KR20200129423A - Method and Apparatus for detecting of Individual Tooth in Teeth Image - Google Patents

Abstract

Provided is a method for detecting an individual tooth in a tooth image which comprises the steps of: generating an average image by averaging a plurality of slices in a Z-axis direction from a tooth image including the plurality of slices; generating a seed image separated between individual teeth by binarizing the average image; and detecting the individual teeth from the tooth image using the seed image.

Description

TECHNICAL FIELD [Method and Apparatus for detecting of Individual Tooth in Teeth Image}

The present invention relates to a method and apparatus for detecting individual teeth in a tooth image.

In general, an implant refers to a substitute that can replace human tissue when the original human tissue is lost, and in particular, in dentistry, including a fixture, an abutment, and a crown. It means implanting an artificial tooth using a prosthesis at the actual tooth position.

In the dental implant procedure, a perforation is formed in the alveolar bone and a fixture is placed in the perforation, and when the fixture is fused to the alveolar bone, the abutment and crown are combined with the fixture. In order to prevent problems such as nerve damage, these implant procedures determine the placement position, angle, and depth of the implant based on a cone beam computed tomography (CBCT) image before the implant procedure. Prior to confirmation. In this way, after clearly separating the teeth and the alveolar bone, it is possible to design for accurate implant placement by configuring 3D image data.

On the other hand, for accurate implant placement, a method of dividing a tooth in a CBCT image is currently being discussed, but even if a tooth is divided in a CBCT image, all the individual teeth are united into one, so there is a problem that information on individual teeth cannot be provided .

Embodiments of the present invention for solving this problem are a method of generating a seed image that is completely separated between individual teeth, and detecting individual teeth from a tooth image capable of detecting individual teeth from a tooth image based on the seed image. And to provide an apparatus.

In order to achieve the above object, the present invention generates an average image by averaging a plurality of slices in a Z-axis direction from a tooth image including a plurality of slices, and a seed image separated between individual teeth by binarizing the average image. It provides a method of detecting individual teeth in a tooth image, including generating and detecting individual teeth in the tooth image using the seed image.

In addition, the method of detecting individual teeth in a tooth image according to the present invention may further include generating an extended seed image by performing a morphological expansion operation for each of a plurality of blobs included in the seed image.

Here, the step of generating the average image may be a step of generating an average image by correcting a tooth inclination from the tooth image.

In addition, generating the average image may be a step of generating an average image by removing noise blobs from the tooth image.

In addition, generating the average image may be a step of generating an average image by correcting a tooth inclination from the tooth image and removing noise blobs from the tooth image.

In addition, generating the average image may be a step of generating a plurality of average images by averaging the tooth images under different conditions.

In addition, generating the seed image may be a step of generating a seed image by selecting any one of a plurality of average images based on the number and width of effective blobs of the average image.

In addition, the generating of the seed image includes generating a plurality of candidate seed images by binarizing the plurality of average images, and selecting a seed image from the plurality of candidate seed images based on the number and width of effective blobs of the candidate seed image. It may include the step of.

In addition, generating the seed image may include setting a reference value for binarizing the average image.

Here, the step of setting the reference value may be a step of setting the reference value based on the number and width of effective blobs of the seed image.

In addition, the step of detecting an individual tooth may include assigning an index to each individual tooth included in the seed image, analyzing a connection component for each of a plurality of blobs for a plurality of slices included in the tooth image, and giving an index. I can.

In addition, the step of assigning the index for each of the plurality of blobs may be a step of assigning an index according to an overlap ratio between a blob included in a slice and an individual tooth included in the seed image.

In addition, the present invention includes an average image generator for generating an average image by averaging a plurality of slices in a Z-axis direction from a tooth image composed of a plurality of slices, and a seed image separated between individual teeth by binarizing the average image. An apparatus for detecting individual teeth in a tooth image, including a seed image generator and an individual tooth detector for detecting individual teeth in the tooth image by using the seed image.

As described above, in the method and apparatus for detecting individual teeth in a tooth image according to the present invention, a seed image that is completely separated between individual teeth is generated, and an individual tooth can be detected from the tooth image based on the seed image. As a result, there is an effect of improving the accuracy of individual tooth detection.

