JP2002330342A - Radiation image processing unit, image processing system, radiation image processing method, recording medium, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a radiation image processing unit that can obtain an excellent radiation image with a component of a grid pattern eliminated therefrom from a radiation image obtained by radiography employing a grid. SOLUTION: A generating means 117 generates the image component caused by the grid 103 from a radiation image obtained by the radiography employing the grid 103 to eliminate scattered radiation of radiation emitted on an object 102.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a radiographic image processing apparatus and an image processing apparatus for use in an apparatus or a system for obtaining a radiographic image of a subject by radiography using a grid for removing scattered radiation of radiation. System, a radiation image processing method, a computer-readable storage medium storing processing steps for performing the method,
And programs.

[0002]

2. Description of the Related Art Conventionally, for example, the inside of a subject is visualized by irradiating a subject (object) with radiation represented by X-rays and imaging the spatial distribution (radiation distribution) of radiation components transmitted through the subject. Although the technology that makes it possible is used, since radiation generates scattered radiation (scattered radiation) inside the subject, the scattered radiation (scattered radiation) is imaged together with the directly transmitted radiation (direct ray) transmitted through the subject. Would.

[0003] The process of generating scattered radiation depends on the type of radiation, the physical properties or structure of the subject, and is usually unpredictable. For this reason, various measures are required to obtain a radiation image from which scattered radiation has been removed.

In a method for easily obtaining a radiation image from which scattered radiation has been removed, a typical method is to provide a wall of a radiation shielding material such as lead on a radiation receiving surface and to adjust the angle of the radiation reaching the radiation receiving surface. There is a method of blocking the scattered radiation component by regulating.

Specifically, for example, in X-ray photography in the medical field, a device called a “grid” is installed between a subject and an X-ray receiving surface in order to remove scattered X-rays from the subject such as a human body. That is being done.

[0006] The grid is formed so that an X-ray shielding material such as lead and an X-ray transmitting material such as wood, paper, aluminum, or carbon are oriented so as to face each other in a direction converging to an X-ray source (focal point). , And are arranged alternately at a predetermined width.

Since the grid as described above excludes a part of the direct line, a shadow of the grid (a shadow of the grid, hereinafter also referred to as "grid stripe") is seen on the X-ray receiving surface. By adopting a configuration such as "correcting the spatial period in which the X-ray shielding material and the X-ray transmitting material are alternately arranged" and "making the period relatively high in spatial frequency", the X-ray image For the observer, the uncomfortable feeling that grid stripes exist on the X-ray image is reduced as much as possible.

FIG. 21 schematically shows the configuration of the grid 940. In FIG. 21, "910"
Indicates an X-ray generation source, and “920” indicates an X-ray emission direction. “930” indicates a subject such as a human body, and “950”
Indicates an X-ray receiving surface.

[0009] As shown in FIG.
Is 1 on a two-dimensional plane because of its ease of manufacture.
In general, a stripe structure is formed only in the direction (the vertical direction indicated by the arrow in the figure).

As a method of reducing (removing) the contrast of grid fringes, during X-ray exposure, only the grid is moved in a direction perpendicular to the fringes, and the X-rays on the X-ray receiving surface are reduced.
There is a method that utilizes the integration effect during line irradiation.

Here, the image receiving surface (image receiving portion) of the radiation mainly means a radiation film for directly recording a radiation distribution on a photosensitive material.

Meanwhile, in recent years, a method of treating a radiation image as digital data has become more popular than a method of directly imaging a radiation image for medical use by using a radiation film. For example, the radiation distribution is once converted into an electric signal (analog signal), and the analog signal is digitized (A / D converted) and converted into numerical data (digital data). Thus, processing such as filing, image processing, and monitor display can be performed at low cost on a digitized radiation image (digital radiation image).

However, in the case of a digital radiographic image, it is necessary to sample an image signal two-dimensionally, so that the problem of aliasing based on the sampling theorem becomes obvious.

Specifically, in the case of a general image, by setting the sampling period in space to an appropriate period (a fine period or the like), even if some aliasing occurs, there is no problem in observing the image. .

On the other hand, in the case of a digital radiographic image obtained by using a grid, when a periodic striped pattern due to the grid has a very low frequency due to aliasing, or depending on the sampling period, aliasing may occur. Does not occur, but low-frequency amplitude fluctuations may occur. This is a problem that the image observer is aware of.

Therefore, the following method has been proposed as a method of removing an inappropriate grid stripe pattern due to aliasing or the like. Here, the method of digitizing a radiographic image is roughly classified into two methods, and (method 1) and (method 2) for removing grid stripe patterns for these two methods are given.

