JP2001266147A - Method and device for processing picture - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suit at least two pictures respectively optimized as visible pictures suited to observation and interpretation to inter-picture arithmetic in a picture processor. SOLUTION: The picture processor is provided with a reading means 12 for reading two pieces of time series picture information P1 and P2 respectively optimized in accordance with optimizing conditions E1 and E2 and the conditions E1 and E2 respectively corresponding to these pieces of information P1 and P2, which are stored in a storing device 11, and a correcting means 13 for correcting the information P1 older in term time-sequentially of the two pieces of information P1 and P2 based on the conditions E1 and E2 so as to nearly coincide the density (p) of corresponding structures of the two pieces of information P1 and P2 read from the mans 12.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image processing method and apparatus, and more particularly to an improvement in processing of an image to be subjected to an inter-image operation, which is obtained for the same subject.

[0002]

2. Description of the Related Art Conventionally, it has been practiced in various fields to compare and read two or more images of the same subject, check the difference between the two images, and inspect the subject based on the difference. ing.

[0003] For example, in the field of manufacturing industrial products, an image of a certain product photographed in a new condition and an image photographed after a durability test of the product are compared and read to find a part where the difference between the two is large. Attention has been paid to studying areas where product durability should be improved.In the medical field, physicians use multiple radiographic images taken in chronological order of a patient's diseased area. By performing comparative image interpretation, the progress of the disease and the healing status of the disease are grasped to examine a treatment policy.

[0004] As described above, comparative reading of two or more images is routinely performed in various fields. For comparative reading, these two or more images are displayed on an image display device, or In some cases, it is output to a film or another recording medium. That is, an image is converted into a density signal or a luminance signal and then displayed on an image display device or the like, or output to a medium such as a film by a printer or the like.

When two or more images to be compared are output, it is common to simply output those images side by side, but the most interesting thing for the reader is the difference between these images. It is. However, as described above, it is difficult to find the difference by simply arranging two images, for example, as the difference is smaller, and there is a demand for an improvement in the performance of comparative image interpretation.

Therefore, in general, an inter-image operation including a subtraction process in which a structural position (anatomical feature position) is made to correspond between two images to be compared and interpreted is performed. The above differences are extracted and emphasized (Japanese Patent Application No. 11-342900). Since only the differences between the images are extracted and emphasized in this way, it is possible for the radiologist to reliably recognize the differences between the images, so that it is possible to prevent an overlook of a lesion that progresses or heals. it is conceivable that.

At the time of this inter-image operation, it is necessary to perform position adjustment in which the positions of the structures appearing in each image (structural positions) correspond between the two images. For example, as this alignment, a first alignment using a global transformation (for example, affine transformation or the like) of translation, rotation and enlargement / reduction between two images, and an image after the first alignment After dividing the image into a number of local areas and performing matching between the corresponding local areas, curve fitting (for example, 2
Nonlinear distortion conversion (warping) by 10-dimensional polynomial)
There is known a technique of performing the second alignment using a two-stage method (Japanese Patent Laid-Open No. 8-335721).

[0008]

By the way, for example, even when the structure is completely aligned by the above-mentioned alignment technique (Japanese Patent Laid-Open No. 8-335721), the structure corresponding to the two images is not aligned. When specific signal values such as density and luminance do not match, artifacts due to this signal value difference may occur in an image obtained by inter-image operation (inter-image operation image).

That is, the images to be subjected to comparative image interpretation, ie, the images to be subjected to inter-image operation, are not originally obtained for the purpose of inter-image operation, and each image itself is suitable for observation and interpretation. This is an image that has been optimized according to optimization conditions set for each image so as to be reproduced as a visible image. In particular, since images acquired at different times in time series are acquired for the purpose of properly grasping the state at each acquisition time, optimization conditions for only the acquired images are set. Is optimized.

Here, the image optimization is performed, for example, as shown in FIG.
A histogram of the density distribution is created for the original image as shown in (2) (FIG. 2 (2)), and the density portion (A area in FIG. 2 (2)) corresponding to the region of interest to be observed and interpreted is The conversion is performed by inputting the density of the original image and outputting the optimized density as an output so that the visible image having the optimum density and contrast is output (reproduced). In the case of this example, the gradient and the shift amount representing the input / output ratio, which define the input / output relationship of the conversion process, are the optimization conditions.

