Dental prosthesis identification program, dental prosthesis identification device, and dental prosthesis identification method

Abstract

To provide a dental prosthesis specification program, a dental prosthesis specification apparatus, and a dental prosthesis specification method capable of accurately specifying design data used to produce a desired dental prosthesis from a plurality of pieces of design data.SOLUTION: A prosthesis data acquisition part 71 acquires three-dimensional point group coordinate data of a desired dental prosthesis. A design data acquisition part 72 acquires three-dimensional point group coordinate data of a plurality of pieces of design data to be a candidate. A coordinate transformation part 74 performs coordinate transformation so that the three-dimensional point group coordinate data of each design data matches three-dimensional coordinate system of the three-dimensional point group coordinate data of the dental prosthesis. An approximation degree calculation part 75 calculates an approximation degree between the three-dimensional point group coordinate data of the dental prosthesis and that of each design data after the coordinate transformation. A design data specification part 76 specifies design data most approximate to the dental prosthesis as the design data of the desired dental prothesis.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a dental prosthesis identification program, a dental prosthesis identification apparatus, and a dental prosthesis identification method for identifying design data of a desired dental prosthesis from a plurality of pieces of design data designed by dental CAD software for manufacturing the dental prosthesis. [Background Art]

[0002] As a treatment for dental caries, a treatment of cutting away the carious portion is generally performed. If the portion cut away in this caries treatment is large, it is necessary to restore the tooth morphology with a crown. A crown or the like used for such tooth restoration is called a dental prosthesis.

[0003] Conventionally, manufacturing of a dental prosthesis goes through a process in which a dentist takes an impression of a tooth, a model is made from the impression using plaster or the like, and a dental technician manufactures the dental prosthesis based on the prepared model.

[0004] In recent years, digitization of the manufacturing process for dental prostheses has progressed. The inside of the oral cavity is read with a three-dimensional scanner, the dental prosthesis is designed directly in a computer from scan data without using a model, and based on the design data, it can be manufactured by an automatic machine tool such as an NC processing machine.

[0005] For example, Japanese Unexamined Patent Publication No. 2022-27071 proposes a dental prosthesis provision management device including: a prosthesis target data receiving unit that receives prosthesis target data including at least shape information of a target to be prosthodontically treated with a dental prosthesis; a design information receiving unit that receives design information indicating a desired design of the dental prosthesis; and an output unit that adds information identifying the dental prosthesis to the prosthesis target data and the design information corresponding to the same dental prosthesis and outputs the same (Patent Document 1).

[0006] The digitalization of dental prosthesis design and the automation of manufacturing have drastically reduced the time required for production compared to the past, allowing a single dental technician to produce multiple dental prostheses in a short period of time. [Prior art documents] [Patent Documents]

[0007] [Patent Document 1] Japanese Patent Publication No. 2022-27071 [Overview of the project] [Problems that the invention aims to solve]

[0008] By the way, after the completion of dental prostheses, a delivery check is performed to identify which design data each prosthesis was created based on. Until now, this delivery check has been performed by visually comparing the design data with the completed dental prosthesis, but since there are multiple completed dental prostheses, some of which have similar forms, mistakes in selecting the wrong one frequently occur.

[0009] In particular, with the recent digitalization, multiple dental prostheses are now manufactured in short periods of time, which has increased the amount of design data to compare for each prosthesis, making visual identification difficult.

[0010] In this regard, the invention described in Patent Document 1 manages the process up to manufacturing and does not have a delivery check function. Therefore, a technology that can perform a delivery check to identify the design data of a manufactured dental prosthesis from multiple design data has not yet been established.

[0011] The present invention was made to solve the above-mentioned problems, and aims to provide a dental prosthesis identification program, a dental prosthesis identification device, and a dental prosthesis identification method that can accurately identify the design data used to manufacture a desired dental prosthesis from multiple design data. [Means for solving the problem]

[0012] The dental prosthesis identification program according to the present invention solves the problem of accurately identifying desired design data from among multiple design data for the manufacture of dental prostheses, and is a dental prosthesis identification program for identifying desired dental prosthesis design data from multiple design data designed by dental CAD software for the manufacture of dental prostheses, and causes a computer to function as follows: a prosthesis data acquisition unit that acquires three-dimensional point cloud coordinate data of the desired dental prosthesis; a design data acquisition unit that acquires three-dimensional point cloud coordinate data of a plurality of candidate design data; a coordinate transformation unit that performs coordinate transformation processing so that the three-dimensional point cloud coordinate data of each design data matches the three-dimensional coordinate system of the dental prosthesis's three-dimensional point cloud coordinate data; an approximation calculation unit that calculates the degree of approximation between the three-dimensional point cloud coordinate data of the dental prosthesis and the three-dimensional point cloud coordinate data of each design data after coordinate transformation; and a design data identification unit that identifies the design data that most closely resembles the dental prosthesis as the desired dental prosthesis design data.

[0013] Furthermore, in one aspect of the present invention, when a dental prosthesis and design data match, the system focuses on points located at positions where the coordinate point data of the three-dimensional point cloud coordinate data obtained by coordinate transformation are approximately the same, and in order to solve the problem of quickly calculating an approximate value while suppressing the processing load without using a complex algorithm and identifying the desired design data, the approximation calculation unit includes a corresponding point determination unit that determines whether or not there is a corresponding coordinate point data in the three-dimensional point cloud coordinate data of each design data within a predetermined range centered on each coordinate point data in the three-dimensional point cloud coordinate data of the dental prosthesis, and a corresponding point count unit that counts the number of coordinate point data for which the corresponding coordinate point data has been determined by the corresponding point determination unit to exist as a value indicating the degree of approximation, and the design data identification unit may identify the design data with the largest number of counted corresponding points as the design data of the desired dental prosthesis.

[0014] Furthermore, in one aspect of the present invention, in order to solve the problem of quickly calculating approximate values ​​and identifying desired design data while suppressing processing load without using a complex algorithm similar to that in the above aspect, and enabling comparison of the degree of similarity between design data with different numbers of coordinate point data used for identification, the degree of similarity calculation unit includes: a corresponding point determination unit that determines whether or not there is a corresponding coordinate point data in the three-dimensional point cloud coordinate data of each design data within a predetermined range centered on each coordinate point data in the three-dimensional point cloud coordinate data of the dental prosthesis; a corresponding point count unit that counts the number of coordinate point data for which the corresponding coordinate point data has been determined by the corresponding point determination unit to exist; and a corresponding point ratio calculation unit that calculates the ratio of the number of corresponding points counted by the corresponding point count unit to the total number of points in the three-dimensional point cloud coordinate data of the dental prosthesis as a value indicating the degree of similarity, and the design data identification unit may identify the design data with the highest calculated corresponding point ratio as the design data for the desired dental prosthesis.