(Method 1) In the digitizing method, first, a radiation distribution (radiation intensity distribution) is temporarily converted into another energy distribution (energy distribution such as fluorescence), and the resultant image is scanned to obtain an image. An analog video signal (time signal) that is spatially sampled only in one direction is generated, and the time signal is A / D-converted based on a separately prepared time period. As a typical example of the digitizing method, there is a method in which energy stored in a stimulable phosphor is sequentially fluoresced / condensed by laser scanning to obtain a video signal, and then A / D conversion is sequentially performed. No.

In obtaining a digital radiation image by the above-described digitizing method, as a method of removing an inappropriate grid stripe pattern by aliasing or the like, for example, Japanese Patent No. 250659, Patent No. 2754
068 and JP-A-8-088765.

Specifically, first, scanning for another energy distribution corresponding to the radiation distribution is performed in a direction orthogonal to the grid stripe, and the grid stripe is converted into a periodic signal on a video signal, and the analog signal is scanned. After performing low-pass filtering on the periodic signal, sampling on the time axis is performed. With such a normal anti-aliasing filter configuration, inappropriate grid stripes can be removed.

As a method similar to the above-mentioned method, Japanese Patent No. 2,076,659 discloses a method in which the existence and frequency of a grid-striped image are detected by performing a Fourier transform on a pre-sampled image. A method has been proposed in which a low-pass filter is selected based on the result, and low-pass filtering is performed using the low-pass filter to remove inappropriate grid stripes.

Further, Japanese Patent No. 2754068 and Japanese Patent Application Laid-Open No.
No. 088765 and the like, instead of the analog low-pass filtering process proposed in Japanese Patent No. 250659, etc., sampling on the time axis is made shorter than a desired interval to eliminate aliasing of grid stripe information. After obtaining image information including a grid image, a digital low-pass filtering for removing a grid image component is performed, and then digitally decimated (resampling) to obtain an image at a desired sampling interval. Proposed.

(Method 2) In the digitizing method, first, the radiation intensity distribution is temporarily converted into another energy distribution (energy distribution such as fluorescence or electric field intensity), and then two-dimensionally arranged. By a plurality of electrical signal conversion elements (photodiodes, capacitors, etc.)
Dimensional sampling is performed, and signals from the conversion elements are sequentially extracted and A / D converted. A typical example of the digitization method is a large-screen flat sensor whose technology has been developed in recent years. A so-called radiation flat panel sensor.

By the way, when acquiring a digital radiation image by the digitizing method of the method 2, it is very difficult to remove an inappropriate grid stripe pattern due to aliasing or the like. The reason is that a plurality of electric signal conversion elements such as a radiation flat panel sensor or the like (hereinafter, also simply referred to as “sensor” or “flat panel sensor”) directly sample in a two-dimensional space. This is because an anti-aliasing filter cannot be applied to a typical electric signal.

In order to solve the above-mentioned problem, direct sampling is performed by a sensor in such a degree that aliasing does not occur in a two-dimensional space, and a digital anti-aliasing filter is applied. There is a method of sampling, but in such a method,
Due to the configuration of the electric signal conversion element of the sensor, it is difficult to perform fine sampling in a two-dimensional space, which degrades the performance of the electric signal conversion element itself and causes a huge increase in cost.

Therefore, a method of moving the grid during X-ray irradiation as in the prior art has been performed.

As another method, Japanese Patent Application Laid-Open No. 9-75332 and the like aim to remove grid stripes when a digital X-ray image is obtained by directly sampling in a two-dimensional space by a sensor. A method of preventing the occurrence of inappropriate grid stripes on an image by completely matching the interval and the sampling pitch (pixel pitch of the sensor) and matching the gap between the pixel and the area where direct lines are blocked by grid stripes. Has been proposed.

Further, JP-A-9-78970 and U.S. Pat.
S. In Patent 5801385 and the like, the grid stripe contrast is reduced by making the interval between the grid stripes smaller than the sampling pitch to be equal to or close to the width of the aperture of the light receiving portion of one pixel (one electric signal conversion element). There has been proposed a method for reducing the noise.

Also, U.S. Pat. S. Patent 5.050.
In 198 and the like, an image of a grid stripe pattern (grid image) is stored in advance under a plurality of shooting conditions, and when shooting using a grid is actually performed, the shooting conditions in the plurality of grid images stored in advance are stored. There has been proposed a method of removing a grid image by dividing an image obtained by shooting by a grid image that is equal to or approximated to.

