JP4956815B2 - Dental X-ray equipment - Google Patents

Description

The present invention relates to a dental X-ray imaging apparatus capable of automatically setting X-ray imaging conditions of a subject by image recognition.

Conventionally, for example, when performing panoramic X-ray imaging or cephalo X-ray imaging for continuous imaging along a dental arch, the X-ray imaging conditions include the thickness of the head of the subject and a fixture that tightens both sides of the head And the volume resistance, and the tube voltage and / or tube current is determined according to the head thickness. “See Patent Document 1”.
At this time, the wider the fixture, the stronger the X-rays (a larger X-ray dose) are applied to the subject's head. Accordingly, the fixture is attached to the subject's head. If the user forgets to close the camera, loosely tightens it, or does not adjust it properly, the person to be imaged will receive an X-ray dose more than necessary.

JP-A 62-98600

  As described above, in the conventional head thickness measurement, the optimum X-ray imaging conditions may not be selected for the subject, such as when the head fixing tool is forgotten to be closed. That is, in the conventional measurement using a head fixture (head support), the head thickness is measured by the head fixture and reflected in the X-ray irradiation dose. However, there is a certain proportional relationship between the head thickness and the optimal dose of the subject, but this is not perfect.

  Therefore, by constructing a database for the relationship between the shape of the subject's head and jaws and the optimum dose using image recognition means, not only the thickness of the head but also the examination by image It was confirmed that a more optimal dose of X-rays can be obtained by patterning the person's jaw shape and performing pattern matching.

  Therefore, even if the head fixture (head support) is forgotten to be closed, X-ray imaging conditions suitable for the subject can be selected.

The term “optimum dose” as used herein means an X-ray irradiation dose that can obtain a better image contrast with a small exposure.
“Head thickness” means the width of the head at the position where the head support hits (slightly behind the temple), that is, the width of the part facing the front of the camera and slightly above both ears, This width shall be measured.

As described above, the head thickness can also be measured by image recognition using a camera.
According to this image recognition, in determining the hair portion of the subject, the recognition of the head thickness differs depending on whether the hair is large or small. At this time, the face position can be limited because the head thickness is grasped by image recognition by the camera after the subject is positioned. Therefore, it is possible to measure the head thickness of the subject by separating the hair and the face by sampling the hue (hue), saturation, and lightness at a certain position and determining the threshold value. At this time, the person to be photographed needs processing such as preventing the hair from being applied to the face part such as putting the hair on the ear.

In judging the subject's face by comparing it with the background color,
When judging a face only by color ... The background needs to be limited to colors other than red.
When using the differential image method to determine the face ... If you use a method that takes the difference between the image t time ago and the current image and extracts the part that moved, you can shoot without limiting the background. The person's face part can be specified.

It is an object of the present invention to provide an apparatus for automatically selecting an optimum X-ray imaging condition for a subject by measuring head thickness by image recognition.
Another object of the present invention is to provide an apparatus for obtaining more optimal X-ray imaging conditions based not only on the thickness of the head but also on the depth and shape of the jaw.
Furthermore, an object of the present invention is to provide an apparatus for automatically selecting an optimum X-ray imaging condition for a subject even if he forgets to close the head fixture.

The dental X-ray imaging apparatus according to the present invention includes one or a plurality of cameras installed in or near the apparatus main body, an arithmetic device that processes and calculates a video signal from the camera, and an output signal from the arithmetic device. In a dental X-ray imaging apparatus provided with a control unit for controlling X-ray imaging conditions, by processing the video signal of the subject's head thickness and jaw shape output from the camera by the arithmetic unit The thickness of the head, the width of the jaw, and the shape are grasped, thereby determining optimum tube voltage and tube current X-ray imaging conditions for jaw imaging, and based on the X-ray imaging conditions in the control unit It is characterized by controlling tube voltage and tube current.
Further, according to the present invention, the control unit in the dental X-ray imaging apparatus patterns the shape of the head fixture in advance, and acquires the width of the head fixture from the camera when recognizing the width of the head fixture. The position of the head fixture in the image is scanned with the patterned pattern of the head fixture, and the position where the head fixture pattern most closely matches is determined as the position of the head fixture. The present invention is characterized in that the width of the part fixing tool can be detected.