1 is a diagram illustrating a main configuration of an electronic device for detecting individual teeth in a tooth image according to an embodiment of the present invention.
2 is a diagram illustrating in detail a control unit of an electronic device for detecting individual teeth in a tooth image according to an exemplary embodiment of the present invention.
3 is a flowchart of a method for generating a plurality of average images by an average image generator according to an embodiment of the present invention.
4 is a diagram illustrating a plurality of seed images generated by a seed image generator according to an embodiment of the present invention.
5 is a diagram illustrating a seed image and an extended seed image generated by a seed image generator according to an embodiment of the present invention.
6 is a diagram illustrating a method of detecting an individual tooth from a tooth image by using a seed image by the individual tooth detector 154 according to an exemplary embodiment of the present invention.
7 is a flowchart illustrating a method of detecting individual teeth in a tooth image according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The detailed description to be disclosed hereinafter together with the accompanying drawings is intended to describe exemplary embodiments of the present invention, and is not intended to represent the only embodiments in which the present invention may be practiced. In the drawings, parts irrelevant to the description may be omitted in order to clearly describe the present invention, and the same reference numerals may be used for the same or similar components throughout the specification.

In an embodiment of the present invention, expressions such as "or" and "at least one" may represent one of words listed together, or a combination of two or more. For example, “A or B” and “at least one of A and B” may include only one of A or B, and may include both A and B.

1 is a diagram illustrating a main configuration of an electronic device for detecting individual teeth in a tooth image according to an embodiment of the present invention.

As illustrated in FIG. 1, the electronic device 100 according to the present invention may include a communication unit 110, an input unit 120, a display unit 130, a memory 140, and a control unit 150.

The communication unit 110 may transmit or receive tooth image data by performing communication with external devices such as an image acquisition device (not shown) and a server (not shown). To this end, the communication unit 110 5G (5 ^th generation communication), LTE-A (long term evolution-advanced), LTE (long term evolution), Bluetooth, BLE (bluetooth low energe), NFC (near field communication), etc. Wireless communication can be performed, and wired communication such as cable communication can be performed.

The input unit 120 generates input data in response to a user input of the electronic device 100. The input unit 120 includes at least one input means. Input unit 120 is a keyboard (key board), keypad (key pad), dome switch (dome switch), touch panel (touch panel), touch key (touch key), mouse (mouse), menu button (menu button), etc. It may include.

The display unit 130 displays tooth image data according to an operation of the electronic device 100. The display unit 130 includes a liquid crystal display (LCD), a light emitting diode (LED) display, an organic light emitting diode (OLED) display, and a micro electro mechanical systems (MEMS) display. Including ray and electronic paper displays. The display unit 130 may be combined with the input unit 120 to be implemented as a touch screen.

The memory 140 stores operation programs of the electronic device 100. The memory 140 may store an algorithm related to a convolutional neural network (CNN) such as U-Net and UDS-Net. The memory 140 may store a plurality of image data received from an image acquisition device or the like.

The controller 150 performs a series of operation processes for detecting individual teeth from tooth image data.

2 is a diagram illustrating in detail a control unit of an electronic device for detecting individual teeth in a tooth image according to an exemplary embodiment of the present invention.

As shown in FIG. 2, the control unit 150 may include an average image generation unit 152, a seed image generation unit 153, and an individual tooth detection unit 154.

The average image generator 152 generates an average image by averaging a plurality of slices in a Z-axis direction from a tooth image (eg, a cone beam computed tomography (CBCT) image) including a plurality of slices.

On the other hand, there may be cases where the teeth are inclined in the captured teeth image because the imaging conditions are different for each patient, and in this case, due to the overlapping problem between individual teeth, the average between individual teeth is relatively well separated. It may be difficult to generate an image or a seed image.

In order to solve this problem, the control unit 150 may further include an image correction unit 151.

The image correction unit 151 estimates the tooth slope using a tooth slope estimation algorithm, and based on the tooth slope, calculates the slope of the tooth image to generate an average image with relatively good separation between individual teeth. Correct. Accordingly, the average image generation unit 152 may generate an average image by averaging the slices whose inclination is corrected to generate an average image in which individual teeth are relatively well separated.