[0015] Furthermore, in one aspect of the present invention, when a dental prosthesis and design data match, the distance between neighboring coordinate points in the three-dimensional point cloud coordinate data after coordinate transformation is small. To solve the problem of quickly calculating an approximate value while suppressing processing load without using a complex algorithm and identifying the desired design data, the approximation calculation unit includes a neighboring point identification unit that identifies the nearest coordinate point in the three-dimensional point cloud coordinate data of each design data for each coordinate point in the three-dimensional point cloud coordinate data of the dental prosthesis, and a mean square error calculation unit that calculates a mean square error based on the distance between each coordinate point in the three-dimensional point cloud coordinate data of the dental prosthesis and the neighboring coordinate point identified by the neighboring point identification unit as a value indicating the degree of approximation. The design data identification unit may identify the design data with the smallest calculated mean square error as the design data for the desired dental prosthesis.

[0016] Furthermore, in one aspect of the present invention, in order to solve the problem of reducing the load on the computer when calculating the degree of approximation and speeding up the processing speed for identifying the design data used in the manufacture of a desired dental prosthesis, the present invention may have a random extraction unit that randomly extracts a predetermined number of coordinate point data from the three-dimensional point cloud coordinate data of the dental prosthesis and the three-dimensional point cloud coordinate data of each design data, and the coordinate transformation unit may perform coordinate transformation processing on the randomly extracted point cloud data of the dental prosthesis and the point cloud data of each design data.

[0017] Furthermore, in one aspect of the present invention, in order to solve the problem of providing a processing method that balances high processing speed and accuracy of identification by verifying the correctness of identified design data while increasing processing speed, and recalculating if correctness cannot be confirmed, the present invention may include: an abnormal value determination unit that determines whether the degree of approximation of the design data identified as the design data of a desired dental prosthesis by the design data identification unit is an abnormal value compared to the degree of approximation of all other design data; a design data narrowing unit that, if the abnormal value determination unit determines that it is not an abnormal value, narrows down a predetermined number of design data in descending order of degree of approximation from each of the design data as a preprocessing step when identifying the design data again; and a re-extraction unit that randomly re-extracts a number of coordinate point data greater than the number of coordinate point data extracted immediately before from the three-dimensional point cloud coordinate data of the predetermined number of narrowed-down design data and the three-dimensional point cloud coordinate data of the dental prosthesis.

[0018] The dental prosthesis identification device according to the present invention solves the problem of accurately identifying desired design data from among multiple design data for the manufacture of dental prostheses, and is a dental prosthesis identification device for identifying desired design data for a dental prosthesis from multiple design data designed by dental CAD software for the manufacture of a dental prosthesis, and comprises: a prosthesis data acquisition unit that acquires three-dimensional point cloud coordinate data of a desired dental prosthesis; a design data acquisition unit that acquires three-dimensional point cloud coordinate data of a plurality of candidate design data; a coordinate transformation unit that performs coordinate transformation processing so that the three-dimensional point cloud coordinate data of each design data matches the three-dimensional coordinate system of the three-dimensional point cloud coordinate data of the dental prosthesis; an approximation calculation unit that calculates the degree of approximation between the three-dimensional point cloud coordinate data of the dental prosthesis and the three-dimensional point cloud coordinate data of each design data after coordinate transformation; and a design data identification unit that identifies the design data that most closely resembles the dental prosthesis as the design data of the desired dental prosthesis.

[0019] The dental prosthesis identification method according to the present invention solves the problem of accurately identifying desired design data from among multiple design data related to the manufacture of a dental prosthesis, and is a dental prosthesis identification method for identifying desired dental prosthesis design data from multiple design data designed by dental CAD software for the manufacture of a dental prosthesis, comprising: a prosthesis data acquisition step of acquiring three-dimensional point cloud coordinate data of the desired dental prosthesis; a design data acquisition step of acquiring three-dimensional point cloud coordinate data of a plurality of candidate design data; a coordinate transformation step of performing a coordinate transformation process so that the three-dimensional point cloud coordinate data of each design data matches the three-dimensional coordinate system of the three-dimensional point cloud coordinate data of the dental prosthesis; an approximation calculation step of calculating the degree of approximation between the three-dimensional point cloud coordinate data of the dental prosthesis and the three-dimensional point cloud coordinate data of each design data after coordinate transformation; and a design data identification step of identifying the design data that most closely resembles the dental prosthesis as the design data of the desired dental prosthesis. [Effects of the Invention]

[0020] According to the present invention, design data used for manufacturing a desired dental prosthesis can be accurately specified from a plurality of pieces of design data. [Brief Description of the Drawings]

[0021] [Figure 1] FIG. 1 is a block diagram showing one embodiment of a dental prosthesis specifying apparatus according to the present invention. [Figure 2] FIG. 2 is a perspective view showing a state where design data before coordinate transformation by a coordinate transformation unit in the present embodiment and three-dimensional point cloud coordinate data of a dental prosthesis are displayed in the same three-dimensional coordinate system. [Figure 3] FIG. 3 is a perspective view showing a state where the design data after coordinate transformation by the coordinate transformation unit in the present embodiment and the three-dimensional point cloud coordinate data of the dental prosthesis are displayed in the same three-dimensional coordinate system. [Figure 4] FIG. 4 is a diagram showing a flowchart of a dental prosthesis specifying program according to the present invention. [Figure 5] FIG. 5 is a diagram showing a flowchart of a first approximation degree calculation unit in the dental prosthesis specifying program of the present embodiment. [Figure 6] FIG. 6 is a diagram showing a flowchart of a second approximation degree calculation unit in the dental prosthesis specifying program of the present embodiment. [Figure 7] FIG. 7 is a diagram showing a flowchart of a third approximation degree calculation unit in the dental prosthesis specifying program of the present embodiment. [Figure 8] FIG. 8 is a bar graph showing results obtained by performing coordinate transformation processing by Fast Global Registration on ten pieces of design data with file names 1-001.stl to 1-010.stl in the dental prosthesis specifying program of the present embodiment, and calculating the number of corresponding points for each design data as an approximation degree. [Figure 9] FIG. 9 is a bar graph showing results obtained by performing the same design data and coordinate transformation processing as in FIG. 8, and calculating the ratio of the number of corresponding points for each design data as an approximation degree. [Figure 10]This bar graph shows the results of calculating the mean squared error for each design data set as the degree of approximation, using the same design data and coordinate transformation process as in Figures 8 and 9. [Figure 11] This is a perspective view showing the three-dimensional data of the design data created using dental CAD software to manufacture the dental prosthesis used in Example 1. [Figure 12] This is a perspective view showing the state after the three-dimensional data in the design data of Figure 11 has been converted into three-dimensional point cloud coordinate data consisting of 10,000 coordinate points. [Figure 13] This perspective view shows the state in which the three-dimensional point cloud coordinate data of the dental prosthesis manufactured in this embodiment 1 and the three-dimensional point cloud coordinate data of its design data have been matched and superimposed using (1) RANSAC, (2) Point to Plane ICP, and (3) Fast Global Registration. [Figure 14] In this embodiment 1, coordinate transformation processing was performed using RANSAC on 10 design data files named 1-001.stl to 1-010.stl, and the bar graph shows the results of calculating the degree of approximation, including the number of corresponding points, the percentage of corresponding points, and the mean squared error for each design data file. [Figure 15] In this embodiment 1, a coordinate transformation process was performed using Point to Plane ICP on 10 design data files named 1-001.stl to 1-010.stl. The bar graph shows the results of calculating the degree of approximation, including the number of corresponding points, the percentage of corresponding points, and the mean squared error for each design data file. [Figure 16] In this embodiment 1, coordinate transformation processing was performed using Fast Global Registration on 10 design data files named 1-001.stl to 1-010.stl, and the bar graph shows the results of calculating the number of corresponding points, the percentage of corresponding points, and the mean squared error for each design data as the degree of approximation. [Modes for carrying out the invention]

[0022] Hereinafter, an embodiment of the dental prosthesis identification program, dental prosthesis identification device, and dental prosthesis identification method according to the present invention will be described with reference to the drawings.

