JP2758570B2 - Method for detecting rotation center coordinates of subject in X-ray CT apparatus - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention makes it possible to use an X-ray TV apparatus (apparatus for displaying an X-ray fluoroscopic image) and to perform an inspection using a tomographic image (X-ray CT image) as well as an inspection using a fluoroscopic image of a subject. The present invention relates to a method of detecting the rotation center coordinates of a subject in a simple X-ray CT apparatus.

[0002]

2. Description of the Related Art Conventionally, an X-ray TV apparatus for observing a fluoroscopic image is generally used for non-destructive inspection of the inside of an object such as an electronic component because real-time inspection is possible and the operation is simple. ing.

However, in the case of a subject having a complicated internal structure or a multi-layer structure, the above-mentioned perspective image (two-dimensional projection of a three-dimensional component) is inspected because a portion to be focused on overlaps with another component. Becomes difficult. For this reason, an inspection using an X-ray CT apparatus capable of nondestructively observing a tomographic image of a subject is required.

However, in the conventional X-ray CT apparatus, the X-ray generator, the X-ray detector, and the rotational axis of the subject are accurately positioned with the accuracy required for tomography, and the rotational motion of the subject does not occur. The rotating device is likely to be large-scale and expensive. That is, the conventional X
Since the X-ray TV apparatus is dedicated to a fluoroscopic image, the position of the rotation center axis of the subject does not always coincide with the center position of the X-ray detector, and the positioning accuracy of the subject mounting table is insufficient for tomography. In addition, the rotating device is configured to easily cause rotational blur due to the size, shape, and gripping mechanism of the subject. In particular, positioning of the rotation axis of the subject is important.
Therefore, in order to perform tomography using a conventional X-ray TV apparatus, it is necessary to replace the subject mounting table, the gripping mechanism, and the rotating device with those dedicated to CT, which is inconvenient in that the operation is inconvenient. Was seen.

An X-ray CT apparatus has been proposed which detects the center axis of rotation of the subject from the contour position of the fluoroscopic image and enables tomography without performing advanced positioning of the subject. (JP-A-06-118030). According to this proposal, for the X coordinate of the rotation center axis, the rotation angle when the X coordinate of the subject reaches the minimum value is set to θn, and the rotation angle when the X coordinate reaches the X coordinate maximum value is set to θm. Xns, Xne, X are the minimum and maximum X coordinate values of the data.
ms, Xme, the true rotation center axis is ((Xns +
Xme) / 2).

However, many electronic components are molded with a resin having a relatively low X-ray absorption coefficient.
It is difficult to determine the contour of the subject from the projected image. Therefore, it is difficult to detect the rotation center axis of the subject by the proposed system.

[0007]

SUMMARY OF THE INVENTION An object of the present invention is to provide an inspection system for both a fluoroscopic image and a CT image by an X-ray TV apparatus, in which a subject is molded with a resin having a relatively low X-ray absorption coefficient. An object of the present invention is to provide a method capable of detecting a rotation center axis even in the case of an electronic component or the like.

[0008]

According to the present invention, in order to achieve the above object, an X-ray generator and an X-ray detector are arranged opposite to each other with a subject interposed therebetween. Image processing means for the X-ray image transmitted through the sample, means for detecting the contour position of the subject in each direction from the density information of the X-ray image obtained by the image processing means, and rotation center of the subject from the contour position In an X-ray CT apparatus having means for detecting an axis,
X outside the subject so that it rotates together with the subject
By installing a material that absorbs X-rays, detecting the true rotation center axis of the subject from the contour projection image of the X-ray fluoroscopic image of the material, detecting the rotation center coordinates of the subject, enabling tomography It is in.

The main point of the present invention is that when an X-ray CT apparatus that detects the rotational center axis of a subject from the contour position of a fluoroscopic image reconstructs a CT image, the X-ray absorption around the subject is weak. Thus, the true rotation center axis can be easily detected even for a subject whose outline of the fluoroscopic image is unclear.