For this reason, if there is a difference between the two images to be subjected to the inter-image calculation, for example, a change in the lesion due to a change over time exists, the distribution of the density histogram differs between the two images. Optimization conditions are also different (input / output ratio and shift amount). Therefore, since both images are converted into images optimized under different optimization conditions, the same structure having the same density in the original image will have different densities in the image after optimization.

As the above-mentioned optimization processing (also referred to as standardization processing), for example, a photographing condition (image generation condition) by performing so-called pre-reading prior to main reading, disclosed in Japanese Patent Application Laid-Open No. Sho 56-16511, is used. Standardization processing and JP 63
No. 261175, which includes standardization of conditions for image processing so that an original image obtained only by main reading without performing pre-reading has optimum density and contrast. The present invention has been made in view of the above circumstances,
An object of the present invention is to provide an image processing method and apparatus capable of making two or more images, each optimized as a visible image suitable for observation and interpretation, suitable for inter-image calculation.

[0013]

SUMMARY OF THE INVENTION An image processing method and apparatus according to the present invention is provided for optimizing each of image information individually optimized so that signal values of structures in a corresponding image substantially coincide with each other. Is corrected, and then the image information is corrected based on the optimization condition.

That is, according to the image processing method of the present invention, among two or more pieces of image information optimized according to optimization conditions set for each image so as to be reproduced as a visible image suitable for observation and interpretation, Correcting at least one of the two pieces of image information based on the optimization condition so that the signal values of the corresponding structures in the images of the two pieces of image information to be subjected to the inter-image operation are substantially the same. It is characterized by doing.

Here, a medical radiation image can be applied as the image information, and it is preferable that the image information is a time-series image whose photographing time points are different from each other. This is because the optimization conditions often differ for each image, and the effect of the present invention is more remarkably exhibited. As the subject of the image, any object such as animals and plants, industrial products, terrain, celestial bodies, and landscapes can be applied in addition to the human body.

In the correction of image information based on the optimization conditions, for example, a second optimization condition such that the signal values of the corresponding structures of the two image information substantially match is obtained, and the original optimization condition and the second optimization condition are obtained. The image information may be corrected based on the optimization condition, or the corresponding image information may be returned to the original image information before optimization based on the optimization condition based on the optimization condition. Correcting the optimization conditions so that the signal values of the structures in the corresponding image after the new optimization substantially match, and subjecting the original image information to optimization processing based on the corrected optimization conditions; Accordingly, the image information optimized by the original optimization condition may be corrected to the image information optimized by the corrected optimization condition.

As the operation between the images, it is particularly preferable to apply a subtraction process in which pixels between image information representing two images are associated with each other. In this case, simple subtraction or weighting may be used. The subtraction after the execution may be performed. The inter-image operation image obtained by the subtraction processing is generally called a subtraction image. As the subtraction image, two original images having different energy distributions (= original image) obtained by photographing at substantially the same time in time series are used. An energy subtraction image (simple subtraction or load subtraction) obtained based on a high-pressure image (normal radiographic image) and a low-pressure image (high-pressure suppression image), and two original images obtained at different times in time series. Examples include a temporal subtraction image obtained based on an image, a DSA (digital subtraction angiography) image obtained based on two original images of a blood vessel obtained before and after injection of a contrast agent, and the like.

The image information processed by the image processing method of the present invention is preferably subjected to a predetermined alignment process prior to an inter-image operation. The alignment process includes a parallel movement between two images, First alignment using global transformation (for example, affine transformation) such as rotation and enlargement / reduction, and dividing the image after the first alignment into a number of local regions, Based on the result of matching between the regions, curve fitting (for example, two-dimensional 10
A technique for performing the second alignment using the non-linear distortion conversion (warping) based on the degree polynomial in two stages (Japanese Patent Laid-Open No.
No. 335721) is preferably applied. This is because high-accuracy positioning can be realized.