表１から明らかなように、頭部厚が厚くなればなるほどX線の管電圧が高くなり、管電圧が高くなればなるほど、透過力の強いX線が発生する。
即ち、頭部固定具による頭部厚検知だけでは頭部固定具（ヘッドサポート）を閉じ忘れた場合、より多くのX線量を被撮影者に照射し得るようになるが、本発明によれば、被撮影者の頭部、顎部形状や奥行き等と最適線量との関係のデータベース（辞書画像）を構築しておくことにより、頭部の厚さだけでなく、画像認識による被撮影者の顎部形状をパターン化し、パターンマッチングすることによって、被撮影者に対して、より最適なX線量を求めることができる。
従って、頭部固定具（ヘッドサポート）を閉め忘れても、被撮影者に対して最適のＸ線撮影条件を選択することができる。 According to the present invention, when the head fixing tool is forgotten to be adjusted, the control unit of the dental X-ray imaging apparatus according to the present invention controls the optimal X for the head thickness that is the test site of the subject. The radiographic conditions can be automatically selected. That is, even if the head fixture is forgotten to be adjusted, the subject is irradiated with strong X-rays (a large amount of X-ray) as shown in Table 1 (relationship with head thickness-X-ray imaging conditions). Can be prevented.

As is apparent from Table 1, the tube voltage of X-rays increases as the head thickness increases, and X-rays with higher transmission power are generated as the tube voltage increases.
That is, if the head fixing device (head support) is forgotten to be closed only by detecting the head thickness by the head fixing device, more X-ray dose can be irradiated to the subject. By constructing a database (dictionary image) of the relationship between the head, jaw shape, depth, etc. of the subject and the optimum dose, not only the thickness of the head but also the subject's By patterning the jaw shape and pattern matching, a more optimal X-ray dose can be obtained for the subject.
Therefore, even if the head fixture (head support) is forgotten to be closed, the optimum X-ray imaging conditions for the subject can be selected.

Image recognition The procedure for introducing a subject to the X-ray imaging apparatus shown in FIG. 1 in order to perform panoramic X-ray imaging is shown below.

  As shown in FIG. 1, the X-ray imaging apparatus is provided with a slide body 10 on a support 1, a chin rest (chin rest) 9 at a lower part thereof, and a drive unit 7 that is movable up and down at an upper part thereof. 7 is provided with a rotatable rotating arm 2, and an X-ray head 5 including an X-ray imaging unit and an X-ray image receiving unit is fixed to the rotating arm 2, and the X-ray imaging unit and X of the X-ray head 5 are fixed. The head of the person to be imaged 11 is interposed between X-ray imaging units during X-ray imaging. Further, the head support 6 is provided so as to be rotatable in the horizontal direction with respect to the chin rest 9. By attaching the slide body 10 slidably up and down with respect to the column 1, the X-ray imaging unit and X-ray image receiving unit, the chin rest 9 and the head support 6 of the X-ray head 5 can be moved up and down.

First, the person to be imaged 11 instructs to stand at the X-ray imaging position by slightly pulling his chin and facing the front.
Next, a chin rest (chin rest) 9 provided at the center of the X-ray imaging apparatus is substantially matched to the height of the chin of the subject 11.
As shown in FIG. 2 and FIG. 3, the person to be imaged 11 is introduced into the imaging position in the X-ray imaging apparatus, the back is stretched, the jaw is placed on the marking portion 12 on the chin rest base 8 of the chin rest 9, and marking is performed. The jaw is brought into contact with the tip 13 of the part, and the bite block 14 is held. In this state, the head support 6 is brought into contact with the temporal region of the person 11 to be photographed from both sides, and accordingly, as shown in FIG. 4, the face 11's face is prevented from being greatly displaced in the X and Y directions. To get.

