary region.

[0048] The information processing device 900 assigns identification information for identifying the clustered regions (more specifically, vertices belonging to the regions) to the attribute information of the clustered regions. Such identification information may include information indicating the type of the region, such as a gingival region, a dental region, and a boundary region. Furthermore, an index for identifying the dental regions may be assigned as identification information to the dental regions. The identification information of the dental region may be a dental formula code. In addition, any identification information for uniquely identifying the cluster, such as a color, may be assigned to each cluster.

[0049] 4 shows the state of the three-dimensional data D202 after processing in this step. In this specification, it is assumed that each cluster is assigned a color as identification information. Therefore, as shown in FIG. 4, the gingival region G, the tooth region T, and the boundary region B are displayed in different colors. Furthermore, a different color is displayed for each tooth region T, such as the tooth region T41 and the tooth region T42 being displayed in different colors. Unless otherwise stated, this is assumed to be the same in the subsequent figures.

[0050] The processing in this step is an example of a first segmentation processing in which clustering using the first region growing method is performed on multiple vertices included in the pre-processed 3D data, and a region that is not shown to have a steep concave shape based on the curvature is extracted as a tooth region or a gingival region, and a region other than the tooth region or the gingival region is extracted as a boundary region. The processing in this step tentatively extracts the tooth region, the gingival region, and the boundary region.

[0051] At least a part of the boundary region is deleted in the subsequent processing. That is, by provisionally extracting the boundary region in the processing of this step, it is possible to delete the unnecessary boundary region in the subsequent processing, and as a result, it is possible to improve the extraction accuracy of the tooth region.

[0052] (Step S204: Removing noise from boundary areas) Next, the information processing device 900 removes noise from the boundary region.

[0053] In detail, the information processing device 900 performs a second clustering by the area expansion method on multiple vertices included in the boundary area. That is, the information processing device 900 integrates adjacent vertices included in the boundary area into one boundary area. As a result, the vertices belonging to the provisionally extracted boundary area are clustered into multiple boundary areas.

[0054] 5 shows the state of the three-dimensional data D204-1 after the second clustering by the region growing method. In the three-dimensional data D204-1 shown in FIG 5, color coding for each tooth region T is omitted.

[0055] Next, the information processing device 900 deletes the boundary region based on the phase information. Deleting the boundary region means deleting the identification information attached to the region that indicates that the region is a boundary region. The phase information is information about adjacent regions, and more specifically, it refers to the number of adjacent regions and their identification information.

[0056] In detail, the information processing device 900 deletes boundary areas that do not satisfy the conditions as tooth boundaries from among the boundary areas based on the phase information. For example, tooth boundaries are considered to be adjacent to a predetermined number of teeth (e.g., 1 to 2). Therefore, the information processing device 900 deletes boundary areas where the number of adjacent tooth areas is not within the predetermined number range. With this configuration, boundary areas with too few or too many adjacent tooth areas are deleted, making it possible to prevent, for example, one tooth from being erroneously extracted as multiple tooth areas.

[0057] After that, the information processing device 900 deletes a boundary region surrounded by one cluster. For example, the information processing device 900 deletes a boundary region that is surrounded by a gingival region on its entire periphery and is not adjacent to a tooth region. With this configuration, for example, it is possible to prevent a gingival region from being erroneously extracted as a tooth region.

[0058] FIG. 6 shows the state of the three-dimensional data D204-2 after the processing of this step. 6, the three-dimensional data D204-2 does not use different colors for each tooth region T. It can be seen that the three-dimensional data D204-2 has a portion of the boundary region B deleted, as compared with the three-dimensional data D202 shown in FIG.

[0059] The processing in this step is an example of a second segmentation process in which clustering using the second region growing method is performed on the three-dimensional data after the first segmentation process to integrate multiple boundary regions, and then boundary regions that satisfy a predetermined condition are deleted from the multiple integrated boundary regions. The processing in this step makes it possible to delete noise in the boundary regions, i.e., unnecessary boundary regions that are not actually boundaries. As a result, it becomes possible to improve the extraction accuracy of the tooth region.

[0060] (Step S206: Extraction of tooth region) Next, the information processing device 900 extracts the tooth region.