An image processing apparatus according to the present invention is an apparatus for implementing the image processing method according to the present invention. The image processing apparatus is provided with an optimization set for each image so as to be reproduced as a visible image suitable for observation and interpretation. A storage device for storing two or more pieces of image information optimized according to conditions and the respective optimization conditions associated with the respective pieces of image information, and an image inter-image among the image information stored in the storage device 2 to be calculated
Reading means for reading out one image information and the optimization condition corresponding to these pieces of image information from the storage device, so that the signal values of the corresponding structures of the two pieces of read out image information substantially coincide with each other. Correction means for correcting at least one of the two pieces of image information based on the optimization condition.

Here, the correction means finds a second optimization condition such that the signal values of the corresponding structures of the two pieces of image information substantially match, and obtains the original optimization condition and the second optimization condition. Based on, based on the optimization of the image information and optimization conditions, the corresponding image information is returned to the original image information before optimization by the optimization conditions, the optimization conditions,
The correction is performed so that the signal values of the corresponding structures in the image after the optimization are substantially the same, and the original image information is subjected to optimization processing under the optimized optimization conditions after the correction, thereby obtaining the image information. Is preferably applied.

It is preferable that the image information is a medical radiation image, particularly a time-series image having different imaging times.

[0022]

According to the image processing method and apparatus of the present invention, each image is individually optimized in accordance with the optimization conditions set for each image so as to be reproduced as a visible image suitable for observation and interpretation. Among the above image information, the two image information based on the optimization condition are set so that the signal values of the corresponding structures in the image of the two pieces of image information to be subjected to the inter-image calculation substantially match. By correcting at least one of the information, the signal values (density etc.) of the corresponding structures in the two optimized images can be matched, and the signal value difference between the two images can be eliminated. It is possible to prevent the occurrence of an artifact based on a density difference in an inter-image operation image.

[0023]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, specific embodiments of an image processing method and an image processing apparatus according to the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of an image processing apparatus which is an embodiment of the image processing method and the image processing apparatus according to the present invention. The illustrated image processing apparatus 10 includes optimization conditions Ei (i = 1, 2, 2) that are individually set for each image so as to be reproduced as a visible image suitable for observation and interpretation.
..) And two or more pieces of time-series image information Pi (i = 1, 2,...) Optimized in accordance with each of the image information Pi and the respective optimization conditions Ei stored therein.
1 and the image information Pi stored in the storage device 11,
Two pieces of image information P1 (i = 1 in Pi) and P2 (i = 2 in Pi) to be subjected to an inter-image operation (temporal subtraction processing in this embodiment), and correspond to these pieces of image information P1 and P2, respectively. Reading means 12 for reading out the optimization conditions E1 and E2 to be performed from the storage device 11
And the two pieces of image information P read from the reading unit 12
Based on the optimization conditions E1 and E2, the older image information P1 out of the two image information P1 and P2 is corrected so that the densities p of the corresponding structures of P1 and P2 substantially match. And a correction means 13 for performing the correction.

Here, the image information Pi is obtained by irradiating the chest of the human body with radiation, storing the radiation transmitted through the chest of the human body (accumulating and recording the energy of the irradiated radiation, and subsequently generating the excitation light). By irradiating the radiation, the radiation is accumulated and recorded on a sheet-shaped radiation detection panel having a stimulable phosphor layer having a characteristic of emitting a stimulable emission light in an amount corresponding to the energy of the radiation accumulated and recorded. Digital original image information (image information before optimization processing) representing a human chest radiographic image obtained by reading from the stimulable phosphor sheet in which the image is stored was subjected to conversion processing according to predetermined optimization conditions Ei. Image information (image information after optimization processing)
It is.

In the optimizing process, for example, a histogram of the density distribution of the original image (image information before the optimizing process) Pi 'as shown in FIG. A region of interest to be interpreted (for example, a portion excluding a direct radiation incident portion (a so-called plain portion) where radiation is directly incident on the stimulable phosphor sheet without passing through the subject; region A in FIG. 2B) ) Is output (reproduced) as a visible image having an optimum density and contrast by a conversion process in which the density of the original image Pi ′ is input and the density after optimization is output. In the case of this example, the slope and shift amount representing the input / output ratio, which defines the input / output relationship of the conversion process, are the optimization conditions (see FIG. 3C) Ei. The image information after being optimized by the optimization condition Ei is Pi.