[0023] The dental prosthesis identification device 1 is a device for identifying design data for a desired dental prosthesis from a plurality of design data. In this embodiment, as shown in Figure 1, it mainly consists of a display means 2, an input means 3, a three-dimensional scanner means 4, and a computer 5 equipped with a storage means 6 and a calculation processing means 7.

[0024] Display means 2 consists of a liquid crystal display or the like, and displays various information to the user, such as an input image for inputting calculation conditions, and an output image for displaying the output results of design data identified by the calculation process.

[0025] Input means 3 consists of a mouse, keyboard, etc., and is used to input various selections and instructions from the user. In this embodiment, it is used for inputting calculation conditions and selecting methods for calculating the degree of approximation.

[0026] In this embodiment, the display means 2 and input means 3 are provided separately, but the system is not limited to this configuration. It may also be configured with a display input means that combines display and input functions, such as a touch panel.

[0027] The three-dimensional scanner means 4 consists of a three-dimensional scanner device and the like, and reads the three-dimensional data of the manufactured dental prosthesis. In this embodiment, the three-dimensional scanner means 4 generates STL (Stereolithography) data of the read three-dimensional data of the dental prosthesis, and can convert the STL data into three-dimensional point cloud coordinate data consisting of multiple coordinate point data and output it.

[0028] The three-dimensional scanner means 4 may be a commercially available three-dimensional scanner device, or it may be a three-dimensional scanner device specifically designed for use only with the dental prosthesis identification device 1. Furthermore, the output format from the three-dimensional scanner means 4 is not particularly limited and may be output in a format other than point cloud data, such as STL data. In that case, the calculation processing means 7 may acquire the three-dimensional data and convert it into three-dimensional point cloud coordinate data using a general-purpose program or the like.

[0029] Computer 5 is a device for executing the dental prosthesis identification program 1a, and in this embodiment, as shown in Figure 1, it includes a storage means 6 and an arithmetic processing unit 7. Although not shown, computer 5 also includes a power supply device for supplying power to the arithmetic processing unit 7, etc., and communication means such as LAN (Local Area Network), Wi-Fi (Wireless Fidelity), and Bluetooth.

[0030] The storage means 6 stores various types of data and also functions as a working area when the arithmetic processing means 7 performs arithmetic processing. In this embodiment, the storage means 6 is composed of a hard disk, ROM (Read Only Memory), RAM (Random Access Memory), flash memory, etc., and as shown in Figure 1, it has a program storage unit 61, a prosthesis data storage unit 62 that stores three-dimensional point cloud coordinate data for a desired dental prosthesis, and a design data storage unit 63 that stores three-dimensional point cloud coordinate data for design data for manufacturing a dental prosthesis.

[0031] The program storage unit 61 has a dental prosthesis identification program 1a installed, which identifies the design data used to manufacture the desired dental prosthesis from among several candidate design data. The arithmetic processing means 7 then executes the dental prosthesis identification program 1a, causing the computer 5 to function as one of the components described later.

[0032] The usage of the dental prosthesis identification program 1a is not limited to the above configuration. For example, the dental prosthesis identification program 1a may be stored on a non-temporary recording medium readable by the computer 5, such as a CD-ROM or DVD-ROM, and then read and executed directly from that recording medium. Alternatively, the dental prosthesis identification program 1a may be used from an external server or the like using a cloud computing method or an ASP (Application Service Provider) method.

[0033] The prosthesis data storage unit 62 is a storage area for storing three-dimensional point cloud coordinate data of dental prostheses, and in this embodiment, it stores the three-dimensional point cloud coordinate data of dental prostheses output from the three-dimensional scanner means 4.

[0034] The three-dimensional point cloud coordinate data of dental prostheses consists of multiple coordinate point data, each consisting of an identifier name to identify the individual point and the X, Y, and Z coordinate values ​​of the three-dimensional coordinate system for that point.

[0035] The design data storage unit 63 is a storage area that stores three-dimensional point cloud coordinate data of multiple design data designed by dental CAD software for manufacturing dental prostheses. In this embodiment, it stores three-dimensional point cloud coordinate data of design data previously acquired from dental CAD software.

[0036] The three-dimensional point cloud coordinate data for design data, like the three-dimensional point cloud coordinate data for dental prostheses, is composed of multiple coordinate point data, each consisting of an identifier name to identify the individual point and the X, Y, and Z coordinate values ​​of the three-dimensional coordinate system for each point. In this embodiment, identification information used for delivery checks, such as the hospital name and patient name, is associated with and stored for each design data.

[0037] Furthermore, the dental CAD software used to create the design data is not particularly limited; commercially available dental CAD software may be used, or software specifically programmed for use only with the dental prosthesis identification device 1 may be used.

[0038] The arithmetic processing unit 7 consists of a CPU (Central Processing Unit) and the like, and by executing the dental prosthesis identification program 1a installed in the program storage unit 61, it functions as a prosthesis data acquisition unit 71, a design data acquisition unit 72, a random sampling unit 73, a coordinate transformation unit 74, an approximateness calculation unit 75, a design data identification unit 76, an abnormal value discrimination unit 77, a design data filtering unit 78, and a re-extraction unit 79, as shown in Figure 1. Each component will be described in more detail below.

[0039] The prosthesis data acquisition unit 71 acquires three-dimensional point cloud coordinate data of a desired dental prosthesis. In this embodiment, it acquires the three-dimensional point cloud coordinate data of a desired dental prosthesis that is output from the three-dimensional scanner means 4 and stored in the prosthesis data storage unit 62.

[0040] The design data acquisition unit 72 acquires three-dimensional point cloud coordinate data of multiple candidate design data. In this embodiment, it acquires three-dimensional point cloud coordinate data of multiple design data stored in the design data storage unit 63. At this time, the range of design data to be acquired may be all the design data stored in the design data storage unit 63, or it may be a portion (for example, each folder) of the design data selected by the input means 3. Furthermore, the system may include a process to automatically exclude design data identified in a previously executed process from the candidates.

[0041] The random sampling unit 73 randomly extracts a predetermined number of coordinate points from all the coordinate point data of the three-dimensional point cloud coordinate data of dental prostheses acquired by the prosthesis data acquisition unit 71 and the three-dimensional point cloud coordinate data of each design data acquired by the design data acquisition unit 72, in order to reduce the processing load on the coordinate transformation unit 74 described below and to increase the processing speed. In this embodiment, the random sampling unit 73 can arbitrarily select the number of points to extract from 3,000, 5,000, and 10,000 points.

