Back to results
US 12,653,650 B2Granted
System to remove from an image unnecessary objects identified by an intraoral scanner
AssigneeJ. Morita Mfg. Corp.
InventorsRyosuke Kaji, Mikinori Nishimura
Life of this patent
Estimating the term…
Abstract
An identification apparatus includes an input interface to which position data including position information of each of a plurality of objects is input, and processing circuitry that is configured to identify at least one object among the plurality of objects based on the position data that is input from the input interface and an estimation model that is trained to estimate a type of each of the plurality of objects based at least on the position data of each of the plurality of objects, the position data including a relative positional relationship which is fixed inside the mouth.
Claims14
An identification apparatus for identifying a plurality of objects inside a mouth, the identification apparatus comprising:
- an input interface to which position data including position information of each of the plurality of objects is input;
- and
- processing circuitry configured to identify at least one object among the plurality of objects based on the position data that is input from the input interface and based on an estimation model, the estimation model being trained to estimate a type of each of the plurality of objects based at least on the position data of each of the plurality of objects, the position data including a relative positional relationship which is fixed inside the mouth,
- wherein the estimation model is trained to estimate the type of each of the plurality of objects based on the position data by using training data including the position data, the position data in the training data including the position information of each of the plurality of objects and a ground truth label indicating each of the plurality of objects, and
- wherein the training data includes a positional relationship label indicating a relative positional relationship between the plurality of objects, and the positional relationship label is associated with the position data of each of the plurality of objects in the training data,
- wherein the positional relationship label includes data that indicates:
- a first region where a tooth is present, a second region where a tongue is present, a third region where a lip is present, a fourth region between the first region and the second region, and a fifth region between the first region and the third region, and
- wherein the estimation model is trained to estimate the type of each of the plurality of objects based on the position data by using training data having the positional relationship label including the first, second, third, fourth and fifth regions, which is used for indicating the relative position between the plurality of objects.
5 more elements
An identification method of identifying a plurality of objects inside a mouth, the identification method comprising:
- receiving, by processing circuitry, position data including position information of each of the plurality of objects;
- and
- identifying, by the processing circuitry, at least one object among the plurality of objects based on the position data that is received and based on an estimation model, the estimation model being that is trained to estimate a type of each of the plurality of objects based at least on the position data of each of the plurality of objects, the position data including a relative positional relationship which is fixed inside the mouth,
- wherein the estimation model is trained to estimate the type of each of the plurality of objects based on the position data by using training data including the position data, the position data in the training data including the position information of each of the plurality of objects and a ground truth label indicating each of the plurality of objects, and
- wherein the training data includes a positional relationship label indicating a relative positional relationship between the plurality of objects, and the positional relationship label is associated with the position data of each of the plurality of objects in the training data,
- wherein the positional relationship label includes data that indicates:
- a first region where a tooth is present, a second region where a tongue is present, a third region where a lip is present, a fourth region between the first region and the second region, and a fifth region between the first region and the third region, and
- wherein the estimation model is trained to estimate the type of each of the plurality of objects based on the position data by using training data having the positional relationship label including the first, second, third, fourth and fifth regions, which is used for indicating the relative position between the plurality of objects.
5 more elements
Dependencies are inferred from the claim text.