Hereinafter, a tooth tilt estimation algorithm will be described.

First, a slice having a certain number of effective blobs (eg, 4) or more is obtained from a tooth image. Here, an index may be allocated to valid blobs included in the obtained slice. For example, the effective blob may be a blob having 80 or more pixels, an ellipse width-to-pixel ratio of 0.8 to 1.3, and an aspect ratio of 0.7 to 1.6. In addition, since the effective blob that satisfies such a condition has a relatively high probability of teeth, it is preferable to estimate the inclination of the teeth based on the effective blob.

Next, tracking is performed for each effective blob acquired in each slice. For example, check the overlapping ratio between the effective blob acquired in the current slice and the effective blob acquired in the next slice, and if the overlap ratio is 80% or more, the index allocated to the effective blob of the current slice is transferred to the effective blob of the next slice. Tracking is carried out in a manner.

Next, while tracking is performed, the amount of change in the center point of the effective blob included in each slice is measured.

Meanwhile, when the overlapping ratio is less than a certain level, tracking may be automatically terminated.

Next, when tracking is finished for each slice, the z-axis and y-axis changes of the center points of the first and last effective blobs are calculated.

Next, after adding all the changes in the z-axis and y-axis of the center point of the effective blob obtained as a result of tracking all effective blobs, the tooth inclination is estimated based on this.

The image correction unit 151 may remove noise blobs from the tooth image. Accordingly, the average image generator 152 may generate an average image by averaging the slices from which the noise blobs have been removed, thereby generating an average image in which individual teeth are relatively well separated. That is, the average image generator 152 may generate an average image by averaging only effective blobs from which noise blobs have been removed from the tooth image.

Here, the effective blob may have different conditions from the effective blob of the above-described slope estimation algorithm. For example, the effective blob may be a blob having 50 or more pixels and a ratio of the ellipse width to the number of pixels of 0.8 to 1.3.

Meanwhile, the average image generator 152 may generate an average image by correcting a tooth inclination from the tooth image and removing noise blobs from the tooth image.

3 is a flowchart of a method for generating a plurality of average images by an average image generator according to an embodiment of the present invention.

The average image generator 152 may generate a plurality of average images by averaging the tooth images under different conditions. That is, referring to FIG. 3, the average image generator 152 includes an uncorrected tooth image (a), a tooth image from which noise blobs are removed (b), a tooth image (c) with corrected tooth tilt, and a tooth. A total of four z-axis average images may be generated based on the tooth image (d) from which the inclination is corrected and noise blobs are removed.

4 is a diagram illustrating a plurality of seed images generated by a seed image generator according to an embodiment of the present invention.

Referring to FIG. 4, the seed image generator 153 generates a seed image separated between individual teeth by binarizing the average image.

(A), (b), (c) and (d) shown in FIG. 4 are a tooth image that is not corrected, a tooth image from which noise blobs are removed, a tooth image with corrected tooth tilt, and a tooth tilt is corrected. And the z-axis average image averaged based on the tooth image from which the noise blob was removed. In addition, (a'), (b'), (c'), and (d') are seed images generated by binarizing the average images of (a), (b), (c) and (d), respectively.

The seed image generator 153 may set a reference value for binarizing the average image in an adaptive manner. Here, the reference value may be set based on the number and width of effective blobs of the seed image. That is, based on the number and area of the effective blobs of the seed image obtained by binarizing the average image while varying the reference value, the reference function is designed according to the following [Equation 1], and the reference value at the point where the reference function value becomes the maximum It is possible to determine a reference value for binarizing the average image.

[Equation 1]

X=

Here, N is the number of effective blobs, and A is the area of the effective blobs.

The seed image generator 153 may generate a seed image by selecting any one of a plurality of average images generated by the average image generator 152 based on the number and width of the effective blobs of the average image.

In contrast, the seed image generator 153 generates a plurality of candidate seed images by binarizing the plurality of average images, respectively, and generates a seed image from the plurality of candidate seed images based on the number and width of effective blobs of the candidate seed image. You can also choose.

