Oral cavity area identification method, electric toothbrush and computer device

Abstract

The invention discloses an oral cavity area identification method, an electric toothbrush and a computer device, dual positioning of spatial attitude and distance is realized through a nine-axis inertia unit and a distance measuring sensor in the electric toothbrush, and easy-to-mix oral cavity areas such as left inner / right outer, right inner / left outer and the like can be identified more accurately. According to the method, the target random forest model is constructed based on the user sample data set of the user, so that the target random forest model can adapt to the oral cavity characteristics of the user, and personalized modeling is realized, so that the personalized recognition capability adaptive to different user individuals is provided while the oral cavity region recognition precision and stability are improved, and the user experience is improved.

Description

Technical Field

[0001] This application relates to the field of small household appliance technology, and in particular to a method for recognizing oral regions, an electric toothbrush, and a computer device. Background Technology

[0002] With the increasing awareness of oral health management, smart electric toothbrushes have become an important tool in daily life.

[0003] Currently, some electric toothbrushes can identify oral cavity regions using triaxial accelerometers and gyroscopes. For example, patent CN110608753A discloses an oral cavity region identification method. This method acquires acceleration and angular velocity data of the electric toothbrush within the oral cavity region. The acceleration and angular velocity data are collected by a triaxial accelerometer and gyroscope mounted on the electric toothbrush, respectively. Then, the acceleration and angular velocity data are smoothed and filtered to obtain target acceleration and target angular velocity data. Next, the target acceleration and target angular velocity data are fused to obtain the target roll angle. Then, a first preset threshold range and a second preset threshold range corresponding to the target roll angle are obtained, along with a preset first threshold corresponding to the first target data. Finally, the oral cavity sub-region corresponding to the acceleration data is determined based on the target roll angle, the first target data, the second target data, the first preset threshold range, the second preset threshold range, and the first preset threshold. It is evident that this method, relying solely on acceleration and angular velocity for oral cavity region identification, has low accuracy. Summary of the Invention

[0004] This application provides a method for identifying oral cavity regions, an electric toothbrush, and a computer device. Data is collected through a nine-axis inertial unit and a ranging sensor in the electric toothbrush, and a target random forest model is constructed based on the user's user sample dataset to improve the accuracy of oral cavity region identification.

[0005] In a first aspect, a method for identifying oral cavity regions is provided, applied to an electric toothbrush, comprising: responding to a user brushing their teeth with the electric toothbrush, obtaining the three-axis acceleration, three-axis angular velocity, and three-axis magnetic force components of the electric toothbrush using a nine-axis inertial unit; obtaining the distance value of the electric toothbrush using a distance sensor of the electric toothbrush; determining the pitch angle and roll angle of the electric toothbrush based on the three-axis acceleration and the three-axis angular velocity; determining the azimuth angle of the electric toothbrush based on the three-axis magnetic force components; and converting the pitch angle, the roll angle, and the azimuth angle into a single data set. The angle and the distance value are sent to the server, so that the server uses a target random forest model to determine the target oral cavity region corresponding to the brushing operation. The target random forest model is constructed based on the user's user sample dataset, which is collected during the user's calibration operation using the electric toothbrush according to the target standard brushing method. The user sample dataset includes sample triaxial acceleration, sample triaxial angular velocity and sample triaxial magnetic force components collected using the nine-axis inertial unit, as well as sample distance values ​​collected using the distance sensor.

[0006] In some embodiments, determining the pitch angle and roll angle of the electric toothbrush based on the triaxial acceleration and the triaxial angular velocity includes: determining a first pitch angle and a first roll angle using the triaxial acceleration; determining a second pitch angle and a second roll angle using the triaxial angular velocity; determining the pitch angle based on the first pitch angle and the second pitch angle; and determining the roll angle based on the first roll angle and the second roll angle.

[0007] The pitch angle is determined by the first pitch angle and the second pitch angle, and the roll angle is determined by the first roll angle and the second roll angle. By comprehensively considering the three-axis acceleration and the three-axis angular velocity, the data fusion of the accelerometer and gyroscope is achieved, which improves the accuracy of the pitch angle and roll angle.

[0008] In some embodiments, determining the pitch angle based on the first pitch angle and the second pitch angle includes: fusing the first pitch angle and the second pitch angle based on a first weight and the x-axis angular velocity of the three-axis angular velocities to determine the pitch angle; determining the roll angle based on the first roll angle and the second roll angle includes: fusing the first roll angle and the second roll angle based on a second weight and the y-axis angular velocity of the three-axis angular velocities to determine the roll angle. This further improves the accuracy of the pitch angle and roll angle.

[0009] In some embodiments, the method further includes: in response to the user performing the calibration operation using the electric toothbrush, acquiring sample data of each region corresponding to each standard oral cavity region using the nine-axis inertial unit and the ranging sensor at a target sampling frequency, wherein each standard oral cavity region is the user's brushing region corresponding to the target standard brushing method; adding a label to each of the region sample data to the standard oral cavity region to determine the user sample dataset; determining an attitude data set including sample pitch angle, sample roll angle, sample azimuth angle, and sample ranging value based on the user sample dataset; and sending the attitude data set to the server, so that the server uses the attitude data set to construct the target random forest model.

[0010] By obtaining a user sample dataset during calibration and determining a pose data set based on the user sample dataset, the pose data set is sent to the server, enabling the server to efficiently and accurately construct a target random forest model that conforms to the user's oral cavity characteristics.

[0011] In some embodiments, determining the attitude data set including sample pitch angle, sample roll angle, sample azimuth angle, and sample ranging value based on the user sample dataset includes: determining the sample pitch angle and sample roll angle of the electric toothbrush based on the sample triaxial acceleration and the sample triaxial angular velocity; determining the sample azimuth angle of the electric toothbrush based on the sample triaxial magnetic force components; and determining the attitude data set based on the sample pitch angle, sample roll angle, sample azimuth angle, and sample ranging value corresponding to each of the standard oral cavity regions.

[0012] By comprehensively considering the three-axis acceleration and three-axis angular velocity of the sample, data fusion of accelerometer and gyroscope is achieved, which improves the accuracy of the sample pitch angle and roll angle, and thus improves the accuracy of the attitude data set.