[0029]

However, the conventional radiographic image processing as described above, in particular, a radiographic image is obtained by using a sensor and a grid which are directly sampled in a two-dimensional space as represented by (method 2). In the image processing when the image is acquired, there are the following problems.

First, in the configuration proposed in Japanese Patent Application Laid-Open No. 9-75332 or the like, it is very difficult to completely match the interval between grid stripes and the sampling pitch.
That is, a flat panel sensor generally created in a semiconductor manufacturing process and a grid created by a combination of a relatively thick lead plate are independently created, or the grid itself is These factors make it extremely difficult to completely match the interval between the grid stripes and the pixel pitch (sampling pitch) of the sensor, because it must be made detachable depending on the situation at that time.

Further, JP-A-9-78970 and U.S. Pat. S.
In the configuration proposed in Patent 5801385, etc., it is effective to make the interval between grid stripes smaller than the sampling pitch to be equal to or close to the width of the aperture of the light receiving portion of one pixel, but it is effective to use a sensor (flat panel sensor). When the sampling pitch is reduced to, for example, 0.1 mm or less, very fine intervals of 10 or more grid stripes per mm are required. In the case of such a fine striped grid, the thickness of the lead plate for blocking scattered radiation is almost fixed, so there is no choice but to narrow the area through which direct rays pass. Becomes extremely low, and good radiation imaging cannot be performed.

Further, in the conventional configuration as described above, the grid itself is moved during radiation exposure. However, the movement of the grid increases cost due to a driving system for moving the grid. In addition to increasing the size of the apparatus or system, it is necessary to provide a configuration for adjusting and controlling the relationship between the drive timing and the radiation exposure timing, the relationship between the drive speeds, and the like. Therefore, the configuration of the grid movement is effective for removing grid stripes, but has the above-mentioned limitation and cannot always be adopted.

Therefore, in order to solve the above problems,
Since the obtained radiographic image is digital data, a method of removing grid stripes by digital filtering can be considered. According to this method, if the spatial frequency of the grid stripe and the spatial frequency component of the effective image information based on the subject are completely separated, the grid stripe can be removed with a simple filtering configuration.

As a specific example of the above method, Japanese Patent Application Laid-Open No.
No. 2,785 and the like propose a method of removing or reducing data corresponding to the spatial frequency of grid stripes using Fourier transform.

Further, a normal FIR (Finite In)
A method of removing or reducing data corresponding to the spatial frequency of grid stripes using a pulse response (pulse response) filter is also conceivable.

The grid fringe image is a shadow in which the transmittance of radiation is reduced by an X-ray shielding material such as lead, and is superimposed almost multiplying on the signal. Will be superimposed. Therefore, the above-described filtering can be performed.

In general, the process of manufacturing a grid for removing scattered radiation is controlled with very high precision, and grid stripes having uniform spatial frequency characteristics over the entire image are widely used. Have been. Therefore, there is a possibility that the above-described filtering may be performed only for a single spatial frequency.

Further, actually, an image of a grid stripe (shadow)
Is not an accurate sinusoidal shape, there is a possibility that spatial frequency components of double, triple,... Which are multiplication frequencies may exist. In this case, the conversion process (energy conversion process) in the sensor It is assumed that only the fundamental wave component is received due to blur due to two-dimensional spatial dependence.

However, a problem with the above-described filtering is that it is almost impossible to limit the spatial frequency band of the image component itself.

More specifically, for example, Japanese Patent No. 2754068
The spatial sampling pitch is made very fine, the effective Nyquist frequency is increased after the sampling, and the effective bandwidth (below the Nyquist frequency or lower) is typified by the method disclosed in Japanese Patent Application Laid-Open No. 8-087655. It is clear that grid stripes can be removed by normal filtering without any problem if the image component and the grid stripe component are completely separated from each other if the bandwidth is widened. As a problem, increasing the spatial sampling pitch solely to remove grid fringes is not effective because it leads to a significant increase in the cost of the sensor due to semiconductor processes and other factors, and also causes a reduction in radiation acquisition efficiency.

Therefore, it is cost-effective and efficient to configure the sensor itself with a spatial sampling pitch such that an effective image component is substantially contained in a band lower than the Nyquist frequency, but such a configuration is employed. In such a case, it is unavoidable that there is some overlap between the spatial frequency of the grid stripes and the spatial frequency of the effective image component.