Next, a positioning beam switch (not shown) is pressed, and the median sagittal line of the subject 11 is aligned with the bite block 14 (located at the center of the median beam).
Since the positioning beam represents the tomographic area of the anterior tooth portion, fine adjustment is performed by moving the chin rest 9 back and forth so that the anterior tooth root apex (center of the lower jaw 3 as a guide) is aligned with the tomographic area.

Next, the eye-ear line is adjusted to determine whether or not the eye-ear line of the subject 11 is horizontal. If the eye-ear line is not horizontal, the chin rest 9 is moved slightly up and down to become horizontal. Make fine adjustments.
Further, as shown in FIG. 3, the head thickness is measured by adjusting the head support 6 so as to be pressed against the positions A of the head supports on both sides of the subject 11.

  At this time, the photographed person 11 is caused to grip the chin rest handle 15 in such a state so that the photographed person can be stably held during photographing.

Here, in order to set the X-ray imaging conditions suitable for the head thickness of the subject, that is, the tube voltage, tube current, etc. of the X-ray tube, the camera 20 installed in the panoramic X-ray imaging apparatus body is used. The head of the subject is photographed from the front.
The installation position of the camera 20 is arranged so that the camera is placed in front of the face of the subject when the subject is positioned in the X-ray imaging apparatus.
The installation location of the camera 20 is not limited to the present embodiment. Similarly, the number of cameras is not limited.

  The video signal of the camera is processed as described later by the arithmetic unit of the X-ray imaging condition automatic setting device of the present invention.

  By the processing of the arithmetic unit, the thickness of the head of the subject, the width of the jaw, and the shape can be grasped, and thereby the optimum X-ray imaging conditions can be determined.

  As shown in FIG. 5, one or more cameras 20 are installed in the apparatus main body so that the head of the subject 11 can be photographed from the front. As shown in FIG. 6, the video signal of the camera is processed in the arithmetic unit of the automatic X-ray imaging condition setting apparatus of the present invention. This signal processing is performed according to the following procedure.

  In other words, the X-ray imaging condition automatic setting device of the present invention includes a computing device 21 and a control unit 22, and the computing device 21 includes a video decoder unit 23 and a computing unit 24 connected thereto. An NTSC output from the camera 20 is supplied to the video decoder unit 23. Further, as indicated by a dotted line, when the output from the camera 20 is a digital output, the output is directly supplied to the arithmetic unit 24.

  When the output signal 25 from the camera 20 is an analog signal (NTSC or the like), it is converted into a digital signal such as 24-bit RGB by the video decoder unit 23 in the arithmetic unit 21. When the output signal from the camera 20 is a digital signal 26, the digital signal 26 is directly supplied to the arithmetic unit 24, not to the video decoder unit 23.

24bit HSV colorimetric system data of RGB data in the calculating unit 24 of converting into (hue (H ue), the saturation (S aturation), lightness (V alue)).
The outline of the conversion method to the HSV color system will be described below.

That is, conversion from the RGB color system to the HSV color system is performed as follows.
If we define V = max {R, G, B},
When V = 0
S = 0, H = undefined.
When V ≠ 0 and defining v = min {R, G, B},
S = (Vv) / V
When R = V, H = (GB) / (Vv)
H = 2 + (BR) / (Vv) when G = V
When B = V, H = 4 + (RG) / (Vv).
At this time, H = H / 6.
However, H <0 H = H + 1.
(Shuji Nakaji, “A Study on Tracking of Part Images in Moving Images”, [online], April 23, 1999, [Search February 25, 2004], Internet, <url: http: // (See www.ics.kagoshima-u.ac.jp/~kimi/prmu98-1/tsld001.htm>)

（「平野，中村：“顔の回転にロバストな個人識別方式”, 電気学会論文誌C, Vol.120-C, No.6, pp.800-808(2000-6).」参照） The conversion formula of the modified HSV color system data converted from the RGB color system data according to the above definition is expressed by Equation 1. “See Non-Patent Document 2.”

(See "Hirano, Nakamura:" Robust personal identification method for face rotation ", IEEJ Transactions C, Vol.120-C, No.6, pp.800-808 (2000-6).))