[0061] In detail, the information processing device 900 performs a third clustering by the region growing method on the 3D data after the noise in the boundary region is deleted. That is, the information processing device 900 merges the vertices from which the identification information as the boundary region is deleted into the region to which the adjacent vertices belong. Furthermore, the information processing device 900 clusters adjacent vertices included in the tooth region into one tooth region. More simply, the information processing device 900 merges tooth regions that are not separated by a boundary region or a gingival region.

[0062] Fig. 7 shows the state of the three-dimensional data D206 after the processing of this step. In the three-dimensional data D206 shown in Fig. 7, it can be seen that the vertices belonging to the boundary region deleted in the three-dimensional data D204-2 shown in Fig. 6 have been extracted as tooth regions or gingival regions.

[0063] The process in step S206 is an example of a third segmentation process in which clustering is performed using the third region growing method on the three-dimensional data after the second segmentation process, and at least some of the vertices belonging to the boundary region deleted in the second segmentation process are clustered into the tooth region or the gingival region. This process makes it possible to correctly re-extract the vertices that were erroneously provisionally extracted as the boundary region as the tooth region or the gingival region.

[0064] (Step S208: Thinning of boundary area) Next, the information processing device 900 thins the boundary region.

[0065] In detail, the information processing device 900 transfers multiple vertices belonging to the boundary region to another adjacent cluster (tooth region or gingival region) in order from the outermost vertex. The information processing device 900 repeats the above process until the width of the boundary region becomes zero, that is, until the vertices belonging to the boundary region are connected in a line. With this configuration, it is possible to minimize the boundary region.

[0066] Fig. 8 shows the state of three-dimensional data D208 after processing in this step. In the three-dimensional data D208 shown in Fig. 8, it can be seen that the boundary region B has become so thin that it is difficult to visually recognize it, compared with the three-dimensional data D206 shown in Fig. 7. On the other hand, it can be seen that one tooth has been erroneously extracted as two tooth regions T81 and T82.

[0067] (Step S300: Post-processing) Finally, the information processing device 900 executes post-processing, which is processing for adjusting clusters (more specifically, tooth regions) after the segmentation processing. The post-processing includes the processing described below.

[0068] The information processing device 900 integrates multiple dental regions based on the phase information of the dental regions. In particular, the information processing device 900 integrates a dental region that is not adjacent to a gingival region and is adjacent to one or more other dental regions with another dental region having the longest adjacent portion length. Furthermore, the information processing device 900 integrates a dental region that is adjacent to three or more other dental regions with another dental region having an adjacent portion length equal to or greater than a predetermined value. With this configuration, when one tooth is unintentionally clustered as multiple dental regions, it is possible to integrate these multiple dental regions into one dental region.

[0069] The information processing device 900 may delete the tooth region based on statistics of the tooth region. The statistics of the tooth region are values ​​calculated based on the shape of the tooth region. Deleting the tooth region means deleting identification information attached to the region that indicates that the tooth region is a tooth region. As an example, the information processing device 900 may delete the tooth region when the bounding box of the tooth region (a region surrounding the tooth region with a rectangular parallelepiped) is extremely small, extremely large, or extremely thin compared to the size of a standard tooth. As another example, the information processing device 900 may delete a tooth region whose center of gravity is significantly different from a U-shaped arrangement formed by connecting the positions of the centers of gravity of a plurality of extracted tooth regions. With this configuration, it is possible to delete an erroneously extracted tooth region.

[0070] Fig. 9 shows the state of three-dimensional data D300 after post-processing. In the three-dimensional data D300 shown in Fig. 9, it can be seen that one tooth that was extracted as multiple tooth regions T81 and T82 has been correctly extracted as one tooth region T, as compared with the three-dimensional data D208 shown in Fig. 8.

[0071] The information processing device 900 outputs the three-dimensional data after the post-processing. For example, the information processing device 900 outputs point cloud data or mesh data including identification information as a segmentation result as attribute information of the vertices.

[0072] <3. Supplementary Information> Although the preferred embodiment of the present invention has been described in detail above with reference to the accompanying drawings, the present invention is not limited to such an example. It is clear that a person having ordinary knowledge in the technical field to which the present invention pertains can conceive of various modified or altered examples within the scope of the technical ideas described in the claims, and it is understood that these also naturally belong to the technical scope of the present invention.