More specifically, for example, FIG.
To explain using (d), first, FIG. 12A shows the image signal when the image to be processed (original image) is viewed in one dimension, and the image signal is 256. It consists of point numbers. FIG. 22B illustrates a case where the image signal illustrated in FIG.
9 shows response characteristics of a filter in a spatial frequency domain. In FIG. 22B, the frequency domain is represented by numerical values of “0” to “128” in consideration of the discrete Fourier transform, and band cut filtering is performed at a position where the value of the spatial frequency is “100”. . FIG. 22 (c) shows the result of applying the filtering shown in FIG. 22 (b) to the image signal shown in FIG. 22 (a). As is clear from the figure, the image signal in FIG. 22C hardly changes from the characteristics of the image signal shown in FIG. FIG. 22D shows a difference between the image signal shown in FIG. 22A and the image signal shown in FIG. 22C for confirmation. FIG. 22 above
As is apparent from (d), there is almost no signal component removed by the filtering.

FIG. 23 (a) is the same as FIG.
This is a signal obtained by adding a portion (so-called edge portion) that rises steeply in the middle to the image signal (original image signal) indicated by. FIG. 23 (b) corresponds to FIG.
23 (b) shows the response characteristics of the filter in the spatial frequency domain when filtering the image signal shown in FIG. 23 (a). FIG. 23 (c) above
Shows the result of applying the filtering shown in FIG. 2B to the image signal shown in FIG. In FIG. 23 (c), it is apparent from the circled portion that the original image signal is deviated from the original image signal and is unstable (artifact). FIG. 23D shows a difference between the image signal shown in FIG. 23A and the image signal shown in FIG. 23C for confirmation. As is clear from FIG. 23D, many vibration components appear in the portion that fluctuates sharply (including both ends of the signal).

FIGS. 22 (a) to 22 (d) and FIG.
As shown in (a) to (d), in the case of a normal image signal, the Nyquist frequency (in FIG.
8 ") and below, which are components that have little information and are not the main components of the image signal, there is little problem even if a steep filtering process is performed at this position. When the image signal has a steep portion (edge portion), the image signal is expressed by using a considerably high frequency component equal to or lower than the above-mentioned Nyquist frequency. Problems (artifacts) occur.

FIGS. 24 (a) to 24 (d) show a simulation of a grid, with respect to the original image signal shown in FIG. 22 (a).
It shows a signal state when a sine wave (sin (2π100χ / 256)) is applied. FIG.
As can be seen from (d), grid stripes are almost removed by the filtering having the response characteristic of the filter shown in FIG. (B) (see FIG. (C)).

FIGS. 25 (a) to 25 (d) illustrate the grid by imitating the original image signal shown in FIG.
It shows a signal state when a sine wave (sin (2π100χ / 256)) is applied. FIG.
As is apparent from (d), the filtering having the response characteristic of the filter shown in (b) of FIG. 23 causes the same artifact as that of (c) of FIG. 23 (see (c) of FIG. 25). .

That is, in a simple filtering process as described in JP-A-3-12785,
When the components of the grid stripes are removed, there is a possibility that the above-described artifacts will be strongly generated. Further, if the width of the impulse response of the filter is reduced in order to reduce the artifact, the response characteristics of the filtering will be reduced in a wide range, and the image itself will have a strong form.

Accordingly, the present invention has been made to eliminate the above-mentioned drawbacks, and it has been proposed to convert a radiographic image obtained by radiography using a grid from a radiographic image obtained by removing grid fringe components. It is an object to provide a radiation image processing apparatus, an image processing system, a radiation image processing method, a storage medium, and a program that can be obtained.

[0049]

For such a purpose,
A first invention is a radiographic image processing apparatus for processing a radiographic image obtained by radiography using a grid for removing scattered radiation from a subject, wherein the radiographic image processing apparatus originates in the grid from data of the radiographic image. Creating means for creating an image component.

According to a second aspect, in the first aspect,
The creation means creates the image component based on a feature that the image component superimposed on the radiation image is stationary over the entire image.

According to a third aspect, in the first aspect,
The creation means has an analysis means for obtaining at least the spatial frequency among the spatial frequency and the angle of the periodic pattern presented by the image component by analyzing the radiation image.

According to a fourth aspect, in the third aspect,
The creating means extracts the predetermined component including the image component from the radiation image based on the analysis result of the analyzing means, and obtains the image component by processing the predetermined component obtained by the extracting means. It is characterized by having processing means and removing means for removing the image component obtained by the processing means from the radiation image.

According to a fifth aspect, in the fourth aspect,
The extraction means performs filtering for extracting the component having the spatial frequency obtained by the analysis means from the radiation image.

According to a sixth aspect, in the fourth aspect,
The processing means may execute, as the processing of the predetermined component, processing of estimating an unsteady portion of the predetermined component from steady portions before and after the predetermined component.