FIG. 7 shows an image of hue H (hue) data of the converted modified HSV color system data. (In the embodiment, a strong reddish color is displayed as a high value (white) and a weakly reddish color is displayed as a low value (black).)
Since the subject's face is more reddish than the background, the width of the face can be obtained by obtaining a horizontal histogram of the image and extracting a portion having a higher value than the background.

  If the background cannot be limited, a histogram can be obtained in a state where a known background such as an image receiver (X-ray image receiving unit) overlaps the subject 11. For example, when the horizontal center of the image coincides with the horizontal center of the face of the subject 11, the image receiver (X-ray image receiving unit) overlaps the right side of the subject 11 in the image of FIG. 7. It is also possible to obtain the face width of the virtual subject 11 by obtaining the width from the center of the image to the right side from the histogram and obtaining a double value thereof.

In addition to the hue data, separation from the background can also be performed using saturation data and lightness data. In addition, the data can be separated from the background from edge information by applying a sharpening filter or analyzing frequency characteristics by Fourier transform or wavelet transform. The face of the subject can also be separated from the background by obtaining a difference image from the image (background image) when the subject is absent.
If the background can be limited, the X-ray head 5 is rotated and scanning is performed so that the subject and the image receiving unit do not overlap, thereby reducing the face width of the subject 11 without taking the above-described means. Can be sought.
If it is difficult to separate the subject's face and background, the background can be made known by rotating the rotary arm 2 to the position shown in FIG. Separation can be facilitated.

  The control unit 22 converts information such as the head thickness of the subject obtained by processing in the calculation unit 24 into X-ray imaging conditions according to a table created in advance, and the converted signal Is supplied to the X-ray irradiation unit 27, the drive circuit 28, and the display unit 29, and the X-ray imaging condition (optimum X-ray dose) is determined by checking with the display unit 29. The X-ray irradiation can be performed from the X-ray irradiation unit 27.

Hereinafter, an example of means for determining the optimum X-ray imaging conditions from the thickness of the head, the width of the jaw, and the shape based on information obtained by image recognition will be described below.
First, pattern matching using the subspace method will be described. First, eigenvectors for each group are obtained. With reference to the flowchart in FIG. 8, eigenvectors are obtained by the following procedure when developing software for determining optimum X-ray imaging conditions from the thickness of the head, the width of the jaw, and the shape. In the following description, an example in which the optimum X-ray imaging conditions are determined from the width and shape of the jaw will be described. However, when the optimum X-ray imaging conditions are determined from the thickness of the head, the following means can be applied. .

ここに、Nは辞書画像の次元数（ピクセル数）、Kは辞書画像数、Cはグループ数である。 First, in step S1, the positioning of the subject is completed. Next, image data is acquired in step S2, and RGB-HSV image conversion is performed on the image data as described above in step S3. In step S4, the image is cut out, and the jaw position of the subject is cut out from the image. Next, in step S5, the extracted information on the jaw position of the subject is grouped. That is, a group is formed for each optimum dose and classified into the dictionary images shown in Table 2.

Here, N is the number of dimensions (number of pixels) of the dictionary image, K is the number of dictionary images, and C is the number of groups.

この行列Sから次式、数３を得ることができる。

（情報統計学II 第5回 主成分分析（１）２変量の主成分分析、
http://kuva.mis.hiroshima-u.ac.jp/~asano/Kougi/99s/JouhouToukei2/5-19.html-12k-2004年8月7参照） Next, referring to FIG. 9, in step S6, a variance-covariance matrix S expressed by the following equation 2 is created. In this case, a variance covariance matrix is obtained for each group.

From this matrix S, the following equation (3) can be obtained.

(Information Statistics II 5th Principal Component Analysis (1) Bivariate Principal Component Analysis,
http://kuva.mis.hiroshima-u.ac.jp/~asano/Kougi/99s/JouhouToukei2/5-19.html-12k-See August 7, 2004)

ここに、jは主成分により圧縮された次元数である。
（SII’99、第5回画像センシングシンポジウム、チュートリアル講演会、基礎コース、テキスト、平成11年6月16日(水)、パシフィコ、主催：画像センシング技術研究会、参照）。 Next, principal component analysis is performed in step S7 to obtain the following equation (4). An eigenvalue λ _j and an eigenvector U _j are obtained for each group by analyzing the principal components.