[0042] The method by which the random sampling unit 73 randomly extracts a predetermined number of coordinate point data from all coordinate point data is not particularly limited, but it is preferable to extract the points so that the spacing between the extracted points is approximately constant, so that the overall shape of the tooth designed and the shape of the manufactured dental prosthesis are represented by the extracted point cloud data.

[0043] The coordinate transformation unit 74 adjusts the three-dimensional point cloud coordinate data of each design data to match the three-dimensional coordinate system of the three-dimensional point cloud coordinate data of the dental prosthesis. In this embodiment, the coordinate transformation process is performed to match the point cloud data of each design data extracted by the random extraction unit 73 to the three-dimensional coordinate system of the point cloud data of the dental prosthesis.

[0044] In other words, the three-dimensional point cloud coordinate data of design data created with dental CAD software and the three-dimensional point cloud coordinate data of dental prostheses acquired by a three-dimensional scanner do not necessarily have the same coordinate system, as shown in Figure 2, depending on the position and angle at which the design or reading was performed. Therefore, even if the three-dimensional point cloud coordinate data of the dental prosthesis is compared directly with the three-dimensional point cloud coordinate data of candidate design data, it is not possible to accurately determine whether the shapes are similar or not. To address this, the coordinate transformation unit 74 performs a coordinate transformation process, as shown in Figure 3, to match the three-dimensional point cloud coordinate data of each design data to the three-dimensional coordinate system of the three-dimensional point cloud coordinate data of the dental prosthesis, thereby enabling accurate identification.

[0045] In this embodiment, the coordinate transformation unit 74 uses methods such as RANSAC (RANDOM SAmple Consensus), Point to Plane ICP, and Fast Global Registration to perform coordinate transformations. Conventionally, RANSAC and similar methods are used to align three-dimensional point cloud data obtained by photographing the same object from multiple angles into a single coordinate system. However, the coordinate transformation unit 74 uses these methods to match the coordinates of design data and three-dimensional point cloud coordinate data of dental prostheses, which are different objects. In this regard, as explained in Example 1 below, it has been confirmed that coordinates can be accurately matched by using these methods.

[0046] The approximation calculation unit 75 calculates an approximation score that serves as a criterion for accurately determining whether the design data is the design data for the desired dental prosthesis when comparing the three-dimensional point cloud coordinate data of the dental prosthesis with the three-dimensional point cloud coordinate data of the candidate design data.

[0047] In this embodiment, the approximation degree calculation unit 75 calculates the approximation degree using one method (one approximation degree calculation unit 75) selected by the input means 3 from among the three methods described below, from the first approximation degree calculation unit 75a to the third approximation degree calculation unit 75c.

[0048] In this embodiment, as shown in Figure 1, the system has all three approximation degree calculation units 75, from the first approximation degree calculation unit 75a to the third approximation degree calculation unit 75c. However, the system is not limited to this configuration, and although not shown, it may have one or two of these three approximation degree calculation units 75.

[0049] The first approximation calculation unit 75a calculates the degree of approximation by determining the number of corresponding points among the coordinate point data of the design data after coordinate transformation for each coordinate point data of the dental prosthesis. In other words, when the dental prosthesis and the design data match, the shapes represented by the three-dimensional point cloud coordinate data match, so the coordinate point data of the three-dimensional point cloud coordinate data after coordinate transformation should be in almost the same position, and it is thought that there will be many matching coordinate point data even if randomly selected.

[0050] Therefore, the first approximation calculation unit 75a has a corresponding point determination unit 751 that determines whether or not there are corresponding points, and a corresponding point counting unit 752 that counts the number of corresponding points, in order to calculate the number of corresponding points (points whose positions are almost the same).

[0051] The corresponding point determination unit 751 determines whether or not the coordinate point data in the three-dimensional point cloud coordinate data of the design data after coordinate transformation exists within a predetermined range centered on the coordinate point data in the three-dimensional point cloud coordinate data of the dental prosthesis.

[0052] For example, in three-dimensional point cloud coordinate data for a dental prosthesis, if the X, Y, and Z coordinate values ​​of one coordinate point are 100, and the predetermined range for determining whether or not they correspond is ±0.1 for all three coordinate values, then in the coordinate point data of the three-dimensional point cloud coordinate data of the design data after coordinate transformation, if there is coordinate point data where the X, Y, and Z coordinate values ​​are all within the range of 99.9 to 100.1, then it is determined that there is corresponding coordinate point data.

[0053] The corresponding point counting unit 752 counts the number of coordinate point data of dental prostheses that the corresponding point determination unit 751 has determined to have corresponding coordinate point data. If there are multiple points corresponding to the coordinate point data of a dental prosthesis (where multiple points correspond to one point), the unit may count only 1 regardless of the number of corresponding points, or it may add up the count according to the number of points.

[0054] The second approximation calculation unit 75b calculates the proportion of corresponding points as the degree of approximation. In other words, similar to the first approximation calculation unit 75a, when a dental prosthesis and design data match, many coordinate points correspond to each other in the three-dimensional point cloud coordinate data after coordinate transformation. However, since the number of corresponding points is proportional to the number of randomly extracted coordinate point data, data sets with different numbers of extracted coordinate point data cannot be compared. Therefore, the second approximation calculation unit 75b calculates the proportion of corresponding points as the degree of approximation so that comparison is possible even when the number of extracted coordinate point data differs.

[0055] Therefore, the second approximation calculation unit 75b has a corresponding point determination unit 751, a corresponding point count unit 752, and a corresponding point ratio calculation unit 753 in order to calculate the proportion of corresponding points. Here, the corresponding point determination unit 751 and the corresponding point count unit 752 perform the same processing as the first approximation calculation unit 75a, so their explanation is omitted.

[0056] The correspondence point ratio calculation unit 753 calculates the ratio of the number of corresponding points counted by the correspondence point count unit 752 to the total number of points in the three-dimensional point cloud coordinate data of the dental prosthesis. In this embodiment, the number of extracted point cloud data is used as the denominator, and the number of corresponding points counted by the correspondence point count unit 752 is used to calculate the correspondence point ratio. The second approximation degree calculation unit 75b uses this correspondence point ratio as the approximation degree.

[0057] The third approximation calculation unit 75c calculates the degree of approximation for each coordinate point in the three-dimensional point cloud coordinate data of the dental prosthesis based on the distance (error) between it and the nearest coordinate point in the three-dimensional point cloud coordinate data of the design data. In other words, if the dental prosthesis and the design data match, the shapes match, so the distance between neighboring points in the three-dimensional point cloud coordinate data after coordinate transformation is considered to be small (the error is small).

[0058] Therefore, the third approximation calculation unit 75c in this embodiment includes a neighboring point identification unit 754 that identifies neighboring points in order to calculate the distance to the nearest coordinate point data, and a mean squared error calculation unit 755 that calculates the average distance (error) between neighboring points.

[0059] The nearest point identification unit 754 identifies the nearest coordinate point data in the three-dimensional point cloud coordinate data of each design data for each coordinate point data in the three-dimensional point cloud coordinate data of the dental prosthesis.