Hereinafter, in the present invention, a material placed outside the subject to detect the true rotation center axis is referred to as a “mark”. The mark of the present invention has two installation methods. One method is to set a CT image on the section to be constructed. The other is a method of setting marks above and below a cross section where a CT image is to be constructed. Hereinafter, the former will be referred to as Installation Method 1 and the latter will be referred to as Installation Method 2.

First, the installation method 1 will be described. Installation method 1
In this case, the detection section of the rotation center axis and the section constituting the CT image are the same. Therefore, a mark is drawn in the reconstructed CT image. In this case, the appropriate amount of X-ray absorption of the mark material is
It is determined relatively to the X-ray absorption amount of the subject itself. Specifically, it is preferable that the gradation of the deepest part including the mark of the fluoroscopic image obtained by the image processing means is just saturated. In other words, if the amount of X-rays absorbed by the mark is too small, the detection of the position of the mark becomes inaccurate, and the rotation center axis may not be accurately detected. On the other hand, if the absorption amount of X-rays is too large, there is a possibility that the X-ray fluoroscopic information on the subject existing in the shadow of the mark will not be used for the reconstruction of the CT image.

The material and shape of the mark material are closely related to the amount of X-ray absorption. Therefore, it is important to take that point into account. The material and shape appropriate for the mark are X
The determination can be made generally from the density information of the perspective image when the line is irradiated. Specifically, as described above, it is preferable that the tone of the deepest part including the mark of the fluoroscopic image is just saturated, and the tone of the darkest part of the subject before the mark is set is saturated (100%). Of 5
It is preferably in the range of 0 to 95%. As a material of the mark, generally, a material having a relatively high X-ray absorption coefficient is preferable. Specifically, metals such as aluminum, copper, iron, nickel, and silver, or materials containing such metal elements are preferable.

The appropriate amount of X-ray output is also relatively determined by the X-ray absorption of the mark and the subject. It is preferable that the appropriate amount is such that the gradation of the darkest part including the mark of the fluoroscopic image obtained by the image processing means is just saturated as described above. That is, when the output of the X-ray is too large, the perspective image of the mark becomes unclear, and the rotation center axis may not be detected accurately. On the other hand, when the output of the X-ray is too small, there is a possibility that the X-ray fluoroscopic information on the subject existing in the shadow of the mark is not utilized for the reconstruction of the CT image. In addition, as a method of setting a mark belonging to the setting method 1, it is possible to cover the entire outer periphery of the subject with the mark. In this case, there is an advantage that the contour of the subject in the CT image after the reconstruction becomes clear.

Next, the case of the installation method 2 will be described.
In the case of the setting method 2, the rotation center axis detection section where the mark exists is different from the CT image configuration section, and there are two rotation center axis detection sections. Specifically, the respective rotation center axes are detected from the data of the two cross-sectional perspective images with marks, the average coordinates of the rotation center axes are calculated as the rotation center axis of the CT section to be constructed, and then the CT image is constructed. The perspective image data of the section to be obtained is taken in, and a CT image of the target section is constructed. It is preferable that the X-ray absorption of the mark material be as large as possible, but any material may be used.
In other words, the rotation center axis positioning by detecting the mark position and the data acquisition of the section to be reconstructed are completely independent, and the X-ray intensity can be freely set in each case. For example, when silver particles are used as a mark for an object whose main component is an organic substance, strong X-rays are emitted when detecting the mark position, and X-rays are used when cross-sectional imaging of the object is performed. Can be made sufficiently weak. Of course, when the X-ray absorption amounts of the subject and the mark are almost the same, the detection of the mark position and the acquisition of the cross-sectional data can be performed simultaneously with the same X-ray intensity.