Here, j is the number of dimensions compressed by the main component.
(See SII'99, 5th Image Sensing Symposium, Tutorial Lecture, Basic Course, Text, June 16, 1999 (Wednesday), Pacifico, Organizer: Image Sensing Technology Study Group).

被撮影者Aに最適な最適X線の線量は既知であるので、最適線量が最も近くなるグループ１（表３）に、画像１_１を分類する。

The actual group classification will be described below with reference to FIGS. 11 (a) and 11 (b) and FIGS. 12 (a) and 12 (b).
The means for creating a group for each optimal dose of X-ray and classifying the dictionary image is as follows.
(1) Obtain image data of the subject A whose optimum dose is known in advance (FIG. 11 (a)).
(2) Since the position of the chin rest is known, the jaw position of the subject A can be determined from the acquired image data.
(3) From the acquired image data, the jaw position of the subject A is picked up as shown in FIG.
For example, the jaw position is picked up with a size of 12 × 15 pixels.
(4) the image data of the photographer A and the image 1 _1, the number of pixels constituting the image 1 ₁ becomes 12 × 15 = 180 pixels.
Considering where the image 1 ₁ as a vector, image 1 _1: X ₁ vector variables X _11, X _12, equation a 180-dimensional vector consisting of · · · · · X _1N, as Equation 5 I can express.

Since the optimum X-ray dose optimal for the subject A is known, the image ₁₁ is classified into group 1 (Table 3) where the optimum dose is closest.

(5) Further, the image data of the subject B with the optimum X-ray dose known in advance is acquired (FIG. 12 (a)).
(6) As in the case of (3) above, the jaw position of the subject B is picked up as shown in FIG. 12 (b).
(7) to image 2 ₁ and the image data of the photographer B. This image 2 ₁ is also represented as X ₂ vector composed of the same manner as described above 12 × 15 = 180 dimensions. Since the optimal dose for the subject B is also known, the image 2 ₁ (see Equation 6) is classified into the group with the closest optimal X-ray dose.

When the subject B is close to the optimum X-ray dose of the subject A, the X ₂ vector is classified into group 1 shown in Table 4.

(8) The operations (1) to (7) are performed for each optimum dose, and grouping is performed as shown in Table 5.

Next, group specifying processing in the calculation unit will be described with reference to FIG.
As shown below, pattern matching is performed on the position image of the jaw of the subject to be cut out to determine which group has a high correlation, and the optimum X-ray dose is determined.
The calculation is performed using the eigenvector Uj for each group in the group classification obtained as described above.
As described above, first, the positioning of the subject is completed in step S1. Next, image data is acquired in step S2, and RGB-HSV image conversion is performed on the image data as described above in step S3. In step S4, an image is cut out, and the jaw position of the subject is cut out from the image.
The subject's jaw position is limited to the chin rest. At this time, the relative position of the camera and the chin rest is known. Therefore, the position of the face in the image can be specified.

最後に、ステップS9においてグループの特定を完了し、選択されたグループによる最適X線の最適値を制御部に送り、最適X線の最適線量制御指令をX線照射部２７に供給して最適X線線量を被撮影者に照射する。 The clipped image is a vector X with the number of images as the number of dimensions.
Next, pattern matching processing is performed in step S8. For each group, the matching value obtained by the following expression, Equation 7, is obtained, and it is determined which group (1, 2,... C) is highly correlated, and the optimum group is selected.

Finally, the identification of the group is completed in step S9, the optimum value of the optimum X-ray by the selected group is sent to the control unit, and the optimum dose control command of the optimum X-ray is supplied to the X-ray irradiation unit 27 to obtain the optimum X Irradiate the patient with a radiation dose.

In the above embodiment, the group classification for each optimum dose is described. However, it is also possible to create a group classification such as a dental dentition size and a dentition tip shape in addition to the creation of the group classification for each optimum dose.
From the group classification for each optimum dose, it is possible to further classify by the dentition size, the dentition tip shape, and the like.