[0060] In this embodiment, the nearest point identification unit 754 calculates the distance between two points using the Pythagorean theorem or the like, based on the three-dimensional coordinate values ​​of each coordinate point in the three-dimensional point cloud coordinate data of the dental prosthesis and the three-dimensional coordinate values ​​of each coordinate point in the design data, and identifies the coordinate point data with the shortest calculated distance as the nearest coordinate point data.

[0061] The mean squared error calculation unit 755 calculates the mean squared error based on the distance between each coordinate point data in the three-dimensional point cloud coordinate data of the dental prosthesis and the neighboring coordinate point data identified by the neighboring point identification unit 754. Specifically, it calculates the mean squared error by squaring the distance between each coordinate point data of the dental prosthesis and the corresponding coordinate point data of the neighboring design data identified by the neighboring point identification unit 754, adding all of these together, dividing the added value by the number of coordinate point data in the three-dimensional point cloud coordinate data of the dental prosthesis, and taking the square root of the divided value. The third approximation degree calculation unit 75c uses this mean squared error value as the approximation degree.

[0062] Next, the design data identification unit 76 will be described. The design data identification unit 76 identifies the desired dental prosthesis design data from a plurality of candidate design data based on the degree of approximation calculated by the degree of approximation calculation unit 75. In this embodiment, identification is performed by comparing the magnitude of the degree of approximation, and specifically, depending on the selected degree of approximation calculation unit 75 (first degree of approximation calculation unit 75a to third degree of approximation calculation unit 75c), it is identified as follows.

[0063] When the first approximation calculation unit 75a is selected, as described above, if the dental prosthesis and the design data match, the number of corresponding coordinate point data is large, and if the dental prosthesis and the design data do not match, the number of corresponding points is small. Therefore, the design data identification unit 76 compares the number of corresponding points for each candidate design data and identifies the design data with the most corresponding points as the design data for the desired dental prosthesis.

[0064] When the second approximation calculation unit 75b is selected, as described above, the proportion of corresponding points is high when the dental prosthesis and the design data match, and low when the dental prosthesis and the design data do not match. Therefore, the design data identification unit 76 compares the proportions of corresponding points calculated for each candidate design data and identifies the design data with the highest calculated proportion of corresponding points as the design data for the desired dental prosthesis. This makes it possible to compare design data with different numbers of coordinate point data.

[0065] If the third approximation calculation unit 75c is selected, as described above, if the dental prosthesis and the design data match, the average distance between neighboring points is small (close), and if the dental prosthesis and the design data do not match, the average distance between neighboring points is large (far). Therefore, the design data identification unit 76 compares the mean squared errors calculated for each candidate design data and identifies the design data with the smallest calculated mean squared error as the design data for the desired dental prosthesis.

[0066] The abnormal value discrimination unit 77 confirms that the design data identified by the design data identification unit 76 is the design data for the desired dental prosthesis, thereby improving the accuracy of the identification.

[0067] In other words, the design data identified by the design data identification unit 76 is highly likely to be the design data for the desired dental prosthesis, as described above, and the degree of similarity of the identified design data is considered to be significantly different from the degree of similarity of other candidate design data.

[0068] However, in this embodiment, in order to speed up processing, the number of coordinate point data points to be processed is reduced, and the design data is identified using randomly selected coordinate point data. Therefore, even if the dental prosthesis and the design data match, the distance between coordinate point data points will increase as the number of points decreases. Thus, even if the coordinates match, the coordinate point data of the design data for the dental prosthesis may be far enough apart that they fall outside the range of the corresponding points, or the average distance between neighboring points may increase.

[0069] Therefore, the abnormal value discrimination unit 77 determines whether the degree of approximation of the design data identified as the desired design data by the design data identification unit 76 is an abnormal value compared to the degree of approximation of all other design data, thereby confirming whether the identified design data is the design data for the desired dental prosthesis. If it is not an abnormal value, it determines that confirmation cannot be obtained and performs the identification process again.

[0070] In this embodiment, the abnormal value discrimination unit 77 determines whether the degree of approximation of the identified design data is an abnormal value relative to the degree of approximation of all other design data, based on the Mahalanobis distance, which is calculated as a distance from the relationship with a known sample based on the correlation between multiple variables. In other words, the Mahalanobis distance is calculated based on the degree of approximation of all other design data, and if the degree of approximation of the identified design data is outside the normal range of the Mahalanobis distance specified in advance, it is determined to be an abnormal value, and if it is within the normal range, it is determined not to be an abnormal value.

[0071] Furthermore, the method for identifying outliers is not limited to the Mahalanobis distance; other methods used for anomaly detection, such as the Smirnov-Grubbs test or OneClassSVM, may be appropriately selected.

[0072] The design data filtering unit 78, when the abnormal value determination unit 77 determines that a value is not abnormal, filters a predetermined number of design data from each design data set in order of the degree of similarity as a preprocessing step before re-identifying the design data. This is done to reduce the load on reprocessing and minimize processing speed delays by narrowing down the target for reprocessing. The number of items to be filtered is not particularly limited, but in this embodiment, it can be arbitrarily selected within the range of 5 to 10 items.

[0073] The re-extraction unit 79 randomly re-extracts a larger number of coordinate point data than the number of coordinate point data extracted immediately before, from the three-dimensional point cloud coordinate data of a predetermined number of design data narrowed down by the design data narrowing unit 78 and the three-dimensional point cloud coordinate data of dental prostheses. In other words, the reason why there is no significant difference between the degree of similarity of the identified design data and the degree of similarity of other candidate design data is that the number of points randomly extracted was small and the distance between the coordinate point data was large. Therefore, the number of points is increased to reduce the distance between the coordinate point data and perform identification again in a denser state.

[0074] In this embodiment, if the random sampling unit 73 has selected 3,000 or 5,000 points to extract, the re-extraction unit 79 randomly re-extracts 10,000 coordinate point data points. Then, as will be described in detail in the following description of the operation, the coordinate transformation unit 74 performs a coordinate transformation process, the approximation calculation unit 75 performs a calculation process, and the design data identification unit 76 performs an identification process based on the re-extracted coordinate point data.

[0075] Next, the operation of the dental prosthesis identification program 1a, the dental prosthesis identification device 1, and the dental prosthesis identification method of this embodiment will be described.

[0076] As shown in Figure 4, the dental prosthesis identification program 1a of this embodiment identifies the design data used to manufacture the desired dental prosthesis from among multiple design data sets in each step.

[0077] First, the prosthesis data acquisition unit 71 acquires the three-dimensional point cloud coordinate data of the dental prosthesis stored in the prosthesis data storage unit 62 (prosthesis data acquisition step: S1). The three-dimensional point cloud coordinate data of the dental prosthesis stored in the prosthesis data storage unit 62 is the three-dimensional point cloud coordinate data of the manufactured dental prosthesis output by the three-dimensional scanner means 4. The reading of the dental prosthesis by the three-dimensional scanner means 4 may be done immediately before starting the dental prosthesis identification program 1a, or it may be read in advance before that.

[0078] Next, the design data acquisition unit 72 acquires three-dimensional point cloud coordinate data of the design data designed by dental CAD software for manufacturing the desired dental prosthesis from the design data storage unit 63 (design data acquisition step: S2). If the design data is managed by folder, the three-dimensional point cloud coordinate data of each design data in the target folder may be acquired.