[0013] Secondly, a method for identifying oral cavity regions is provided, applied to a server, comprising: obtaining pitch angle, roll angle, azimuth angle, and distance values ​​received from an electric toothbrush, wherein the pitch angle, roll angle, and azimuth angle are determined based on the triaxial acceleration, triaxial angular velocity, and triaxial magnetic force components of the electric toothbrush, wherein the triaxial acceleration, triaxial angular velocity, triaxial magnetic force components, and the distance value are obtained during a user's brushing operation using the electric toothbrush, utilizing the nine-axis inertial unit and distance sensor of the electric toothbrush; and obtaining the pitch angle, roll angle, azimuth angle, and distance values ​​based on the triaxial acceleration, roll angle, azimuth angle, and distance values ​​of the electric toothbrush. The rolling angle, the azimuth angle, and the distance measurement value are used to determine the target oral cavity region corresponding to the brushing operation using a target random forest model. The target random forest model is constructed based on the user's user sample dataset, which is collected during the user's calibration operation using the electric toothbrush according to the target standard brushing method. The user sample dataset includes sample triaxial acceleration, sample triaxial angular velocity, and sample triaxial magnetic force components collected using the nine-axis inertial unit, as well as sample distance measurement values ​​collected using the distance measurement sensor.

[0014] In some embodiments, before determining the target oral cavity region corresponding to the brushing operation using a target random forest model based on the pitch angle, roll angle, azimuth angle, and distance measurement value, the method further includes: obtaining a set of posture data received from the electric toothbrush, including sample pitch angles, sample roll angles, sample azimuth angles, and sample distance measurement values, wherein the posture data set is determined by the electric toothbrush based on the user sample dataset; dividing the posture data set into a training set and a test set according to a target partitioning ratio; standardizing the training set and the test set to obtain a standardized training set and a standardized test set; and constructing the target random forest model using the standardized training set and the standardized test set. This achieves efficient and accurate construction of a target random forest model that conforms to the user's oral cavity characteristics.

[0015] In some embodiments, after determining the target oral cavity region corresponding to the brushing operation using a target random forest model based on the pitch angle, roll angle, azimuth angle, and distance value, the method further includes: determining the data frame corresponding to the pitch angle, roll angle, azimuth angle, and distance value as the current data frame; determining at least one historical data frame preceding the current data frame, and determining the historical target oral cavity region corresponding to the historical data frame; and outputting the region identifier of the target oral cavity region when the number of current data frames and historical data frames reaches a target number, the current data frames and historical data frames form consecutive data frames, and the historical target oral cavity region is the same as the target oral cavity region. This improves the accuracy of the output recognition result.

[0016] Thirdly, an electric toothbrush is provided, including a nine-axis inertial unit, a distance sensor, and a controller. The controller is configured to: in response to a user brushing their teeth with the electric toothbrush, obtain the three-axis acceleration, three-axis angular velocity, and three-axis magnetic force components of the electric toothbrush using the nine-axis inertial unit; obtain the distance value of the electric toothbrush using the distance sensor; determine the pitch angle and roll angle of the electric toothbrush based on the three-axis acceleration and the three-axis angular velocity; determine the azimuth angle of the electric toothbrush based on the three-axis magnetic force components; and input the pitch angle, the roll angle, and the... The azimuth angle and the distance measurement value are sent to the server, which then uses a target random forest model to determine the target oral cavity region corresponding to the brushing operation. The target random forest model is constructed based on the user's user sample dataset, which is collected during the user's calibration operation using the electric toothbrush according to the target standard brushing method. The user sample dataset includes sample triaxial acceleration, sample triaxial angular velocity, and sample triaxial magnetic force components collected using the nine-axis inertial unit, as well as sample distance measurement values ​​collected using the distance sensor.

[0017] Fourthly, a computer device is provided, including a processor, a memory, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the oral cavity region identification method as described in the first or second aspect.

[0018] By applying the above technical solutions, in response to a user's brushing operation with an electric toothbrush, the toothbrush's three-axis acceleration, three-axis angular velocity, and three-axis magnetic force components are obtained using its nine-axis inertial unit, and its distance measurement value is obtained using its ranging sensor. The pitch and roll angles of the toothbrush are determined based on the three-axis acceleration and angular velocity, and the azimuth angle is determined based on the three-axis magnetic force components. The pitch, roll, azimuth, and distance values ​​are then sent to the server, which uses a target random forest model to determine the target oral cavity region corresponding to the brushing operation. This dual positioning of spatial attitude and distance is achieved through the nine-axis inertial unit and ranging sensor in the electric toothbrush, enabling more accurate identification of easily confused oral cavity regions such as left inner / right outer and right inner / left outer. By constructing a target random forest model based on the user's sample dataset, the model can be adapted to the user's oral cavity characteristics, achieving personalized modeling. This improves the accuracy and stability of oral cavity region recognition while providing personalized recognition capabilities tailored to different individual users, thus enhancing the user experience. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 The flowchart of a method for identifying oral cavity regions according to an embodiment of this application Figure 1 ;

[0021] Figure 2 A flowchart illustrating the determination of the pitch and roll angles of an electric toothbrush according to an embodiment of this application;

[0022] Figure 3 The flowchart of a method for identifying oral cavity regions according to an embodiment of this application Figure 2 ;

[0023] Figure 4 The flowchart of a method for identifying oral cavity regions according to an embodiment of this application Figure 3 ;

[0024] Figure 5 The flowchart of a method for identifying oral cavity regions according to an embodiment of this application Figure 4 ;

[0025] Figure 6 The flowchart of a method for identifying oral cavity regions according to an embodiment of this application Figure 5 ;

[0026] Figure 7 A schematic diagram of the 16 oral cavity zones corresponding to the Bass brushing technique;

[0027] Figure 8 This is a structural block diagram of an electric toothbrush according to an embodiment of this application;

[0028] Figure 9 This is a structural block diagram of a computer device according to an embodiment of this application. Detailed Implementation

[0029] Various embodiments and features of this application are described herein with reference to the accompanying drawings.

[0030] It should be understood that various modifications can be made to the embodiments described herein. Therefore, the above description should not be considered as limiting, but merely as an example of embodiments. Other modifications within the scope and spirit of this application will be apparent to those skilled in the art.