The advantage of the setting method 2 is that since no mark exists in the CT image reconstruction cross section, all projection data of the X-ray fluoroscopic image is information relating to the subject, and the information can be used 100% as reconstruction data. It is. Further, since the mark installation section and the CT image reconstruction section are independent, the material and shape of the mark can be freely selected as compared with the installation method 1. Accordingly, not only the above-mentioned materials, that is, aluminum, copper, iron, nickel, and silver, but also resin materials can be used as long as they have a different X-ray absorption coefficient from the mold resin of the subject.
However, when the detection of the rotation center axis and the collection of the projection data of the reconstructed cross section are performed at the same time, it is preferable to consider the same as the setting method 1. That is, it is preferable that both the grayscale of the mark and the darkest part of the fluoroscopic image of the subject be saturated.

The appropriate amount of X-ray output is also determined by the X-ray absorption of the mark and the subject. That is, when the contour detection of the mark and the projection data collection of the subject are independently performed, first, the output of the X-ray is adjusted so that the gradation of the deepest part of the fluoroscopic image related to the mark is just saturated, and X of subject
The output of the X-ray is adjusted to such an extent that the tone of the darkest part of the fluoroscopic image relating to the line absorption is just saturated. However, when the detection of the rotation center axis and the collection of the projection data of the reconstructed cross section are performed at the same time,
It is preferable that both the mark and the gradation of the darkest part of the fluoroscopic image of the subject are just saturated. If the output of the X-ray is not an appropriate value, the rotation center axis cannot be accurately detected, or the reconstructed CT image of the subject becomes unclear.

Regarding the setting method of the mark, the material may be adhered or applied to the subject in both the setting method 1 and the setting method 2. When it is desired to remove the material placed on the outer periphery after the CT inspection, it is preferable to take into account the adhesive strength or the solvent for cleaning the coating film in advance.

The rotation center axis X in the case of a normal X-ray TV device
c detection is obtained according to the following equation (1). Xc = ((Xns + Xme) / 2) (1) where Xc is the rotation center axis, Xns is the X coordinate minimum value of the projection data of the edge of the subject farthest from the true rotation center axis, Xme is the X coordinate maximum value of the projection data of the edge of the subject farthest from the true rotation center axis. However, in the case of the setting method 2, the average of the rotation center axes detected by the marks is set as the rotation center axis of the subject.

In the case of a micro-focus type X-ray TV device, the rotation center axis of the subject can be detected by performing calculations in consideration of the fact that the X-rays are emitted in a fan beam shape.

Although the contour position of the object differs depending on the setting of the object, by setting the projected images of the object and the mark to be always within the fluoroscopic image, the rotational movement of the central position of the object due to rotational shake can be detected. it can.

After detecting the rotation center axis as described above, the number S of samples is determined, and f (X0, θ) to f (X0 + S−).
(1, θ) is used in the reconstruction process as projection data. Next, the projection data of 500 to 600 pixels per line can be used as the maximum number of samples by the image processing apparatus, but the leading value f (X0, Θ) of the sample of the projection data used for the tomographic image reconstruction processing is the value of the subject. This can be performed by detecting the contour position after detecting the rotation center axis. Then, rotation blur can be detected.

In the present invention, the subject rotation control means may be newly installed in consideration of the shape, dimensions, etc. of the subject, using a rotating jig provided in a normal X-ray TV apparatus. Is also good. The control may use any method,
It is preferable to collectively manage the information including the concentration information of the subsequent sample by a personal computer.

In the present invention, the image processing means for the X-ray image transmitted through the subject is provided by an X-ray detector using an X-ray detector such as a vidicon camera or an image intensifier provided in an ordinary X-ray TV apparatus. 2 shows processing means for converting a perspective image into a two-dimensional video.

This X-ray two-dimensional fluoroscopic image is preferably integrated by the image processing means 5 to 100 times to compensate for the deterioration of the image quality due to the random noise of the X-ray detector and to improve the SN ratio. Then, the contour position of the subject is detected for each direction from the density information of the obtained fluoroscopic image.

[0025]

【Example】

Example 1 The case of the installation method 1 will be described. As shown in FIG. 1, the mark M is on the same cross section as the CT image reconstruction target cross section. Therefore, the density information of the mark M due to X-ray absorption and the density information of the target cross section overlap.