[0079] The random extraction unit 73 randomly extracts a predetermined number of coordinate point data from the three-dimensional point cloud coordinate data of the dental prosthesis acquired by the prosthesis data acquisition unit 71, and also randomly extracts the same number of coordinate point data from the three-dimensional point cloud coordinate data of the design data acquired by the design data acquisition unit 72 (random extraction step: S3). In this embodiment, the number of coordinate point data points input or selected by the input means 3 are randomly extracted.

[0080] The coordinate transformation unit 74 performs a coordinate transformation process to match the point cloud data of each randomly selected design data with the three-dimensional coordinate system of the point cloud data of the randomly selected dental prosthesis (coordinate transformation step: S4). In this embodiment, as a method for performing coordinate transformation, RANSAC, which is conventionally used to align three-dimensional point cloud data acquired from multiple angles of the same object into a single coordinate system, is applied, making it possible to accurately match the coordinates of the design data and the dental prosthesis.

[0081] Next, the approximation calculation unit 75 calculates the degree of approximation between the three-dimensional point cloud coordinate data of the dental prosthesis acquired by the prosthesis data acquisition unit 71 and the three-dimensional point cloud coordinate data of each design data after coordinate transformation by the coordinate transformation unit 74 (approximation calculation step: S5). In this embodiment, as shown in Figures 5 to 7, the approximation calculation unit 75 calculates the degree of approximation based on one of the methods (programs) of the first approximation calculation unit 75a to the third approximation calculation unit 75c, which are selected in advance by the input means 3.

[0082] If the first approximation calculation unit 75a is selected, as shown in Figure 5, the correspondence point determination unit 751 determines whether there is a corresponding coordinate point data for each design data's three-dimensional point cloud coordinate data within a predetermined range centered on each coordinate point data in the three-dimensional point cloud coordinate data of the dental prosthesis (correspondence point determination step: S51). Then, the correspondence point count unit 752 counts the number of coordinate point data for which a corresponding coordinate point data has been determined to exist (correspondence point count step: S52). As a result, as shown in Figure 8, the correspondence point number for each design data's dental prosthesis is calculated as the degree of approximation. After calculation, the process proceeds to the next step S6.

[0083] In step S6, the design data used to create the dental prosthesis is identified based on the degree of approximation calculated by the approximation calculation unit 75 (design data identification step: S6). As described above, if the first degree of approximation calculation unit 75 is selected and the number of corresponding points for each design data is calculated as the degree of approximation, the number of corresponding points should be the highest if the dental prosthesis and the design data match. Therefore, the design data identification unit 76 compares the number of corresponding points for each design data and identifies the design data with the highest number of corresponding points as the desired design data. In other words, in this embodiment, as shown in Figure 8, the number of corresponding points (count) for file name 1-001.stl is the highest. Therefore, the design data identification unit 76 identifies the design data for file name 1-001.stl as the design data for the desired dental prosthesis.

[0084] Furthermore, if the second approximation calculation unit 75b is selected, as shown in Figure 6, similar to the first approximation calculation unit 75a, the corresponding point determination unit 751 determines whether a corresponding point exists (corresponding point determination step: S51), and the corresponding point count unit 752 counts the number of corresponding coordinate point data (corresponding point count step: S52). Then, the corresponding point ratio calculation unit 753 calculates the ratio of the number of corresponding points to the total number of points in the three-dimensional point cloud coordinate data (or extracted point cloud data) of the dental prosthesis (corresponding point ratio calculation step: S53). As a result, as shown in Figure 9, the corresponding point ratio for each design data is calculated. After calculation, the process proceeds to the next step S6.

[0085] If the dental prosthesis and the design data match, the proportion of corresponding points should be high. Therefore, the design data identification unit 76 compares the proportions of corresponding points for each design data and identifies the design data with the highest proportion of corresponding points as the design data for the desired dental prosthesis (design data identification step: S6). In other words, in this embodiment, as shown in Figure 9, the proportion of corresponding points for file name 1-001.stl is the highest. Therefore, the design data identification unit 76 identifies the design data for file name 1-001.stl as the design data for the desired dental prosthesis.

[0086] Furthermore, if the third approximation calculation unit 75c is selected, as shown in Figure 7, the nearest point identification unit 754 identifies the nearest coordinate point data in the three-dimensional point cloud coordinate data of each design data for each coordinate point data in the three-dimensional point cloud coordinate data of the dental prosthesis (neighborhood point identification step: S54). Then, the mean squared error calculation unit 755 calculates the mean squared error based on the distance (error) between two neighboring points (mean squared error calculation step: S55). As a result, the mean squared error for each design data is calculated as shown in Figure 10. After calculation, the process proceeds to the next step S6.

[0087] If the dental prosthesis and the design data match, the calculated mean squared error should be small. Therefore, the design data identification unit 76 compares the mean squared errors of each design data and identifies the design data with the smallest mean squared error as the design data for the desired dental prosthesis (design data identification step: S6). In other words, as shown in Figure 10, the mean squared error of file name 1-001.stl is the smallest. Therefore, the design data identification unit 76 identifies the design data of file name 1-001.stl as the design data for the desired dental prosthesis.

[0088] Next, in this embodiment, a verification process is performed to confirm whether the identified design data is the design data for the desired dental prosthesis. Specifically, as shown in Figure 4, the abnormal value discrimination unit 77 determines whether the degree of approximation of the identified design data is an abnormal value relative to the degree of approximation of all other design data (abnormal value discrimination step: S7). In this embodiment, the Mahalanobis distance is calculated based on the degree of approximation of all other design data, and it is determined whether the degree of approximation of the identified design data is outside the normal range of the Mahalanobis distance specified in advance.

[0089] If the degree of approximation of the identified design data is outside the normal range of the Mahalanobis distance, the abnormal value discrimination unit 77 determines that the identified design data is an abnormal value (S7: YES). In this case, it is indicated that the degree of approximation of the identified design data was significantly different from the degree of approximation of all other design data, and it is considered highly likely that the identified design data is design data for a dental prosthesis. Therefore, the dental prosthesis identification program 1a is terminated, as it is determined that the identified design data is the desired dental prosthesis design data (end).

[0090] On the other hand, if the degree of approximation of the identified design data is within the normal range of the Mahalanobis distance, the abnormal value discrimination unit 77 determines that the identified setting unit data is not an abnormal value (S7: NO). In this case, it indicates that there is no significant difference between the degree of approximation of all other design data and the degree of approximation of the identified design data, and there is a possibility that the identified design data is not design data for a dental prosthesis. Therefore, the process of identifying the design data again is performed.

[0091] First, as a preprocessing step to re-identify the design data, the design data filtering unit 78 compares the degree of similarity of the calculated design data and filters them to a predetermined number of design data in descending order of similarity (design data filtering step: S8). If the first degree of similarity calculation unit 75a is selected, the data will be sorted in descending order of similarity; if the second degree of similarity calculation unit 75b is selected, the data will be sorted in descending order of similarity; and if the third degree of similarity calculation unit 75c is selected, the data will be sorted in descending order of similarity.