The correction means 13 restores the older image information P1 in chronological order to the original image information P1 'before optimization by the corresponding optimization condition E1, based on the optimization condition Ei. E1 is corrected so that the signal values of the structures in the corresponding image after optimization are substantially equal (the optimized optimization condition after correction is E1 ′), and the corrected original image information P1 ′ is corrected. By performing the optimization process under the optimization condition E1 ', the image information P1 optimized under the original optimization process condition E1 is corrected (the corrected image information is referred to as P1 ").

Next, the operation of the image processing apparatus 10 according to the present embodiment will be described.

First, the reading means 12 reads the first image information P1 from the time-series image information Pi stored in the storage device 11.
And the second image information P2, which is newer in time series than the first image information P1, is read out, and the optimization condition (hereinafter referred to as the first optimization) stored in the storage device 11 in association with the first image information P1 An optimization condition (hereinafter, referred to as a second optimization condition) E2 stored in the storage device 11 in association with the condition E1 and the second image information P2 is read.

Here, original image information (first original image information) of the first image information P1 before optimization by the first optimization condition E1.
P1 'has the image state and the density distribution (histogram; indicated by the solid line) shown in FIG. 2A, and the original image information (the second image information) of the second image information P2 before the optimization under the second optimization condition E2. The original image information) P2 'has the image state and the density distribution (histogram; indicated by a solid line) shown in FIG. 2B. The difference between the two images P1' and P2 'is as follows.
As shown in the figure, only the second original image P2 'has a lesion K having a higher density than the peripheral portion.

The optimization conditions E1 and E2 are obtained for each of the original images P1 'and P2', respectively. The optimization condition E1 for the first original image P1 '(shown by a broken line in FIG. 2A). ) And the optimization condition E for the second original image P2 '
2 (2), the input / output ratio is different mainly due to the difference of the presence or absence of the lesion K. Therefore, the first after optimization
The image P1 and the second image P2 have different signal values (density values) in the corresponding structures. When a subtraction process corresponding to a structure is performed between two images having different densities of the corresponding structures, a subtraction image Psu (= P2-P1) obtained by the subtraction process is applied to each structure. Artifacts will result from different values of concentration in the object.

The image processing apparatus 10 according to the present embodiment prevents artifacts caused by the density difference between the images, and includes two pieces of image information P1, P2 and The two optimization conditions E1 and E2 are input to the correction unit 13.

The correction means 13 outputs the older image information P1 in time series among the two input image information P1 and P2.
To the original original image information P1 'using the optimization condition E1.
Is performed. On the other hand, the first optimization condition E1
Is corrected so as to match the second optimization condition E2, and the optimization process defined by the corrected first optimization condition E1 ′ (= E2; indicated by a dashed line in FIG. 2A) is performed. Performed on the first original image information P1 '. As a result, the first original image P1 'and the second original image P2', which are taken in a state where the radiation tube voltage / tube current and the distance from the radiation source are equal and the corresponding structures have the same density, are respectively corrected. After the first
Even after being converted into the corrected first image P1 ″ and second image P2 by the optimization condition E1 ′ and the second optimization condition E2, the signal values (density values) of the structures corresponding to the corrected first image P1 ″ and the second image P2 are the same. The values are substantially the same.

Therefore, when a subtraction process corresponding to a structure is performed between two images P1 ″ and P2 having substantially the same density of the corresponding structure, a subtraction image Psu (= P2) obtained by the subtraction process is obtained. −
P1 ′) is a substantial difference between the two images (eg, lesion K)
Since a constant value is adopted regardless of each structure except for the portion corresponding to the above, it is possible to prevent the occurrence of artifacts due to the density difference.

Note that the first image P1 and the second image P2 have the first original image information P1 'and the second original image information P having the density distributions as shown in FIGS. 3A and 3B, respectively.
The operation of the image processing apparatus 10 of the present embodiment when 2 ′ is optimized by the optimization conditions E1 and E2 will be described below.

That is, since the first original image information P1 'and the second original image information P2' have different levels of irradiation radiation,
Normally, the skin margin a is not likely to overlap with the lesion or itself, that is, hardly causes a substantial difference between the two images (examples other than the skin margin include the shoulder joint head and , A large difference occurs in the density level (horizontal axis in each figure) of the reference member photographed together with the subject for alignment or the like.