[0031] The accompanying drawings, which are included in and form part of this specification, illustrate embodiments of the present application and, together with the general description of the present application given above and the detailed description of the embodiments given below, serve to explain the principles of the present application.

[0032] These and other features of this application will become apparent from the following description of preferred forms of embodiments given as non-limiting examples, with reference to the accompanying drawings.

[0033] It should also be understood that although this application has been described with reference to some specific examples, those skilled in the art can certainly implement many other equivalent forms of this application.

[0034] The above and other aspects, features and advantages of this application will become more apparent when taken in conjunction with the accompanying drawings and in view of the following detailed description.

[0035] Specific embodiments of this application are described thereafter with reference to the accompanying drawings; however, it should be understood that the claimed embodiments are merely examples of this application, which can be implemented in various ways. Well-known and / or repeated functions and structures are not described in detail to avoid unnecessary or redundant details that could obscure the application. Therefore, the specific structural and functional details claimed herein are not intended to be limiting, but merely serve as the basis and representative basis for the claims to teach those skilled in the art to use this application in a variety of substantially any suitable detailed structures.

[0036] This specification may use the phrases “in one embodiment,” “in another embodiment,” “in yet another embodiment,” or “in other embodiments,” all of which may refer to one or more of the same or different embodiments according to this application.

[0037] This application discloses a method for recognizing oral regions, applied to an electric toothbrush. By utilizing a nine-axis inertial unit and a ranging sensor within the electric toothbrush, dual positioning of spatial attitude and distance is achieved. This enables more accurate recognition of easily confused oral regions such as left inner / right outer and right inner / left outer. A target random forest model is constructed based on a user's sample dataset, allowing the model to adapt to the user's oral characteristics and achieve personalized modeling. This improves the accuracy and stability of oral region recognition while providing personalized recognition capabilities tailored to different individual users, thus enhancing the user experience.

[0038] like Figure 1 As shown, the identification method includes the following steps:

[0039] Step S101: In response to the user brushing teeth with the electric toothbrush, the three-axis acceleration, three-axis angular velocity and three-axis magnetic force components of the electric toothbrush are obtained using the nine-axis inertial unit of the electric toothbrush, and the distance value of the electric toothbrush is obtained using the distance sensor of the electric toothbrush.

[0040] In this embodiment, the nine-axis inertial unit can be a nine-axis MEMS inertial unit integrating an accelerometer, gyroscope, and magnetometer, or it can be an inertial unit formed by independently configured accelerometers, gyroscopes, and magnetometers. The distance sensor is used to measure the distance between the electric toothbrush and the teeth / skin, and can be any of the following sensors: laser TOF (Time of Flight) distance sensor, millimeter-wave radar sensor, infrared distance sensor, etc. For example, if the distance sensor reading is within the range of 1-3 cm, the area is determined to be the inner side of the tooth. If the distance reading is less than or equal to 1 cm, the area is determined to be the outer side of the tooth (skin-covered side). If the distance reading is greater than 3 cm, i.e., the tooth is unobstructed, the area is determined to be the incisor.

[0041] Users can activate the start button on the electric toothbrush to brush their teeth. In response to this brushing operation, the accelerometer, gyroscope and magnetometer in the nine-axis inertial unit are used to obtain the three-axis acceleration, three-axis angular velocity and three-axis magnetic force components of the electric toothbrush, respectively. At the same time, the distance value of the electric toothbrush is obtained by the distance sensor.

[0042] Step S102: Determine the pitch angle and roll angle of the electric toothbrush based on the triaxial acceleration and the triaxial angular velocity.

[0043] In this embodiment, the triaxial acceleration, triaxial angular velocity, and triaxial magnetic force components are first converted into corresponding attitude angles so that the oral cavity area can be identified subsequently using the attitude angles and ranging values ​​of the electric toothbrush. The triaxial acceleration and triaxial angular velocity are converted to determine the pitch and roll angles of the electric toothbrush.

[0044] Step S103: Determine the azimuth angle of the electric toothbrush based on the triaxial magnetic force components.

[0045] The triaxial magnetic components are converted to determine the azimuth angle of the electric toothbrush.

[0046] Step S104: The pitch angle, roll angle, azimuth angle, and ranging value are sent to the server, so that the server uses a target random forest model to determine the target oral cavity region corresponding to the brushing operation. The target random forest model is constructed based on the user's user sample dataset. The user sample dataset is collected during the user's calibration operation using the electric toothbrush according to the target standard brushing method. The user sample dataset includes sample triaxial acceleration, sample triaxial angular velocity, and sample triaxial magnetic force components collected using the nine-axis inertial unit, as well as sample ranging values ​​collected using the ranging sensor.

[0047] In this embodiment, a communication link, such as WiFi, 4G, or 5G, is pre-established between the electric toothbrush and the server. After determining the pitch angle, roll angle, azimuth angle, and distance measurement values, the electric toothbrush sends these values ​​to the server via the communication link. The server then uses a pre-established target random forest model to determine the target oral cavity region corresponding to the brushing operation based on the pitch angle, roll angle, azimuth angle, and distance measurement values. The target oral cavity region represents the brushing position of the electric toothbrush in the user's mouth.

[0048] Before using the electric toothbrush, users can calibrate it. Specifically, the user uses the electric toothbrush to calibrate according to the target standard brushing method. During this process, the electric toothbrush uses a nine-axis inertial unit to collect sample triaxial acceleration, sample triaxial angular velocity, and sample triaxial magnetic force components, as well as sample distance values ​​collected by a ranging sensor, to obtain a user sample dataset corresponding to the current user. The electric toothbrush then sends the user sample dataset to a server, allowing the server to construct a target forest model using the user sample dataset. This target forest model conforms to the user's oral cavity characteristics, achieving personalized modeling. Optionally, the target standard brushing method can be any one of the following: a 4-zone brushing method, a 6-zone brushing method, an 8-zone brushing method, a 12-zone brushing method, or a 16-zone brushing method (i.e., the Bass brushing method).