  In this example, the shape of the head fixture is previously patterned in the X-ray imaging apparatus. This patterning is performed by the arithmetic unit, and when recognizing the head fixture width, the position of the head fixture in the image acquired from the camera is scanned with the pattern patterned as described above, and the head fixture The position where the pattern further matches is determined as the position of the head fixture so that the width of the head fixture can be detected. As described above, the optimal X-ray dose is calculated by the control unit with respect to the width of the head fixture detected in this way, and thus the X-ray imaging conditions based on the width of the head fixture are thus calculated. Can be selected.

  The present invention can be applied to a dental X-ray imaging apparatus to obtain a great effect.

FIG. 1 is a sectional side view showing a configuration of an X-ray imaging apparatus according to the present invention, FIG. 2 is a plan view showing the configuration of the chin rest of the X-ray imaging apparatus, FIG. 3 is a side view showing a state where the head is fixed to the chin rest, FIG. 4 is a front view for explaining the head positioning state, FIG. 5 is an explanatory diagram for explaining a shooting state by the camera of the subject. FIG. 6 is a block diagram showing the configuration of the automatic setting device of the present invention. FIG. 7 is an explanatory diagram showing the outline of the head of the subject who displays the hue (hue) data of the HSV color system data as an image. FIG. 8 is a flowchart showing a procedure for obtaining eigenvectors for each group, FIG. 9 is a flowchart showing a procedure for obtaining eigenvectors for each group, FIG. 10 is a flowchart showing the group specifying process in the calculation unit, FIGS. 11 (a) and 11 (b) are explanatory views showing means for picking up the subject's jaw part from the image data and obtaining its pixel size in the means for classifying the dictionary images, FIGS. 12 (a) and 12 (b) are explanatory views showing a means for picking up the subject's jaw part from the image data and obtaining the pixel size in the means for classifying dictionary images.

Explanation of symbols

1 support 2 rotating arm 5 X-ray head 6 head support 7 drive unit 8 chin rest base 9 chin rest
11 Subject
12 Marking part
13 Tip
14 byte block
15 Chinrest handle
20 Camera
21 Arithmetic unit
22 Control unit
23 Video decoder section
24 Calculation unit
25 Output signal
26 Digital signal
27 X-ray irradiation unit
28 Drive circuit
29 Display section A Head support position

Claims (4)

  One or a plurality of cameras installed in the apparatus main body or in the vicinity thereof, an arithmetic device for processing and calculating a video signal from the camera, and a control unit for controlling X-ray imaging conditions by an output signal from the arithmetic device; In a dental X-ray imaging apparatus comprising: a head thickness, a chin width, and a shape of a head by processing a video signal of the head thickness and chin shape of the subject output from the camera by the arithmetic unit To determine the optimum tube voltage and tube current X-ray imaging conditions for jaw imaging, and to control the tube voltage and tube current based on the X-ray imaging conditions in the control unit. A featured dental X-ray imaging apparatus.   The control unit of the dental X-ray imaging apparatus includes: an X-ray irradiation unit connected to the control unit; a drive circuit connected to the control unit to activate the X-ray irradiation unit and perform X-ray irradiation; The dental X-ray imaging apparatus according to claim 1, further comprising a display unit connected to the control unit and configured to confirm imaging conditions during X-ray imaging.   The control unit of the dental X-ray imaging apparatus patterns the shape of the head fixture in advance, and recognizes the width of the head fixture, the head fixture in the image acquired from the camera The position of the head fixing tool is scanned with the pattern of the shape of the head fixing tool, and the position where the head fixing tool pattern most closely matches is determined as the position of the head fixing tool, and the width of the head fixing tool is detected. The dental X-ray imaging apparatus according to claim 1, wherein the dental X-ray imaging apparatus is configured to be able to do so.   The control unit of the dental X-ray imaging apparatus is configured such that the calculation unit determines the position where the pattern of the head fixture shape most closely matches, and the calculation of the optimal X-ray dose is performed by the control unit. The dental X-ray imaging apparatus according to claim 3.

Images