[0092] On the other hand, one possible reason why the degree of similarity of the identified design data did not show a significant difference compared to the degree of similarity of other design data is that the number of randomly extracted points was small. Therefore, the re-extraction unit 79 randomly re-extracts more coordinate point data than the number of coordinate point data extracted immediately before from the three-dimensional point cloud coordinate data of the narrowed-down predetermined number of design data and the three-dimensional point cloud coordinate data of dental prostheses (re-extraction step: S9).

[0093] Thus, in this embodiment, as a preprocessing step when re-identifying the design data, the number of design data to be processed is reduced while the number of coordinate point data to be processed is increased, thereby suppressing the overall processing load while improving identification accuracy.

[0094] Subsequently, the design data is identified again using the coordinate point data of the design data that has been filtered after preprocessing and whose number of extracted data points has been increased (S4-S6). Then, the identification process is repeated until the degree of similarity of the identified design data becomes an outlier compared to the degree of similarity of all other filtered design data (S4-S9).

[0095] Although not shown in the diagram, if the value is determined to be non-abnormal two or more times, the design data may not be identified again, and the dental prosthesis identification program 1a may be terminated with the last identified design data as the predetermined design data. Alternatively, the program may be terminated without identification, displaying a message such as "No predetermined design data" on the display means 2.

[0096] According to the dental prosthesis identification program 1a, dental prosthesis identification device 1, and dental prosthesis identification method of this embodiment described above, the following effects can be achieved. 1. By matching the three-dimensional point cloud coordinate data of the design data with the three-dimensional coordinate system of the three-dimensional point cloud coordinate data of the dental prosthesis, the degree of approximation based on the coordinate values ​​can be calculated. 2. By using one of the following methods, which has a low computational load—the number of corresponding points, the ratio of corresponding points, or the mean squared error—the degree of approximation used to identify design data can be calculated quickly. Furthermore, using the ratio of corresponding points allows for comparison between approximations with different numbers of coordinate point data used in the calculation. 3. By comparing the degree of approximation calculated without using complex algorithms, the desired design data can be identified. 4. By reducing the number of points in the three-dimensional point cloud coordinate data to be processed and randomly extracting them, the processing load related to identifying design data can be reduced, and the processing speed can be increased. Therefore, even if there are many candidate design data, rapid identification becomes possible. 5. After identifying the design data, the accuracy of the identified design data can be verified by determining whether its degree of similarity is significantly similar to the degrees of similarity of all other design data. If it is not significantly similar, the identification process can be repeated to improve the accuracy of the identification. 6. When re-identifying the design data, pre-processing can be performed by narrowing down the target design data and increasing the number of coordinate point data to be extracted. This reduces the overall processing load while enabling re-identification with higher accuracy than the previous identification process.

[0097] Next, specific examples of the dental prosthesis identification program, dental prosthesis identification device, and dental prosthesis identification method according to the present invention will be described. It should be noted that the technical scope of the present invention is not limited to the features shown in the following examples. [Examples]

[0098] In this first embodiment, a dental prosthesis identification program and a dental prosthesis identification device according to the present invention were created, and a verification experiment was conducted to determine whether these could be used to identify the design data of a dental prosthesis that was actually manufactured from multiple design data for manufacturing a dental prosthesis.

[0099] [Design data and dental prostheses used in the verification experiment] In this embodiment 1, the dental prosthesis used in the verification experiment was a crown for a molar, as shown in Figure 11, and was manufactured using the design data. The design data was converted into three-dimensional point cloud coordinate data, as shown in Figure 12. In addition, ten design data sets were prepared as candidate design data, including the design data for the dental prosthesis (the correct design data).

[0100] [Three-dimensional point cloud coordinate data and three-dimensional scanner for dental prostheses] The manufactured dental prostheses were scanned using a dental 3D scanner and converted into 3D point cloud coordinate data. The dental 3D scanners used were the EDGE from DOF, the DORA from Digital Process, and the Trios3 from 3shape. No significant difference in accuracy was observed between the different models of dental 3D scanners.

[0101] [Coordinate transformation processing method] Three methods—RANSAC, Point to Plane ICP, and Fast Global Registration—were used to perform coordinate transformations to match the coordinates of the three-dimensional point cloud data of dental prostheses with those of the design data. As shown in Figure 13, all of these methods successfully matched the coordinates of the three-dimensional point cloud data of dental prostheses with those of the design data.

[0102] [Calculation of the degree of approximation of multiple design data] We calculated the degree of similarity between the three-dimensional point cloud coordinate data of each design data and the three-dimensional point cloud coordinate data of the dental prosthesis, using 10 design data sets, including the design data used to design the manufactured dental prosthesis. The calculated degree of similarity is expressed as the number of corresponding points, the percentage of corresponding points, and the mean squared error. The calculation results are described below.

[0103] When RANSAC was used for coordinate transformation, as shown in the upper part of Figure 14, the design data with the filename 1-001.stl, which is the correct answer and listed furthest to the left, had the most corresponding points. Furthermore, this result was significantly higher than the number of corresponding points for all other design data. As shown in the middle part of Figure 14, the proportion of corresponding points was also highest for filename 1-001.stl, and was significantly higher than that of all other design data.

[0104] On the other hand, as shown in the bottom row of Figure 14, the mean squared error was smallest for 1-005.stl, and the correct file name 1-001.stl was the second smallest.

[0105] Next, when Point to Plane ICP was used for coordinate transformation, as shown in the upper and middle sections of Figure 15, the number of corresponding points and the proportion of corresponding points were highest for the design data file named 1-001.stl, which is the correct answer and is listed on the far left. This result was significantly higher than the number of corresponding points for all other design data.

[0106] Furthermore, as shown in the bottom row of Figure 15, the mean squared error was also smallest for the correct file name 1-001.stl, indicating that the correct design data could be identified.

[0107] Furthermore, when Fast Global Registration was used for coordinate transformation, as shown in the upper and middle sections of Figure 16, the number of corresponding points and the percentage of corresponding points were highest for the design data file named 1-001.stl, which is the correct answer, and significantly higher than the number of corresponding points for all other design data. Also, as shown in the bottom section of Figure 16, the mean squared error was also smallest for the correct file named 1-001.stl, indicating that the correct design data could be identified.

[0108] As described above, it was demonstrated that the design data used to design the dental prosthesis can be identified by matching the coordinates of the three-dimensional point cloud coordinate data of the dental prosthesis with the coordinates of the three-dimensional point cloud coordinate data of the design data, and by calculating the number of corresponding points, the proportion of corresponding points, and the mean squared error. Furthermore, when RANSAC was used for coordinate transformation, the mean squared error for the correct design data was the second smallest, but at least it was among the top results. Therefore, in such cases, it is thought that accuracy can be improved and precise identification can be achieved by changing the coordinate transformation method or by increasing the number of coordinate points in the three-dimensional point cloud coordinate data being processed.