That is, the seed image generator 153 may select any one of the (a'), (b'), (c'), and (d') seed images shown in FIG. 4. On the other hand, it is impossible to uniformly determine which of the (a'), (b'), (c') and (d') seed images has better separation between individual teeth, and the effective blob of the candidate seed image It is desirable to judge individually based on the number and area.

5 is a diagram illustrating a seed image and an extended seed image generated by a seed image generator according to an embodiment of the present invention.

As shown in FIG. 5, the seed image generator 153 may generate an extended seed image b by performing a morphological expansion operation for each of a plurality of blobs included in the seed image a. Accordingly, there is no empty space between individual teeth, and it is possible to clearly distinguish between individual teeth. Meanwhile, the nearest individual tooth index may be allocated to an area where the overlapping between individual teeth occurs during the morphological expansion operation process.

6 is a diagram illustrating a method of detecting an individual tooth from a tooth image by using a seed image by the individual tooth detection unit 154 according to an exemplary embodiment of the present invention.

The individual tooth detection unit 154 detects individual teeth from a tooth image using a seed image or an extended seed image (hereinafter, these are collectively referred to as a seed image).

Specifically, referring to FIG. 6, the individual tooth detection unit 154 allocates an index for each individual tooth included in the seed image, and analyzes a connection component for each of a plurality of blobs for a plurality of slices included in the tooth image to determine the index. By applying, individual teeth are detected in the tooth image.

Here, the individual tooth detection unit 154 may assign an index according to an overlap ratio between the blob included in the slice and the individual teeth included in the seed image. For example, when the overlapping ratio is 90% or more, the index of the individual tooth may be assigned to the corresponding blob.

On the other hand, when two or more teeth are attached to each other or when peripheral teeth penetrate into a specific tooth, the overlapping ratio may be less than 90%.In this case, the number of pixels of the blob is calculated for each individual tooth index and the histogram is analyzed to determine the nearest Individual tooth indexes can be assigned.

In this way, the apparatus for detecting individual teeth in a tooth image according to an embodiment of the present invention generates a seed image that is completely separated between individual teeth, and can detect individual teeth in the tooth image based on the seed image. As a result, there is an effect of improving the accuracy of individual tooth detection.

7 is a flowchart illustrating a method of detecting individual teeth in a tooth image according to an embodiment of the present invention.

Hereinafter, a method of detecting an individual tooth in a tooth image according to an exemplary embodiment of the present invention will be described with reference to FIGS. 1 to 7, but overlapping content with the foregoing will be omitted.

Referring to FIG. 7, a method of detecting individual teeth in a tooth image according to an embodiment of the present invention includes obtaining a tooth image (S10), correcting a tooth image (S20), and generating an average image ( S30), generating a seed image (S40), and detecting an individual tooth (S50).

First, a tooth image (eg, a CBCT image) data is transmitted by performing communication with an external device such as an image acquisition device (not shown) and a server (not shown) (S10).

Next, the average image is generated by averaging the plurality of slices in the Z-axis direction from the tooth image including the plurality of slices (S30). Here, a plurality of average images may be generated by averaging tooth images under different conditions.

Specifically, the average image is generated by correcting the tooth inclination from the tooth image, or by removing the noise blob from the tooth image to generate an average image, or by correcting the tooth inclination from the tooth image and removing the noise blob from the tooth image. Can be generated (S20).

Next, the average image is binarized to generate a seed image separated between individual teeth (S40).

Here, a seed image may be generated by selecting any one of a plurality of average images based on the number and width of the effective blobs of the average image.

Alternatively, a plurality of candidate seed images may be generated by binarizing a plurality of average images, and a seed image may be selected from a plurality of candidate seed images based on the number and width of effective blobs of the candidate seed image.

In addition, in order to generate a seed image with relatively good separation between individual teeth, a reference value for binarizing the average image may be set in an adaptive manner.

Such a reference value may be set based on the number and width of effective blobs of the seed image.

The method of detecting individual teeth from a tooth image according to an embodiment of the present invention may further include generating an extended seed image by performing a morphological expansion operation for each of a plurality of blobs included in the seed image.