[0049] The server can then send the identifier or number of the target oral cavity area to the user's client. The user can then compare this identifier or number with the standard area corresponding to the standard brushing method to determine if the brushing method is accurate. Alternatively, the server can send the identifier or number of the target oral cavity area to the electric toothbrush, allowing the electric toothbrush to perform brushing operations according to the target brushing parameters (such as pressure, direction, etc.) corresponding to the target oral cavity area. Alternatively, the server can also send the identifier or number of the target oral cavity area to the electric toothbrush, which will then issue voice prompts corresponding to the target oral cavity area, allowing the user to identify any remaining uncleaned oral cavity areas based on the voice prompts.

[0050] In some embodiments of this application, the electric toothbrush includes a brush head, and the pitch angle, roll angle, azimuth angle, and distance measurement values ​​of the electric toothbrush are the pitch angle, roll angle, azimuth angle, and distance measurement values ​​of the brush head.

[0051] The oral cavity region recognition method in this embodiment responds to a user's brushing operation with an electric toothbrush. It utilizes the electric toothbrush's nine-axis inertial unit to obtain the electric toothbrush's three-axis acceleration, three-axis angular velocity, and three-axis magnetic force components, and uses the electric toothbrush's ranging sensor to obtain the ranging value. Based on the three-axis acceleration and three-axis angular velocity, it determines the electric toothbrush's pitch and roll angles; based on the three-axis magnetic force components, it determines the electric toothbrush's azimuth angle. The pitch, roll, azimuth, and ranging values ​​are sent to a server, allowing the server to use a target random forest model to determine the target oral cavity region corresponding to the brushing operation. By using the nine-axis inertial unit and ranging sensor in the electric toothbrush, dual positioning of spatial attitude and distance is achieved, enabling more accurate recognition of easily confused oral cavity regions such as left inner / right outer and right inner / left outer. By constructing a target random forest model based on the user's sample dataset, the target random forest model can be adapted to the user's oral cavity characteristics, achieving personalized modeling. This improves the accuracy and stability of oral cavity region recognition while providing personalized recognition capabilities tailored to different individual users, thus enhancing the user experience.

[0052] In some embodiments of this application, the pitch and roll angles of the electric toothbrush are determined based on the triaxial acceleration and the triaxial angular velocity, such as... Figure 2 As shown, it includes the following steps:

[0053] Step S1021: Determine the first pitch angle and the first roll angle using the triaxial acceleration.

[0054] In this embodiment, the first pitch angle and the first roll angle can be determined by the following formulas (1) and (2):

[0055]

[0056] in, Let be the first pitch angle at time t. Let be the first roll angle at time t. These represent the x-axis acceleration, y-axis acceleration, and z-axis acceleration at time t, respectively.

[0057] Step S1022: Determine the second pitch angle and the second roll angle using the triaxial angular velocities.

[0058] In this embodiment, the second pitch angle and the second roll angle can be determined by the following formulas (3) and (4):

[0059]

[0060] in, The second pitch angle at time t. The second roll angle at time t. Let x and y be the angular velocities along the x-axis and y-axis at time t, respectively. Let be the pitch angle at the initial time t0. dt is the roll angle at the initial time t0, and dt is the sampling time interval in seconds.

[0061] Step S1023: Determine the pitch angle based on the first pitch angle and the second pitch angle.

[0062] In this embodiment, after determining the first pitch angle and the second pitch angle, the first pitch angle and the second pitch angle are fused together to determine the pitch angle.

[0063] Step S1024: Determine the roll angle based on the first roll angle and the second roll angle.

[0064] In this embodiment, after determining the first roll angle and the second roll angle, the first roll angle and the second roll angle are merged to determine the roll angle.

[0065] The pitch angle is determined by the first pitch angle and the second pitch angle, and the roll angle is determined by the first roll angle and the second roll angle. By comprehensively considering the three-axis acceleration and the three-axis angular velocity, the data fusion of the accelerometer and gyroscope is achieved, which improves the accuracy of the pitch angle and roll angle.

[0066] In some embodiments of this application, determining the pitch angle based on the first pitch angle and the second pitch angle includes: performing a fusion process on the first pitch angle and the second pitch angle based on a first weight and the x-axis angular velocity in the three-axis angular velocities to determine the pitch angle;

[0067] The step of determining the roll angle based on the first roll angle and the second roll angle includes: performing a fusion process on the first roll angle and the second roll angle based on the second weight and the y-axis angular velocity in the three-axis angular velocities to determine the roll angle.

[0068] In this embodiment, the first weight and the second weight can be the same or different. The first pitch angle and the second pitch angle are fused using the first weight and the x-axis angular velocity from the three-axis angular velocities to obtain the pitch angle. The first roll angle and the second roll angle are fused using the second weight and the y-axis angular velocity from the three-axis angular velocities to determine the roll angle. This further improves the accuracy of the pitch angle and roll angle.

[0069] In some embodiments of this application, the pitch angle and roll angle are determined by fusion processing using the following formulas (5) and (6):

[0070]

[0071] in, and Let represent the pitch angle and roll angle at time t, respectively. The first and second weights are both α, and α can be 0.95 to 0.98.

[0072] In some embodiments of this application, the azimuth angle is determined by the following formula (7):

[0073]

[0074] in, Let m be the azimuth angle. y Let m be the y-axis magnetic force component. x The x-axis magnetic force component.

[0075] In some embodiments of this application, such as Figure 3 As shown, the identification method also includes the following steps:

[0076] Step S201: In response to the user performing the calibration operation using the electric toothbrush, sample data of each region corresponding to each standard oral cavity region is collected using the nine-axis inertial unit and the ranging sensor at the target sampling frequency. Each standard oral cavity region is the user's brushing region corresponding to the target standard brushing method.

[0077] In this embodiment, the user can calibrate the electric toothbrush upon first use or periodically. The user can enter calibration mode by pressing and holding the start button, or by pressing a calibration button on the toothbrush. In some embodiments of this application, after entering calibration mode, the electric toothbrush can issue a voice prompt reminding the user to brush their teeth according to the target standard brushing method, enabling the user to perform the calibration operation accurately.

[0078] In response to the user's calibration operation using an electric toothbrush, the system collects sample data for each region corresponding to a standard oral cavity area at the target sampling frequency using a nine-axis inertial unit and a ranging sensor. Each region's sample data includes the sample's triaxial acceleration, triaxial angular velocity, triaxial magnetic force components, and ranging value, corresponding to the standard oral cavity area. Each standard oral cavity area corresponds to the user's brushing area using the target standard brushing technique.