As in the case of the above-described embodiment, first, the reading means 12 sets the time series image information Pi stored in the storage device 11 to be more time series than the first image information P1 and the first image information P1. The first optimization condition E1 and the second optimization information E1 stored in the storage device 11 in association with the first image information P1 are read out.
The second optimization condition E2 stored in the storage device 11 in association with the image information P2 is read.

Here, the first original image information P1 'before optimization under the first optimization condition E1 has an image state and a density distribution (histogram; indicated by a solid line) shown in FIG. The second original image information P2 'before optimization under the 2 optimization condition E2 has the image state and the density distribution (histogram; indicated by a solid line) shown in FIG.

The optimization conditions E1 and E2 are obtained respectively for the two original images P1 'and P2'. The optimization conditions E1 for the first original image P1 '(shown by broken lines in FIG. 3A). ) And the optimization condition E2 for the second original image P2 '
(Indicated by a broken line in FIG. 3 (2)). However, the first original image P1 'and the second original image P2'
Means that the radiation dose is different and the input / output ratio is different due to the difference in the presence or absence of the lesion K. Since the shift amounts are significantly different, the first image P1 after optimization and the second image
The image P2 differs from the image P2 in signal value (density value) in the corresponding structure. Therefore, when a subtraction process corresponding to a structure is performed between two images having different densities of corresponding structures, a subtraction image Psu obtained by the subtraction process is obtained.
(= P2−P1) means that the density value of each structure takes a different value to cause an artifact.

The image processing apparatus 10 according to the present embodiment is intended to prevent artifacts caused by the density difference between the images, and the two image information P1, P2 and The two optimization conditions E1 and E2 are input to the correction unit 13.

The correction means 13 optimizes the corresponding optimization conditions E for the two pieces of input image information P1 and P2.
A conversion process for returning to the original original image information P1 ', P2' is performed using 1, E2.

The correcting means 13 detects, for example, a skin margin a which is unlikely to become a lesion from each of the original image information P1 'and P2' by image analysis processing, and detects the detected skin margin a. Is set as a in each density histogram. Then, the correcting means 13 sets the first optimization condition E1 to E1 'based on the second optimization condition E2 so that the signals P1a and P2a of the skin periphery a after the optimization become equal (FIG. 1), and is corrected so that its input / output ratio is the same.

The correcting means 13 further performs an optimization process defined by the corrected first optimization condition E1 '(having the same input / output ratio as the input / output ratio of the second optimization condition E2) in the first condition.
Performed on the original image information P1 '. As a result, the corrected first image P1 ″ obtained by correcting the first original image P1 ′ under the corrected first optimization condition E1 ′ and the second image P2 are the structures corresponding to each other. Is substantially the same value. Therefore, when the subtraction processing corresponding to the structure is performed between the two images P1 ″ and P2 having the substantially same density of the corresponding structure, the subtraction processing is performed. Subtraction image Psu obtained
(= P2−P1 ′) is a constant value irrespective of each structure except for a portion corresponding to a substantial difference between the two images (such as a lesion K). Generation can be prevented.

As described above, according to the image processing apparatus of this embodiment, each image is individually optimized in accordance with the optimization conditions set for each image so as to be reproduced as a visible image suitable for observation and interpretation. By correcting at least one of the two pieces of image information, the signal values (density and the like) of the corresponding structures in the two optimized images can be matched, and the signal value difference between the two images can be reduced. Thus, it is possible to prevent the occurrence of an artifact based on the density difference in the subtraction image.

The image processing apparatus of this embodiment is an example of a configuration in which the storage device 11 storing the image information Pi and the optimization condition Ei is included in the image processing apparatus 10.
The image processing apparatus of the present invention is not limited to such an embodiment, and the storage device 11 may be configured outside the image processing apparatus 10 via a network.

As described above, the optimization condition is not limited to the signal conversion condition for image processing, but may be, for example, an optimum generation condition for image generation.

In each of the above embodiments, the optimized image information is returned to the original image information (the image information before optimization) by inverting the conversion based on the optimization condition, and the optimization is performed. The image information is corrected by correcting the conditions and applying the corrected optimization conditions to the original image information and converting the original image information.

The process of this operation will be described using a process shown in equation (1) which is a simple example of the optimization process. Here, S represents original image information, A and B each represent a parameter of an optimization condition (A is an input / output ratio, B is a shift amount), and Q represents image information after optimization.