As shown in FIG. 2, an X-ray TV comprising an X-ray generator 2, an X-ray detector 3 for detecting a fluoroscopic image of a subject P, and the like.
An image input / processing device having a rotation control device 4 for controlling the rotation of a gripping mechanism of a subject P, an image display CRT 6 for executing a process of image input, and displaying a fluoroscopic image and a reconstructed image for CT. 5 and a personal computer 7 for executing device control, image reconstruction processing, and the like.

[0027] The personal computer includes:
The control unit includes a contour detection device 8 for the subject P, a rotation center axis detection device 9 for the subject P, a tomographic image reconstruction processing device 10, and the like. Also, the gripping mechanism of the subject utilizes an electronic component grip.

The rotation control device 4 generates a rotation output signal of the subject P by a rotation input signal from the personal computer 7 so that projection data of the subject P can be collected from all directions of 360 degrees. There is a step angle rotation mode. The continuous rotation mode is for setting the position of the subject P and setting the X-ray output. The rotation angle step can be set once or twice, and can be selected by a step angle setting switch. I do.

Next, the operation of each device during CT image observation will be described with reference to the flowchart shown in FIG. Since the optimum value of the intensity of the incident X-ray differs depending on the X-ray absorption amounts of the subject P and the mark M, the X-ray output is set to the optimum value from the see-through density information (step (1)). At this time, the density information is set to such an extent that the density information of the deepest part relating to the mark M is just saturated.

An input condition of the projection data is set. The input conditions include a rotation angle step required for the perspective image of the projection image, the number of projection directions, the number of integration of the perspective image, and the like (step (2)).

The rotation of the subject P is executed (steps (3) and (4)), and image quality deterioration of the fluoroscopic image due to random noise of the X-ray detector 3 is compensated by image integration of 10 to 100 times (step (3)). 5)), and store the projection data with the improved SN ratio (step (6)).

The projection data necessary for CT is obtained by converting a fluoroscopic image into an 8-bit digital value, and performing input and basic preprocessing by the image input / processing device 5.

Next, the minimum and maximum X coordinate values of the projection data relating to the X-ray absorption of the mark M are determined by using the object contour detection device 8.
And a normal X-ray T
In the case of a V apparatus, the rotation center coordinate Xc is obtained in accordance with the above equation (1), and in the case of a microfocus type X-ray TV apparatus, taking into account that X-rays are emitted in a fan beam shape (step (7) )). Thereafter, the number of samples S is determined (step (8)), and a range X0 to X0 + S-1 including all the projection data of the subject P is set (step (9)), and f
(X0, Θ) to f (X0 + S-1, に) are shifted to the reconstruction processing (step (10)). Note that f (X, Θ) represents the integrated value of the X-ray absorption coefficient of the subject, and Θ represents the projection angle.

The tomographic image reconstruction processing device 10 sets the number of reconstruction matrices, reconstructs a tomographic image by filter-corrected backprojection processing of projection data (step (10)), observes the image on the image display CRT 6, and Store (step (11)). The reconstruction processing of the object contour detection device 8, the rotation center axis detection device 9, the tomographic image reconstruction processing device 10, and the like is executed by software on a personal computer.

Embodiment 2 A case where the detection of the center of rotation and the reconstruction of the tomographic image in the case of the installation method 2 are performed independently will be described. As shown in FIG. 4, the mark M is installed at a position vertically and vertically spaced from the CT image reconstruction cross section. As the X-ray CT apparatus, an apparatus similar to that of the first embodiment is used. Next, a measurement procedure at the time of CT image observation will be described with reference to a flowchart shown in FIG.