[0079] In some embodiments of this application, before collecting sample data of each region corresponding to each standard oral cavity region using the nine-axis inertial unit and the ranging sensor according to the target sampling frequency, the method further includes initializing the triaxial acceleration, triaxial angular velocity, triaxial magnetic force components, ranging value, pitch angle, roll angle, and azimuth angle, including setting:

[0080] Triaxial acceleration vector:

[0081] A = [a x =0,a y =0,a z =0];

[0082] Three-axis angular velocity vector:

[0083] G = [g x =0,g y =0,g z =0];

[0084] Three-axis magnetic force vector:

[0085] M = [m x =0,m y =0,m z =0];

[0086] Distance measurement value:

[0087] D = 0.

[0088] Record the pitch angle at the current moment. Roll angle Azimuth These are the initial attitude angle data.

[0089] Step S202: Add labels to the standard oral cavity regions to each of the aforementioned regional sample data to determine the user sample dataset.

[0090] In this embodiment, the labels for standard oral cavity regions can be numbers or text identifiers other than numbers. Labels for each region's sample data are added to determine the user sample dataset.

[0091] In some embodiments of this application, the target standard brushing method is the Bass brushing technique, and the standard oral cavity areas are as follows: Figure 7 The 16 oral cavity regions shown are sampled at a target frequency of 50Hz, meaning 5 complete data sets are collected per second. During calibration, the user brushes their teeth sequentially across the 16 regions, brushing each region for approximately 5 seconds, collecting 5 data points per second, for a total of 25 region samples. The entire calibration process collects a total of 16 region × 25 region samples = 400 region sample data points. Each region sample data point contains the following input features:

[0092] X = [A, G, M, D].

[0093] Each region sample data is also appended with the current dental region number as the output label Y∈{0,1,2,...,16}.

[0094] Step S203: Determine an attitude data set including sample pitch angle, sample roll angle, sample azimuth angle and sample ranging value based on the user sample dataset.

[0095] In this embodiment, the sample triaxial acceleration, sample triaxial angular velocity, and sample triaxial magnetic force components in the user sample dataset are converted into sample pitch angle, sample roll angle, and sample azimuth angle, and an attitude data set including sample pitch angle, sample roll angle, sample azimuth angle, and sample ranging value is determined.

[0096] Step S204: Send the attitude data set to the server, so that the server can use the attitude data set to construct the target random forest model.

[0097] The attitude data set is sent to the server, which then uses the attitude data set to build a target random forest model. When the user needs to use an electric toothbrush to brush their teeth, the server can use the target random forest model to determine the target oral cavity region corresponding to the brushing operation based on the pitch angle, roll angle, azimuth angle and range values ​​received from the electric toothbrush.

[0098] By obtaining a user sample dataset during calibration and determining a pose data set based on the user sample dataset, the pose data set is sent to the server, enabling the server to efficiently and accurately construct a target random forest model that conforms to the user's oral cavity characteristics.

[0099] In some embodiments of this application, determining the attitude data set including sample pitch angle, sample roll angle, sample azimuth angle, and sample ranging value based on the user sample dataset includes:

[0100] The pitch angle and roll angle of the electric toothbrush are determined based on the triaxial acceleration and triaxial angular velocity of the sample.

[0101] The sample azimuth angle of the electric toothbrush is determined based on the triaxial magnetic force components of the sample.

[0102] The attitude data set is determined based on the pitch angle, roll angle, azimuth angle, and distance measurement value of the sample corresponding to each standard oral cavity region.

[0103] In this embodiment, similar to steps S102-S103, the sample triaxial acceleration and sample triaxial angular velocity are converted to determine the sample pitch angle and sample roll angle of the electric toothbrush. The sample triaxial magnetic components are converted to determine the sample azimuth angle of the electric toothbrush. Then, the attitude data set is determined based on the sample pitch angle, sample roll angle, sample azimuth angle and sample ranging value corresponding to each standard oral cavity area. By comprehensively considering the sample triaxial acceleration and sample triaxial angular velocity, the data fusion of the accelerometer and gyroscope is realized, which improves the accuracy of the sample pitch angle and sample roll angle, and thus improves the accuracy of the attitude data set.

[0104] In some embodiments of this application, the sample pitch angle and sample roll angle, and the set of attitude angles at time t can be determined by referring to formulas (1)-(5).

[0105] This application also proposes a method for recognizing oral cavity regions, applied to a server, such as... Figure 4 As shown, it includes the following steps:

[0106] Step S301: Obtain the pitch angle, roll angle, azimuth angle, and distance value received from the electric toothbrush. The pitch angle, roll angle, and azimuth angle are determined based on the three-axis acceleration, three-axis angular velocity, and three-axis magnetic force component of the electric toothbrush. The three-axis acceleration, three-axis angular velocity, three-axis magnetic force component, and distance value are obtained by using the nine-axis inertial unit and distance sensor of the electric toothbrush during the user's brushing operation.

[0107] In this embodiment, during the brushing operation of the electric toothbrush, the electric toothbrush obtains three-axis acceleration, three-axis angular velocity, and three-axis magnetic force components through the nine-axis inertial unit, determines the distance value through the distance sensor, and determines the pitch angle, roll angle, and azimuth angle based on the three-axis acceleration, three-axis angular velocity, and three-axis magnetic force components. Then, the pitch angle, roll angle, azimuth angle, and distance value are sent to the server so that the server can obtain the pitch angle, roll angle, azimuth angle, and distance value.

[0108] Step S302: Based on the pitch angle, roll angle, azimuth angle, and ranging value, a target random forest model is used to determine the target oral cavity region corresponding to the brushing operation. The target random forest model is constructed based on the user's user sample dataset. The user sample dataset is collected during the user's calibration operation using the electric toothbrush according to the target standard brushing method. The user sample dataset includes sample triaxial acceleration, sample triaxial angular velocity, and sample triaxial magnetic force components collected using the nine-axis inertial unit, as well as sample ranging values ​​collected using the ranging sensor.