[0109] It should be noted that the dental prosthesis identification program, dental prosthesis identification device, and dental prosthesis identification method according to the present invention are not limited to the embodiments described above and can be modified as appropriate. For example, the design data storage unit 63 may store the design data as three-dimensional data such as STL data generated by dental CAD software, and convert it to three-dimensional point cloud coordinate data using a general-purpose program or the like when performing the identification process. [Explanation of Symbols]

[0110] 1. Dental prosthesis identification device 1a Dental Prosthesis Identification Program 2 Display means 3. Input means 4. Three-dimensional scanner means 5 Computers 6 Memory means 7. Calculation processing means 61 Program Storage Unit 62 Prosthetic Data Storage Unit 63 Design data storage unit 71 Prosthetic Data Acquisition Unit 72 Design Data Acquisition Unit 73 Random Sampling Section 74 Coordinate Transformation Unit 75 Approximation calculation part 75a First approximation calculation unit 75b Second approximation calculation unit 75c Third Approximation Calculation Unit 76 Design Data Identification Department 77 Anomaly detection unit 78 Design Data Refinement Section 79 Re-extraction part 751 Corresponding Point Discrimination Unit 752 Corresponding score counting unit 753 Corresponding Score Ratio Calculation Unit 754 Nearby Point Identification Unit 755 Mean Squared Error Calculation Unit

Claims

1. A dental prosthesis identification program for identifying the design data of a desired dental prosthesis from multiple design data sets designed using dental CAD software for the purpose of manufacturing dental prostheses, A prosthesis data acquisition unit that acquires three-dimensional point cloud coordinate data of a desired dental prosthesis, A design data acquisition unit that acquires three-dimensional point cloud coordinate data of multiple candidate design data, A coordinate transformation unit performs coordinate transformation processing to match the three-dimensional point cloud coordinate data of each of the aforementioned design data to the three-dimensional coordinate system of the three-dimensional point cloud coordinate data of the dental prosthesis. A approximation calculation unit calculates the degree of approximation between the three-dimensional point cloud coordinate data of the dental prosthesis and the three-dimensional point cloud coordinate data of each design data after coordinate transformation. A design data identification unit identifies the design data that most closely resembles the aforementioned dental prosthesis as the design data for the desired dental prosthesis. A dental prosthesis identification program that enables the computer to function.

2. The approximation calculation unit is, A correspondence point determination unit that determines whether or not there is a corresponding coordinate point in the three-dimensional point cloud coordinate data of each design data within a predetermined range centered on each coordinate point in the three-dimensional point cloud coordinate data of the dental prosthesis, The system includes a corresponding point counting unit that counts the number of coordinate point data points that the corresponding point discrimination unit has determined to have corresponding coordinate point data, as a value indicating the degree of approximation. The dental prosthesis identification program according to claim 1, wherein the design data identification unit identifies the design data with the highest number of counted corresponding points as the design data for a desired dental prosthesis.

3. The approximation calculation unit is, A correspondence point determination unit that determines whether or not there is a corresponding coordinate point in the three-dimensional point cloud coordinate data of each design data within a predetermined range centered on each coordinate point in the three-dimensional point cloud coordinate data of the dental prosthesis, The correspondence point counting unit counts the number of coordinate point data that the correspondence point determination unit has determined to have corresponding coordinate point data, The system includes a correspondence point ratio calculation unit that calculates the ratio of the number of corresponding points counted by the correspondence point count unit to the total number of points in the three-dimensional point cloud coordinate data of the dental prosthesis, as a value indicating the degree of approximation. The dental prosthesis identification program according to claim 1, wherein the design data identification unit identifies the design data with the highest calculated correspondence point ratio as the design data for the desired dental prosthesis.

4. The approximation calculation unit is, A neighboring point identification unit identifies the nearest coordinate point data in the three-dimensional point cloud coordinate data of each design data for each coordinate point data in the three-dimensional point cloud coordinate data of the dental prosthesis, The system includes a mean squared error calculation unit that calculates a mean squared error based on the distance between each coordinate point data in the three-dimensional point cloud coordinate data of the dental prosthesis and the neighboring coordinate point data identified by the neighboring point identification unit, as a value indicating the degree of approximation. The dental prosthesis identification program according to claim 1, wherein the design data identification unit identifies the design data with the smallest calculated mean squared error as the design data for a desired dental prosthesis.

5. The computer is configured to function as a random extraction unit that randomly extracts a predetermined number of coordinate point data from the three-dimensional point cloud coordinate data of the dental prosthesis and the three-dimensional point cloud coordinate data of each design data. The dental prosthesis identification program according to any one of claims 1 to 4, wherein the coordinate transformation unit performs coordinate transformation processing on the point cloud data of the randomly extracted dental prosthesis and the point cloud data of each design data.

6. An abnormal value determination unit that determines whether the degree of approximation of the design data identified as the design data for a desired dental prosthesis by the design data identification unit is an abnormal value compared to the degree of approximation of all other design data, If the abnormal value detection unit determines that the value is not abnormal, the preprocessing for re-identifying the design data is as follows: A design data filtering unit that narrows down a predetermined number of design data from each of the aforementioned design data in order of decreasing similarity, A re-extraction unit randomly re-extracts a number of coordinate point data points greater than the number of coordinate point data points extracted immediately before, from a predetermined number of the three-dimensional point cloud coordinate data points of the design data and the three-dimensional point cloud coordinate data points of the dental prosthesis. A dental prosthesis identification program according to claim 5, which enables a computer to function.

7. A dental prosthesis identification device for identifying the design data of a desired dental prosthesis from multiple design data sets designed using dental CAD software for the purpose of manufacturing dental prostheses, A prosthesis data acquisition unit that acquires three-dimensional point cloud coordinate data of a desired dental prosthesis, A design data acquisition unit that acquires three-dimensional point cloud coordinate data of multiple candidate design data, A coordinate transformation unit performs coordinate transformation processing to match the three-dimensional point cloud coordinate data of each of the aforementioned design data to the three-dimensional coordinate system of the three-dimensional point cloud coordinate data of the dental prosthesis. A approximation calculation unit calculates the degree of approximation between the three-dimensional point cloud coordinate data of the dental prosthesis and the three-dimensional point cloud coordinate data of each design data after coordinate transformation. A design data identification unit identifies the design data that most closely resembles the aforementioned dental prosthesis as the design data for the desired dental prosthesis. A dental prosthesis identification device having the following:

8. A method for identifying a dental prosthesis, for identifying the design data of a desired dental prosthesis from multiple design data sets designed using dental CAD software for the manufacture of dental prostheses, A prosthesis data acquisition step to obtain three-dimensional point cloud coordinate data of the desired dental prosthesis, A design data acquisition step involves acquiring three-dimensional point cloud coordinate data of multiple candidate design data, A coordinate transformation step is performed to perform a coordinate transformation process so that the three-dimensional point cloud coordinate data of each of the design data matches the three-dimensional coordinate system of the three-dimensional point cloud coordinate data of the dental prosthesis. A approximation calculation step that calculates the degree of approximation between the three-dimensional point cloud coordinate data of the dental prosthesis and the three-dimensional point cloud coordinate data of each design data after coordinate transformation, A design data identification step of identifying the design data that most closely resembles the aforementioned dental prosthesis as the design data for the desired dental prosthesis. A method for identifying dental prostheses, comprising the above-mentioned features.

Images