In consideration of the X-ray absorption amount of the mark M, the X-ray output is set to an optimum value so as to be suitable for detecting the perspective contour image (step (1)). The rotation of the subject is performed, and the minimum and maximum coordinates of the contour position of each mark M are detected by using the subject contour detection device 8 (steps (2) to (5)).
5)). After the rotation is completed, the rotation center coordinates of the cross section including the respective marks M are detected from the detected minimum and maximum coordinates (step (6)), and the average rotation center coordinates are obtained as the rotation center coordinates of the subject (step (6)). Step (7)). Next, the X-ray output is set to an optimum value based on the fluoroscopic density information relating to the X-ray absorption of the subject (step (8)).
Next, the input conditions of the projection data are set in the same manner as in the first embodiment (step (9)). A series of subsequent processing from rotation of the subject P to image reconstruction processing is performed in the same manner as in the first embodiment (steps (14 to 17)).

Embodiment 3 A case will be described in which the installation method 2 simultaneously performs rotation center coordinate detection and projection data collection for tomographic image reconstruction processing.
The setting of the mark on the subject is the same as in FIG. X-ray CT
As the apparatus, the same apparatus as that of the first embodiment is used.

FIG. 6 shows a measurement procedure for observing a CT image.
This will be described with reference to the flowchart shown in FIG. First, the X-ray output is set to an optimum value so that the fluoroscopic images of both the mark M and the subject P have appropriate values (step (1)). Next, the input conditions of the projection data are set in the same manner as in the first embodiment (step (2)). The rotation of the subject is performed, and the detection of the minimum and maximum coordinates of the contour position of each mark M and the collection of projection data of the reconstructed cross section are simultaneously performed (steps (3 to 7)). The rotation center coordinates of the cross section including the respective marks M are detected from the minimum and maximum coordinates detected after the rotation is completed (step (8)), and then the average rotation center coordinates are obtained as the rotation center coordinates of the subject (step (8)). (9)). Next, the process proceeds to a reconstruction process using the collected projection data. Other series of operations and processes are described in the first embodiment.
(Steps (10-13)).

[0039]

The present invention utilizes a fluoroscopic image of an ordinary X-ray TV apparatus or microfocus type X-ray TV apparatus, detects the rotational center axis of the subject from the contour position of the fluoroscopic image, and converts the CT image into a CT image. In an X-ray CT apparatus to be reconstructed, the rotation center axis can be set even for an object such as an electronic component molded with a material having a low X-ray absorption coefficient, in which the detection of the contour position is difficult by the conventional method. Detection and reconstruction of the CT image was made possible.

[Brief description of the drawings]

FIG. 1 is a mark setting example based on a setting method 1.

FIG. 2 is a block diagram showing a configuration outline of the present invention.

FIG. 3 is a flowchart illustrating an operation based on an installation example 1.

FIG. 4 is an example of setting a mark based on a setting example 2;

FIG. 5 is a flowchart illustrating an operation based on a case where the detection of the rotation center axis and the collection of the subject projection data are independently performed in Installation Example 2.

FIG. 6 is a flowchart illustrating an operation based on a case where the detection of the rotation center axis and the collection of the subject projection data are simultaneously performed in Installation Example 2;

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 X-ray TV device 2 X-ray generator 3 X-ray detector 4 Rotation control device 5 Image input / processing device 6 Image display CRT 7 Personal computer 8 Subject outline detection device 9 Rotation center axis detection device 10 Tomographic image reconstruction processing apparatus

Continuation of the front page (56) References JP-A-6-118030 (JP, A) JP-A-62-75772 (JP, A) JP-A-53-134680 (JP, U) (58) Fields studied (Int .Cl. ⁶ , DB name) G01N 23/00-23/227 A61B 6/00-6/14 G03B 42/02

Claims (1)

(57) [Claims] An X-ray generator and an X-ray detector are arranged opposite to each other with a subject interposed therebetween, and the X-ray generator includes a subject rotation control unit, an X-ray image transmitting unit passing through the subject, and an image processing unit. In an X-ray CT apparatus having means for detecting a contour position of a subject in each direction from density information of an obtained X-ray image and means for detecting a rotation center axis of the subject from the contour position, A material that absorbs X-rays is installed so as to rotate integrally with the subject, and a true rotation center axis of the subject is detected from a contour projection image of an X-ray fluoroscopic image of the material. A method of detecting the rotation center coordinates of the subject.