[0109] In this embodiment, before brushing their teeth, the user can calibrate the electric toothbrush. Specifically, the user uses the electric toothbrush to calibrate according to the target standard brushing method. During this process, the electric toothbrush uses a nine-axis inertial unit to collect sample triaxial acceleration, sample triaxial angular velocity, and sample triaxial magnetic force components, as well as sample distance values ​​collected by a ranging sensor, to determine the user sample dataset corresponding to the current user. Then, the electric toothbrush determines an attitude data set including sample pitch angle, sample roll angle, sample azimuth angle, and sample distance values ​​based on the user sample dataset. After obtaining the attitude data set from the electric toothbrush, the server constructs a target random forest model that conforms to the user's oral cavity characteristics based on the attitude data set.

[0110] After obtaining the pitch angle, roll angle, azimuth angle, and distance values ​​from the electric toothbrush, the server uses a target random forest model to determine the target oral cavity region corresponding to the brushing operation based on the pitch angle, roll angle, azimuth angle, and distance values.

[0111] Optionally, the target standard brushing method can be any one of the following: 4-zone brushing method, 6-zone brushing method, 8-zone brushing method, 12-zone brushing method, and 16-zone brushing method (i.e., Bass brushing method).

[0112] The oral cavity region recognition method of this application obtains pitch angle, roll angle, azimuth angle, and distance values ​​received from an electric toothbrush. Based on these values, a target random forest model is used to determine the target oral cavity region corresponding to the brushing operation. This target random forest model is constructed based on a user sample dataset collected during the user's calibration process using the electric toothbrush according to the target standard brushing method. By constructing the target random forest model based on the user sample dataset, the model can adapt to the user's oral cavity characteristics, achieving personalized modeling. This improves the accuracy and stability of oral cavity region recognition while providing personalized recognition capabilities tailored to different individual users, thus enhancing the user experience.

[0113] In some embodiments of this application, before determining the target oral cavity region corresponding to the brushing operation using a target random forest model based on the pitch angle, roll angle, azimuth angle, and ranging value, such as Figure 5 As shown, it also includes the following steps:

[0114] Step S401: Obtain a set of attitude data received from the electric toothbrush, including sample pitch angle, sample roll angle, sample azimuth angle, and sample distance measurement value. The set of attitude data is determined by the electric toothbrush based on the user sample dataset.

[0115] In this embodiment, in response to a user performing a calibration operation using an electric toothbrush, the electric toothbrush uses a nine-axis inertial unit and a ranging sensor to collect a user sample dataset. The toothbrush then converts the sample triaxial acceleration, sample triaxial angular velocity, and sample triaxial magnetic force components from the user sample dataset into sample pitch angle, sample roll angle, and sample azimuth angle, determining an attitude data set including sample pitch angle, sample roll angle, sample azimuth angle, and sample ranging value. The server obtains this attitude data set from the electric toothbrush and subsequently uses it to construct a target random forest model.

[0116] Step S402: Divide the attitude data set into a training set and a test set according to the target division ratio.

[0117] In this embodiment, to prevent model overfitting, the pose dataset is divided into a training set and a test set according to a target partitioning ratio. For example, the target partitioning ratio could be 80% of the pose dataset as the training set and 20% as the test set. In some embodiments of this application, stratified sampling is used to divide the pose dataset into training and test sets, thereby ensuring that the proportion of each category in the training and test sets is consistent with the original dataset.

[0118] Step S403: Standardize the training set and the test set to obtain a standardized training set and a standardized test set.

[0119] In this embodiment, in order to eliminate the differences in the units of measurement between data and improve the efficiency of model training, the training set and the test set are standardized to obtain a standardized training set and a standardized test set.

[0120] In some embodiments of this application, z-score standardization is used to standardize the data so that the mean of each feature is 0 and the variance is 1. For example, for the training set, the following formula (8) is used for standardization:

[0121]

[0122] Where X' j X is the result of standardizing the j-th sample data. j Let be the j-th sample data, μ be the mean of the training set features, and σ be the standard deviation of the training set features.

[0123] Step S404: Construct the target random forest model using the standardized training set and the standardized test set.

[0124] In this embodiment, a standardized training set is used to construct and train a random forest model, and a standardized test set is used to test the trained random forest model to obtain a target random forest model, thereby achieving efficient and accurate construction of a target random forest model that conforms to the user's oral cavity characteristics.

[0125] In some embodiments of this application, constructing the target random forest model using the standardized training set and the standardized test set may include the following process:

[0126] Step S4041, randomly select a feature subset: The total number of features is M = 4. At each node split, m ≤ M features are randomly selected from these 4 features (usually...). Instead of examining all features, we examine only the trees. The goal is to reduce the correlation between trees and enhance the generalization ability of the ensemble.

[0127] Step S4042, calculate the optimal split point for each candidate feature: For any node and each of its selected features f, try different split thresholds and calculate the Gini impurity for each split:

[0128]

[0129] Where D is a node, the standard brushing method is the Bass brushing method, K is the number of categories (16), and p k denoted as the proportion of the k-th class of samples in the node.

[0130] Step S4043, Weighted Gini coefficient after feature segmentation: For a candidate feature A j At a certain split point s, we divide the sample set D into two subsets:

[0131] Left subset (less than or equal to the split point):

[0132]

[0133] Right subset (greater than the split point):

[0134]

[0135] Then the weighted Gini coefficient at this point is:

[0136]

[0137] Step S4044, Select the optimal split: The goal is to find the optimal feature A* and the corresponding split point s*, which minimizes the weighted Gini coefficient, i.e.:

[0138]

[0139] Step S4045, Recursive Splitting: Repeat steps S4042-S4044 until one of the following stopping conditions is met: number of samples < minimum number of splits (e.g., min_samples_split = 2); Gini value is 0 (pure nodes); tree depth reaches maximum depth (e.g., max_depth = 10).

[0140] Step S4046, Model Ensemble: After training N trees, a complete random forest model is formed. For the test sample c in the test set, each tree is given a predicted label T. i (c), the final output is:

[0141]

[0142] Step S4047, Model Evaluation: For each category Y∈{0,1,2,...,16}, we can define: TruePositive(TP) Y ): Predicted as class Y, and actually is Y; FalsePositive(FP) Y ): Predicted as Y, but not actually Y; FalseNegative(FN) Y ): The actual value is Y, but it is predicted to be another category;

[0143] The model evaluation score F1-Score can then be expressed as:

[0144]

[0145] Precision Y This indicates how many samples, out of all those predicted as class Y, actually belong to class Y. Recall Y This indicates how many of the samples that truly belong to category Y were correctly predicted.

