JP2004081330A - Radiation imaging device and stereo radiograph - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce the number of correction images necessary for the stereo radiography. <P>SOLUTION: The apparatus has a means for calculating the gain correction factor from the white image taken from the front, a means for calculating the left shading modulus from the white image taken from the front and the white image taken from the left, a means for calculating the right shading modulus from the white image taken from the front and the white image taken from the right and a means which photographs an object under the almost the same conditions as those for the photographing the white image from the left and the white image from the right. The gain correction of the left object image is carried out by the gain correction factor while the shading correction thereof done by the left shading modulus and the gain correction of the right object image is carried out with the gain correction factor while the shading correction thereof done with the right shading modulus. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a radiographic imaging apparatus and a stereoscopic imaging apparatus, and more particularly to an X-ray digital imaging apparatus using a flat panel. In particular, the present invention relates to shading correction of a stereo photographing device.
[0002]
[Prior art]
In stereo shooting of an X-ray transmission image, it is necessary to secure an appropriate convergence width according to the shooting distance in order to obtain a stereo feeling. It is considered that humans should obtain a convergence angle of about 5 degrees for a distance of about 1 m in order to obtain an appropriate stereo feeling.
[0003]
On the other hand, with respect to oblique insertion during stereo imaging, shading of the X-ray tube and shading formed by the scattered ray grid occur. In order to obtain a suitable stereo feeling, it is necessary to appropriately remove the above-mentioned shading.
[0004]
[Problems to be solved by the invention]
A gain coefficient for performing the gain correction is obtained by taking a plurality of images (for example, eight or more) of images without a subject and improving the S / N (S / N ratio) and averaging them. .
Although it is possible to obtain a plurality of gain images depending on the shooting distance and the angle of oblique insertion, it is necessary to shoot a very large number of white images, which imposes a heavy burden on the operator. Specifically, assuming that there are three types of distances for performing stereo shooting, (3 types) × (two directions) × (average number of 8) = 48 white images are required.
[0005]
[Means for Solving the Problems]
The stereo imaging apparatus of the present invention, a white image from the front, a white image from the left, and a means for shooting a white image from the right, and a means for calculating a gain correction coefficient from the white image from the front, A means for calculating a left shading coefficient from the white image from the front and the white image from the left; a means for calculating a right shading coefficient from the white image from the front and the white image from the right; Means for photographing the subject under substantially the same conditions as the photographing of the white image from the right and the white image from the right, wherein the left subject image is gain-corrected with the gain correction coefficient, and the left shading coefficient is subjected to shading correction. The gain of the right subject image is corrected by the gain correction coefficient, and the shading correction is performed by the right shading coefficient.
[0006]
The radiation image capturing apparatus according to the present invention includes a gain correction unit and a shading correction unit, wherein the parameters of the gain correction unit are generated from a white image, and the parameters of the shading correction unit are generated from a low frequency component of the white image. It is characterized by doing.
[0007]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 shows a conceptual diagram of the present invention.
Generally, white correction is composed of gain correction and shading correction. The gain correction corrects the gain (sensitivity) of each channel of the semiconductor detector, whereas the shading correction performs the X-ray intensity distribution (generally, the X-ray in the central portion is strong and the peripheral portion has an arc shape). (Weak intensity). That is, the gain correction does not depend on the shooting distance or the like. Further, the shading correction varies depending on the distance and the incident angle, but there is no difficulty in diagnosis as long as the distance merely changes. Therefore, conventionally, white correction has been performed in which gain correction and shading correction are performed simultaneously even when the distance changes.
[0008]
However, when shading greatly changes in stereo shooting or the like, it is necessary to separately perform shading correction. In FIG. 1, an image A1, an image B2, and an image C3 are white images in which no subject is inserted, respectively. With respect to the image A1, an average white image is created by averaging a plurality of white images at the same position in order to increase the S / N, and this is set as a gain coefficient (also referred to as a "gain correction parameter") 4. I do. Further, since the image B2 and the image C3 do not affect the S / N ratio, one image is photographed, and after the gain correction, only the low frequency components of these images are extracted. The right image shading coefficient is 6.
[0009]
In FIG. 1, the main stereo shooting includes a left image D7 and a right image E8. The left image D7 and the right image E8 are subjected to gain correction with the same gain correction coefficient according to the block diagram shown in FIG. 1, and then are subjected to shading correction with separately prepared left and right image shading coefficients. The order of the gain correction and the shading correction is not limited, and either may be performed first.
[0010]
In the above description, the stereo imaging has been described. However, the present invention is not limited to this. In the present invention, when shooting in shading that is significantly different from normal (from the front) shading due to oblique insertion or the like, shading can be corrected with one white image without requiring a plurality of white images from that angle become.
[0011]
FIG. 2 is a block diagram illustrating a system configuration example according to the embodiment of the present invention. The white image is captured in a state where the subject 13 is not inserted between the X-ray generator 11 and the image detecting means 12. Operation instructions such as acquisition of a gain image, acquisition of a shading image, and photographing of a stereo image are performed through the user interface unit 14, and moving means of the X-ray tube (not shown) in accordance with the photographing of those images. And the other image processing means 15 and image display means 16 connected to the system bus BS1 are controlled by the system control means 17.
[0012]
The collection of the gain image and the calculation of the gain correction coefficient will be described.
The gain image is obtained by irradiating X-rays from an intermediate position or a standard position with respect to all expected imaging. Generally, an image is shot from the front and at a shooting distance of 180 cm. A plurality of (about eight) gain images are taken without the subject 13 and are transferred to the image processing means 15 one by one.
Further, offset data (data in which X-rays that are imaged almost at the same time as X-ray irradiation are not incident) are collected for each captured image (in this case, a gain image) and are overwritten and stored in the offset storage unit 31. .
[0013]
The gain image is corrected by the offset correction unit 20 based on the offset data. The offset correction is performed by subtracting the offset data from the X-ray incident data on the captured image.
Further, the offset-corrected image is averaged by the averaging means 21. The result of LOG conversion (not shown) of the averaged data is a gain correction coefficient. The obtained gain correction coefficient is stored in the gain coefficient storage means 32.
[0014]
Next, creation of a shading coefficient will be described.
The shading coefficient is photographed with no subject in the same manner as in the gain image, but the position of the X-ray is fixed according to the actual photographing position. In the case of shading for a left image in stereo imaging, the X-ray tube is set at a position moved to the left by a predetermined distance, and irradiation is performed.
[0015]
The same processing as the above-described calculation of the gain correction coefficient is performed on the shading image until the offset correction. After the offset-corrected data is converted by a LOG conversion circuit (not shown), the gain is corrected by the gain correction unit 23 based on the gain correction coefficient stored in the gain coefficient storage unit 32. Gain correction can be realized by subtraction of gain correction.
[0016]
The gain-corrected shading image is filtered by the filter unit 24. The filtering process can be realized by a rectangular average filter (low-pass filter) of N × N pixels. Here, a predetermined value is used for N, and generally a relatively large 100 to 200 pixels is used. The pixel pitch is 200 microns.
[0017]
The result of this filtering is a shading coefficient, which is stored in the corresponding shading coefficient storage means (left shading coefficient storage means 33, right shading coefficient storage means 34). Assuming that the shading coefficients are used by the shading correction means 26, it is conceivable that a fixed linear operation is applied to all the coefficients. Here, the primary operation is a result = input × first constant + second constant. The shading coefficients are calculated for the number of incident directions requiring shading correction, and are stored in respective storage locations.
[0018]
Lastly, data collection of data in the main shooting (stereo main shooting) will be described. Here, a stereo image composed of two left and right images is assumed.
When the subject 13 is inserted and the left image is photographed, the shading correction is performed by the shading correction unit 26 using the left shading coefficient after performing the offset correction and the gain correction. Shading correction can also be realized by subtraction.
[0019]
The image subjected to shading correction is subjected to QA (sharpening and gradation conversion) processing by QA processing means 28 and stored in a memory (not shown).
Similarly, shading correction of the right image is performed on the right image, and when the QA process is completed, a stereo image is created using the two left and right images. The method of creating a stereo image depends on the image display means 16, but in the case of a display method that separates left and right in the vertical direction, the left and right images are fitted into a comb for each line to be a stereo image.
[0020]
FIG. 3 is a simplified summary of the above-described gain correction, shading correction, actual shooting, and processing order for a stereo image.
In the above description, the description is limited to stereo imaging. However, performing gain correction and shading correction separately can also be applied to the case of correcting data from many directions such as digital tomography. Thus, a greater effect can be obtained as the number of directions increases.
[0021]
When the number of directions is very large, such as 100 directions, it is conceivable to use a shading coefficient that can be approximated. In other words, shading coefficients in ten representative directions are obtained in advance, and which shading coefficient is used in actual photographing is determined by a shading coefficient selection unit that determines directional approximation.
[0022]
【The invention's effect】
As described above, according to the present invention, it is possible to reduce the number of corrected images required for stereo shooting by separately providing a gain correction coefficient between channels and a shading coefficient using X-rays and a grid. Specifically, assuming that there are three types of distances for performing stereo shooting, the conventional method requires (three types) × (two directions) × (average number of images) = 48 white images. According to the above, it can be achieved with eight gain images + (three types) × (two directions) = 14 images.
[Brief description of the drawings]
FIG. 1 is a diagram showing the concept of the present invention.
FIG. 2 is a block diagram illustrating an example of a system configuration according to an embodiment of the present invention.
FIG. 3 is a flowchart illustrating a flow of processing up to creation of a stereo image in the embodiment.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 11 X-ray generator 12 Image detection means 13 Subject 14 User interface means 15 Image processing means 16 Image display means 17 System control means 20 Offset correction means 21 Averaging means 22 Gain coefficient calculation means 23 Gain correction means 24 Filter means 25 Shading coefficient Calculation means 26 Shading correction means 27 Shading coefficient selection means 28 QA processing means 29 Stereo image creation means 30 Offset storage means 32 Gain coefficient storage means 33, 34 Shading coefficient storage means