[0073] For example, in the above embodiment, an example is described in which clustering is performed using a region growing method in the segmentation process, but the present invention is not limited to such an example. Any other algorithm may be used.

[0074] The series of processes by each device described in this specification may be realized using any of software, hardware, and a combination of software and hardware. The programs constituting the software are stored in advance, for example, in a recording medium (more specifically, a non-transient storage medium readable by a computer) provided inside or outside each device. Then, each program is read into a RAM when executed by a computer that controls each device described in this specification, and executed by a processing circuit such as a CPU. The recording medium is, for example, a magnetic disk, an optical disk, a magneto-optical disk, a flash memory, etc. In addition, the above computer program may be distributed, for example, via a network without using a recording medium. In addition, the above computer may be an application-specific integrated circuit such as an ASIC, a general-purpose processor that executes a function by reading a software program, or a computer on a server used for cloud computing. In addition, the series of processes by each device described in this specification may be distributed and processed by multiple computers.

[0075] In addition, the processes described in this specification using flowcharts and sequence diagrams do not necessarily have to be performed in the order shown in the drawings. Some processing steps may be performed in parallel. In addition, additional processing steps may be employed, and some processing steps may be omitted. [Explanation of symbols]

[0076] 900 Information processing device 901 CPU 902 ROM 903 RAM 904a Host Bus 904 Bridge 904b External Bus 905 Interface 906 Input Device 907 Output Device 908 Storage device 909 Communication equipment

Claims

1. The 3D data corresponding to the shape of the oral cavity is used pre-processing the three-dimensional data to correct it; A segmentation process that performs clustering using a region growing method multiple times on the three-dimensional data after the preprocessing; and a post-processing for adjusting the clusters after the segmentation processing, thereby extracting a tooth region from the three-dimensional data; An information processing device comprising:

2. The pre-processing includes a process of setting a curvature of a vertex constituting a convex shape to a first predetermined value and a curvature of a vertex constituting a steep concave shape to a second predetermined value. The information processing device according to claim 1 .

3. The segmentation process includes a first segmentation process in which clustering using the region growing method is performed for the three-dimensional data after the preprocessing for a first time, and a region that is not shown to have a steep concave shape based on a curvature is extracted as the tooth region or the gingival region, and a region other than the tooth region or the gingival region is extracted as a boundary region. The information processing device according to claim 1 .

4. The segmentation process includes a second segmentation process for performing clustering using the region growing method for a second time on the three-dimensional data after the first segmentation process to integrate a plurality of the boundary regions, and deleting the boundary regions that satisfy a predetermined condition from among the integrated plurality of the boundary regions. The information processing device according to claim 3 .

5. The segmentation process includes a third segmentation process in which clustering is performed using the region growing method for a third time on the three-dimensional data after the second segmentation process, and at least a part of the vertices belonging to the boundary region deleted in the second segmentation process is clustered into the tooth region or the gingiva region. The information processing device according to claim 4.

6. The post-processing includes integrating a plurality of dental regions based on phase information of the dental regions. The information processing device according to claim 1 .

7. The post-processing includes merging the dental region that is not adjacent to a gingival region but is adjacent to one or more other dental regions with the other dental region that has the longest adjacent portion length. The information processing device according to claim 6.

8. The post-processing includes integrating the tooth region adjacent to three or more other regions with the other tooth regions having adjacent portions whose lengths are equal to or greater than a predetermined value. The information processing device according to claim 6.

9. The post-processing includes removing the dental region based on statistics of the dental region. The information processing device according to claim 1 .

10. 1. A computer-implemented information processing method, comprising: The 3D data corresponding to the shape of the oral cavity is used pre-processing the three-dimensional data to correct it; A segmentation process that performs clustering using a region growing method multiple times on the three-dimensional data after the preprocessing; and and a post-processing step of adjusting the clusters after the segmentation process, thereby extracting a tooth region from the three-dimensional data. An information processing method comprising:

11. Computer, The 3D data corresponding to the shape of the oral cavity is used pre-processing the three-dimensional data to correct it; A segmentation process that performs clustering using a region growing method multiple times on the three-dimensional data after the preprocessing; and a post-processing for adjusting the clusters after the segmentation processing, thereby extracting a tooth region from the three-dimensional data; A program to function as a

Images