Q = A · S + B (1) On the other hand, the parameters obtained by correcting the parameters A and B of the optimization conditions are A ′ and A ′ so that the signal values of the corresponding structures in the image information substantially match. Assuming that B ′, the corrected image information Q ′ is as follows: Q ′ = A ′ · S + B ′ (2)

The operation in the above-described embodiment is as follows. First, the original image information S is obtained by the equation (1), and the obtained original image information S is calculated using the corrected parameters A 'and B'. The corrected image information Q 'is obtained according to 2).

However, the image processing method / apparatus of the present invention is not limited to these embodiments, but includes optimization conditions A and B before correction, optimized image information Q, and optimization conditions A after correction. The corrected image information Q ′ may be obtained based on the image information Q ′ and B ′ without obtaining the original image information S. That is, from equations (1) and (2), A'.QA.Q '= B.A'-B'.A (3) Therefore, Q' = A '(Q-B) / A + B' (4 ), The corrected image information Q 'can be obtained without using the original image information S.

[Brief description of the drawings]

FIG. 1 is a diagram showing an embodiment of an image processing apparatus according to the present invention.

FIG. 2 is a diagram showing each original image and each density histogram of two images to be processed by the image processing apparatus of the embodiment shown in FIG. 1 (part 1)

FIG. 3 is a diagram showing each original image and each density histogram of two images to be processed by the image processing apparatus according to the embodiment shown in FIG. 1 (part 2);

FIG. 4 is a diagram illustrating an optimization process for an image.

[Explanation of symbols]

 10 Image processing device 11 Storage device 12 Readout means 13 Correction means

 ──────────────────────────────────────────────────続 き Continued from the front page F term (reference) GA08 JA01

Claims (10)

[Claims] 1. An image-to-image operation among two or more pieces of image information optimized according to optimization conditions set for each image so as to be reproduced as a visible image suitable for observation and interpretation. Wherein at least one of the two pieces of image information is corrected based on the optimization condition so that the signal values of the corresponding structures in the image of the two pieces of image information substantially match each other. Processing method. 2. Finding a second optimization condition such that the signal values of the corresponding structures of the two pieces of image information substantially match, and based on the original optimization condition and the second optimization condition,
2. The image processing method according to claim 1, wherein the image information is corrected. 3. Returning the corresponding image information to original image information before optimization based on the optimization condition based on the optimization condition, and changing the optimization condition in the corresponding image after optimization to the original image information. The image information is corrected by correcting the signal values of the structures so that they substantially match each other, and performing an optimization process on the original image information based on the corrected optimization condition. 2. The image processing method according to 1. 4. The image information according to claim 1, wherein the image information is time-series images having different photographing points.
An image processing method for an image according to any one of the above. 5. The image processing method according to claim 1, wherein the image information is a medical radiation image. 6. Two or more pieces of image information optimized according to optimization conditions set for each image so as to be reproduced as a visible image suitable for observation interpretation and associated with each of the image information. A storage device for storing the optimization conditions, two of the image information stored in the storage device, two image information to be subjected to an inter-image operation, and the optimization condition corresponding to the image information Reading means for reading out the two image information from the storage device based on the optimization condition so that the signal values of the corresponding structures of the two read image information substantially match. An image processing apparatus, comprising: a correction unit configured to correct at least one of them. 7. The correction means obtains a second optimization condition such that signal values of corresponding structures of the two pieces of image information substantially match, and obtains the original optimization condition and the second optimization condition. 7. The image processing apparatus according to claim 6, wherein the image information is corrected based on a condition. 8. The correction means returns the corresponding image information to original image information before optimization based on the optimization condition, based on the optimization condition, and changes the optimization condition to a response after optimization. The image information is corrected by correcting the signal values of the structures in the image so that they substantially match, and performing an optimization process on the original image information under the optimized optimization condition. 7. The image processing apparatus according to claim 6, wherein: 9. The image information according to claim 6, wherein the image information is a time-series image whose photographing time points are different from each other.
An image processing apparatus for an image according to any one of the preceding claims. 10. The apparatus according to claim 6, wherein the image information is a medical radiation image.
An image processing apparatus according to the item.