Through this step, there is no empty space between individual teeth, and it is possible to clarify the distinction between individual teeth. Meanwhile, the nearest individual tooth index may be allocated to an area where the overlapping between individual teeth occurs during the morphological expansion operation process.

Next, an individual tooth is detected from the tooth image using the seed image (S50).

Specifically, an individual tooth may be detected by assigning an index to each individual tooth included in the seed image, analyzing a connection component for each of a plurality of blobs and assigning an index to a plurality of slices included in the tooth image. Here, an index may be assigned according to an overlap ratio between a blob included in a slice and an individual tooth included in the seed image.

As described above, in the method of detecting individual teeth in a tooth image according to an embodiment of the present invention, a seed image that is completely separated between individual teeth is generated, and an individual tooth can be detected from the tooth image based on the seed image. As a result, there is an effect of improving the accuracy of individual tooth detection.

The embodiments of the present invention disclosed in the present specification and drawings are only provided for specific examples to easily explain the technical content of the present invention and to aid understanding of the present invention, and are not intended to limit the scope of the present invention. Therefore, the scope of the present invention should be construed that all changes or modified forms derived based on the technical idea of the present invention in addition to the embodiments disclosed herein are included in the scope of the present invention.

110: communication department
120: input
130: display
140: memory
150: control unit
151: image correction unit
152: average image generator
153: seed image generation unit
154: individual tooth detection unit

Claims (13)

Generating an average image by averaging the plurality of slices in a Z-axis direction in a tooth image including a plurality of slices;
Generating a seed image separated between individual teeth by binarizing the average image; And
Detecting individual teeth in the tooth image using the seed image
Method for detecting individual teeth in a tooth image comprising a.
The method of claim 1,
Generating an expanded seed image by performing a morphological expansion operation for each of a plurality of blobs included in the seed image
A method of detecting individual teeth in a tooth image further comprising a.
The method of claim 1,
Generating the average image
The step of generating the average image by correcting the tooth inclination from the tooth image
How to detect individual teeth in tooth image.
The method of claim 1,
Generating the average image
Generating the average image by removing noise blobs from the tooth image
How to detect individual teeth in tooth image.
The method of claim 1,
Generating the average image
Compensating a tooth inclination from the tooth image and removing noise blobs from the tooth image to generate the average image
How to detect individual teeth in tooth image.
The method of claim 1,
Generating the average image
A step of generating a plurality of the average images by averaging the tooth images under different conditions
How to detect individual teeth in tooth image.
The method of claim 6,
The step of generating the seed image
The step of generating the seed image by selecting any one of the plurality of average images based on the number and area of the effective blobs of the average image
How to detect individual teeth in tooth image.
The method of claim 6,
The step of generating the seed image
Generating a plurality of candidate seed images by binarizing the plurality of average images; And
And selecting the seed image from the plurality of candidate seed images based on the number and width of effective blobs of the candidate seed image.
How to detect individual teeth in tooth image.
The method of claim 1,
The step of generating the seed image
And setting a reference value for binarizing the average image.
How to detect individual teeth in tooth image.
The method of claim 1,
The step of setting the reference value
The step of setting the reference value based on the number and width of effective blobs of the seed image
How to detect individual teeth in tooth image.
The method of claim 1,
The step of detecting the individual teeth
Allocating an index for each of the individual teeth included in the seed image, analyzing a connection component for each of a plurality of blobs for the plurality of slices included in the tooth image, and giving the index
How to detect individual teeth in tooth image.
The method of claim 11,
The step of assigning the index for each of the plurality of blobs
The step of assigning the index according to the overlap ratio of the blob included in the slice and the individual teeth included in the seed image
How to detect individual teeth in tooth image.
An average image generator configured to generate an average image by averaging the plurality of slices in a Z-axis direction from a tooth image composed of a plurality of slices;
A seed image generator configured to generate a seed image separated between individual teeth by binarizing the average image; And
Individual tooth detection unit for detecting individual teeth in the tooth image using the seed image
Individual tooth detection device in a tooth image comprising a.

Images