[0146] The effectiveness of the target random forest model can be determined by the model evaluation score. If the model evaluation score does not reach the target score, the parameters of the target random forest model can be adjusted to improve the model evaluation score.

[0147] In some embodiments of this application, after determining the target oral cavity region corresponding to the brushing operation using a target random forest model based on the pitch angle, roll angle, azimuth angle, and ranging value, as follows: Figure 6 As shown, it also includes the following steps:

[0148] Step S501: Determine the data frame corresponding to the pitch angle, the roll angle, the azimuth angle, and the ranging value as the current data frame.

[0149] In this embodiment, the electric toothbrush collects triaxial acceleration, triaxial angular velocity, triaxial magnetic force components and ranging values ​​at the target sampling frequency, and sends the corresponding sets of pitch angle, roll angle, azimuth angle and ranging values ​​to the server in sequence. The server determines each set of data received in sequence as a data frame, and each data frame includes a set of pitch angle, roll angle, azimuth angle and ranging values.

[0150] Step S502: Determine at least one historical data frame preceding the current data frame, and determine the historical target oral cavity region corresponding to the historical data frame.

[0151] In this embodiment, at least one historical data frame is determined and the historical target oral cavity region corresponding to the historical data frame is determined.

[0152] Step S503: When the number of current data frames and historical data frames reaches the target number, and the current data frames and historical data frames form a continuous data frame, and the historical target oral cavity region is the same as the target oral cavity region, output the region identifier of the target oral cavity region.

[0153] In this embodiment, if the number of current data frames and historical data frames reaches the target number, and the current data frame and historical data frames form consecutive data frames, and the historical target oral cavity region is the same as the target oral cavity region, it indicates that the recognition result is stable and effective. The region identifier of the target oral cavity region is then output, thereby improving the accuracy of the output recognition result. For example, if the target number is 3 frames, and the target oral cavity region corresponding to the current data frame is the maxillary anterolateral region, then the two historical data frames preceding the current data frame are identified. These two historical data frames are consecutive to the current data frame (3 data frames). If the historical oral cavity regions corresponding to the two previous historical data frames are both the maxillary anterolateral region, then the region identifier of the maxillary anterolateral region is output as the recognition result.

[0154] This application also provides an electric toothbrush, such as... Figure 8 As shown, it includes a nine-axis inertial unit, a ranging sensor, and a controller, wherein the controller is configured to:

[0155] In response to a user brushing their teeth with the electric toothbrush, the three-axis acceleration, three-axis angular velocity, and three-axis magnetic force components of the electric toothbrush are obtained using the nine-axis inertial unit.

[0156] The distance measurement value of the electric toothbrush is obtained using the distance measurement sensor;

[0157] The pitch angle and roll angle of the electric toothbrush are determined based on the triaxial acceleration and the triaxial angular velocity.

[0158] The azimuth angle of the electric toothbrush is determined based on the triaxial magnetic force components.

[0159] The pitch angle, roll angle, azimuth angle, and ranging value are sent to the server, which then uses a target random forest model to determine the target oral cavity region corresponding to the brushing operation. The target random forest model is constructed based on the user's user sample dataset, which is collected during the user's calibration operation using the electric toothbrush according to the target standard brushing method. The user sample dataset includes sample triaxial acceleration, sample triaxial angular velocity, and sample triaxial magnetic force components collected using the nine-axis inertial unit, as well as sample ranging values ​​collected using the ranging sensor.

[0160] The electric toothbrush of this application embodiment achieves dual positioning of spatial attitude and distance through a nine-axis inertial unit and a ranging sensor, which can more accurately identify easily confused oral regions such as left inner / right outer and right inner / left outer. By constructing a target random forest model based on the user's user sample dataset, the target random forest model can be adapted to the user's oral features, realizing personalized modeling. This improves the accuracy and stability of oral region recognition while providing personalized recognition capabilities adapted to different individual users, thus enhancing the user experience.

[0161] This application also proposes a computer device, such as... Figure 9 As shown, it includes a processor, a memory, and a computer program stored in the memory. The processor executes the computer program to implement the steps of the oral cavity region identification method as described in various embodiments of this application.

[0162] The computer device in this application embodiment can be a terminal or other devices besides a terminal. For example, the computer device can be a mobile phone, tablet computer, laptop computer, handheld computer, in-vehicle electronic device, mobile internet device (MID), augmented reality (AR) / virtual reality (VR) device, robot, wearable device, ultra-mobile personal computer (UMPC), netbook, or personal digital assistant (PDA), etc. It can also be a server, network attached storage (NAS), personal computer (PC), television (TV), ATM, or self-service machine, etc. The embodiments disclosed in this disclosure do not impose specific limitations.

[0163] The memory may include RAM (Random Access Memory) or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0164] The processors mentioned above can be general-purpose processors, including CPUs, NPs (Network Processors), etc.; they can also be DSPs (Digital Signal Processors), ASICs (Application Specific Integrated Circuits), FPGAs (Field Programmable Gate Arrays), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0165] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0166] The above embodiments are merely exemplary embodiments of this application and are not intended to limit this application. The scope of protection of this application is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to this application within its substance and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of this application.

Claims

1. A method for identifying oral cavity regions, characterized in that, Applications in electric toothbrushes, including: In response to a user brushing their teeth with the electric toothbrush, the three-axis acceleration, three-axis angular velocity, and three-axis magnetic force components of the electric toothbrush are obtained using the nine-axis inertial unit of the electric toothbrush, and the distance value of the electric toothbrush is obtained using the distance sensor of the electric toothbrush. The pitch angle and roll angle of the electric toothbrush are determined based on the triaxial acceleration and the triaxial angular velocity. The azimuth angle of the electric toothbrush is determined based on the triaxial magnetic force components. The pitch angle, roll angle, azimuth angle, and ranging value are sent to the server, which then uses a target random forest model to determine the target oral cavity region corresponding to the brushing operation. The target random forest model is constructed based on the user's user sample dataset, which is collected during the user's calibration operation using the electric toothbrush according to the target standard brushing method. The user sample dataset includes sample triaxial acceleration, sample triaxial angular velocity, and sample triaxial magnetic force components collected using the nine-axis inertial unit, as well as sample ranging values ​​collected using the ranging sensor.