Claims (5)

Means for capturing a white image from the front, a white image from the left, and a white image from the right,
Means for calculating a gain correction coefficient from the white image from the front,
Means for calculating a left shading coefficient from the white image from the front and the white image from the left,
Means for calculating a right shading coefficient from the white image from the front and the white image from the right,
Means for photographing the subject under substantially the same conditions as the photographing of the white image from the left and the white image from the right,
Gain correcting the left subject image with the gain correction coefficient, performing shading correction with the left shading coefficient, correcting the right subject image with the gain correction coefficient, and performing shading correction with the right shading coefficient. Stereo photography device. 2. The stereo image capturing apparatus according to claim 1, wherein a plurality of white images are taken from the front, and the gain correction coefficient is calculated based on an average of the plurality of white images. 3. The stereo imaging apparatus according to claim 1, wherein the left shading coefficient is created by applying a high-frequency removal filter to a difference image between the white image from the front and the white image from the left. . A gain correction unit and a shading correction unit,
A stereo photographing apparatus, wherein gain correction is performed on the left and right images with the same parameter, and shading correction is performed on the left and right images with different parameters. A gain correction unit and a shading correction unit,
The radiation image capturing apparatus according to claim 1, wherein the parameter of the gain correction unit is generated from a white image, and the parameter of the shading correction unit is generated from a low frequency component of the white image.

Images