2. The method for identifying oral cavity regions as described in claim 1, characterized in that, The step of determining the pitch and roll angles of the electric toothbrush based on the triaxial acceleration and the triaxial angular velocity includes: The first pitch angle and the first roll angle are determined using the triaxial acceleration. The second pitch angle and the second roll angle are determined using the three-axis angular velocities. The pitch angle is determined based on the first pitch angle and the second pitch angle; The roll angle is determined based on the first roll angle and the second roll angle.

3. The method for identifying oral cavity regions as described in claim 2, characterized in that, The step of determining the pitch angle based on the first pitch angle and the second pitch angle includes: performing a fusion process on the first pitch angle and the second pitch angle based on the first weight and the x-axis angular velocity in the three-axis angular velocities to determine the pitch angle; The step of determining the roll angle based on the first roll angle and the second roll angle includes: performing a fusion process on the first roll angle and the second roll angle based on the second weight and the y-axis angular velocity in the three-axis angular velocities to determine the roll angle.

4. The method for identifying oral cavity regions as described in claim 1, characterized in that, Also includes: In response to the user performing the calibration operation using the electric toothbrush, the nine-axis inertial unit and the ranging sensor collect sample data of each region corresponding to each standard oral cavity region according to the target sampling frequency. Each standard oral cavity region is the brushing region of the user corresponding to the target standard brushing method. Add labels to the standard oral cavity regions to the sample data of each region to determine the user sample dataset; Based on the user sample dataset, determine an attitude data set including sample pitch angle, sample roll angle, sample azimuth angle, and sample range value; The attitude data set is sent to the server, so that the server can use the attitude data set to construct the target random forest model.

5. The method for identifying oral cavity regions as described in claim 4, characterized in that, The determination of the attitude data set based on the user sample dataset, including sample pitch angle, sample roll angle, sample azimuth angle, and sample ranging values, includes: The pitch angle and roll angle of the electric toothbrush are determined based on the triaxial acceleration and triaxial angular velocity of the sample. The sample azimuth angle of the electric toothbrush is determined based on the triaxial magnetic force components of the sample. The attitude data set is determined based on the pitch angle, roll angle, azimuth angle, and distance measurement value of the sample corresponding to each standard oral cavity region.

6. A method for identifying oral cavity regions, characterized in that, Applied to servers, including: The pitch angle, roll angle, azimuth angle, and distance measurement value received from the electric toothbrush are obtained. The pitch angle, roll angle, and azimuth angle are determined based on the triaxial acceleration, triaxial angular velocity, and triaxial magnetic force component of the electric toothbrush. The triaxial acceleration, triaxial angular velocity, triaxial magnetic force component, and distance measurement value are obtained by the electric toothbrush using its nine-axis inertial unit and distance sensor during the user's brushing operation. Based on the pitch angle, roll angle, azimuth angle, and ranging value, a target random forest model is used to determine the target oral cavity region corresponding to the brushing operation. The target random forest model is constructed based on the user's user sample dataset, which is collected during the user's calibration operation using the electric toothbrush according to the target standard brushing method. The user sample dataset includes sample triaxial acceleration, sample triaxial angular velocity, and sample triaxial magnetic force components collected using the nine-axis inertial unit, as well as sample ranging values ​​collected using the ranging sensor.

7. The method for identifying oral cavity regions as described in claim 6, characterized in that, Before determining the target oral cavity region corresponding to the brushing operation using a target random forest model based on the pitch angle, roll angle, azimuth angle, and distance measurement value, the method further includes: Obtain a set of attitude data received from the electric toothbrush, including sample pitch angle, sample roll angle, sample azimuth angle and sample range value, wherein the set of attitude data is determined by the electric toothbrush based on the user sample dataset; The attitude data set is divided into a training set and a test set according to the target division ratio; The training set and the test set are standardized to obtain a standardized training set and a standardized test set. The target random forest model is constructed using the standardized training set and the standardized test set.

8. The method for identifying oral cavity regions as described in claim 6, characterized in that, After determining the target oral cavity region corresponding to the brushing operation using a target random forest model based on the pitch angle, roll angle, azimuth angle, and distance measurement value, the method further includes: The data frame corresponding to the pitch angle, the roll angle, the azimuth angle, and the ranging value is determined as the current data frame; Determine at least one historical data frame preceding the current data frame, and determine the historical target oral cavity region corresponding to the historical data frame; When the number of current data frames and historical data frames reaches the target number, and the current data frames and historical data frames form consecutive data frames, and the historical target oral cavity region is the same as the target oral cavity region, the region identifier of the target oral cavity region is output.

9. An electric toothbrush, characterized in that, It includes a nine-axis inertial unit, a ranging sensor, and a controller, wherein the controller is configured to: In response to a user brushing their teeth with the electric toothbrush, the three-axis acceleration, three-axis angular velocity, and three-axis magnetic force components of the electric toothbrush are obtained using the nine-axis inertial unit, and the distance value of the electric toothbrush is obtained using the distance sensor. The pitch angle and roll angle of the electric toothbrush are determined based on the triaxial acceleration and the triaxial angular velocity. The azimuth angle of the electric toothbrush is determined based on the triaxial magnetic force components. The pitch angle, roll angle, azimuth angle, and ranging value are sent to the server, which then uses a target random forest model to determine the target oral cavity region corresponding to the brushing operation. The target random forest model is constructed based on the user's user sample dataset, which is collected during the user's calibration operation using the electric toothbrush according to the target standard brushing method. The user sample dataset includes sample triaxial acceleration, sample triaxial angular velocity, and sample triaxial magnetic force components collected using the nine-axis inertial unit, as well as sample ranging values ​​collected using the ranging sensor.

10. A computer device comprising a processor, a memory, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the oral cavity region identification method as described in any one of claims 1-5 or 6-8.

Images