According to a seventh aspect, in the sixth aspect,
The processing means detects the non-stationary part based on envelope information of the predetermined component.

According to an eighth aspect based on the sixth aspect,
The processing means estimates the amplitude and phase of a sine wave corresponding to the image component based on the stationary part and the spatial frequency before and after the unsteady part of the predetermined component, and based on the estimation result. And repairing the irregular part.

According to a ninth aspect, in the eighth aspect,
The processing means, for the predetermined component after the repair,
The image component is obtained by performing further filtering.

In a tenth aspect based on the sixth aspect, the processing means obtains the image component by replacing the non-stationary portion satisfying a predetermined condition among the non-stationary portions with a predetermined value. It is characterized by the following.

According to an eleventh aspect based on the first aspect, the creating means executes the image component creating process for a predetermined line selected from the radiation image.

In a twelfth aspect based on the eleventh aspect, the creation means executes the creation process on an average result of a plurality of lines.

In a thirteenth aspect based on the eleventh aspect, the creation means obtains, as the image component of the line for which the creation processing has not been performed, from a line near the line where the creation processing has been executed. The above-mentioned image components are used.

According to a fourteenth aspect, in the first aspect, there is provided a cutting means for cutting out a partial image corresponding to a radiation irradiation field from the radiographic image, and the creating means is configured to cut out the partial image obtained by the cutting means. The method is characterized in that the image component of the image is created.

A fifteenth invention according to the first invention, further comprising a detecting means for detecting the mounting of the grid, wherein the creating means executes the processing based on a detection result of the detecting means. It is characterized by.

According to a sixteenth aspect, in the first aspect, an image storage means for storing the radiation image from which the image component has been removed by the creation means is provided.

According to a seventeenth aspect, in the first aspect, the solid-state imaging device has an image receiving surface having a size equal to or greater than the range of the spatial distribution of the radiation intensity through the subject and the grid. And imaging means for acquiring the radiation image.

The eighteenth invention is characterized in that, in the first invention, there is provided an image component storage means for storing the image component created by the creation means.

A nineteenth invention is an image processing system in which a plurality of devices are communicably connected to each other, wherein at least one of the plurality of devices is connected to any one of the first to third embodiments.
18. The radiographic image processing apparatus according to any one of 18.

A twentieth invention is a radiation image processing method for processing a radiation image obtained by radiography using a grid for removing scattered radiation from a subject, wherein the method comprises the steps of: The method includes a creation step of creating an image component caused by the grid.

In a twenty-first aspect based on the twentieth aspect, the creating step is a step of creating the image component based on a feature that the image component superimposed on the radiation image is stationary over the entire image. It is characterized by including.

In a twenty-second aspect based on the twentieth aspect, in the creating step, at least the spatial frequency of the spatial frequency and the angle of the periodic pattern represented by the image component is obtained by analyzing the radiation image. An analysis step is included.

In a twenty-third aspect based on the twenty-second aspect, the creating step includes: an extracting step of extracting a predetermined component including the image component from the radiation image based on an analysis result of the analyzing step; A processing step of processing the predetermined component obtained in the step to obtain the image component; and a removing step of removing the image component obtained in the processing step from the radiation image.

In a twenty-fourth aspect based on the twenty-third aspect, the extracting step includes a step of performing filtering for extracting the component having the spatial frequency obtained in the analyzing step from the radiation image. I do.

In a twenty-fifth aspect based on the twenty-third aspect, in the machining step, the processing of the predetermined component is performed by estimating an unsteady portion of the predetermined component from steady portions before and after the predetermined component. And a step of performing a process of performing the following.

In a twenty-sixth aspect based on the twenty-fifth aspect, the processing step includes a step of detecting the non-stationary part based on envelope information of the predetermined component.

In a twenty-seventh aspect based on the twenty-fifth aspect, the processing step corresponds to the image component based on a stationary portion before and after the unsteady portion of the predetermined component and the spatial frequency. A step of estimating the amplitude and phase of the sine wave and repairing the unsteady portion based on the estimation result.

According to a twenty-eighth aspect based on the twenty-seventh aspect, the processing step includes a step of further performing filtering on the predetermined component after the repair to obtain the image component. .

In a twenty-ninth aspect based on the twenty-fifth aspect, in the processing step, the image component is obtained by replacing the unsteady portion satisfying a predetermined condition among the unsteady portions with a predetermined value. It is characterized by including a step.

In a thirtieth aspect based on the twentieth aspect, the creating step includes a step of executing the image component creating process for a predetermined line selected from the radiation image.

In a thirty-first aspect based on the thirtieth aspect, the creating step includes a step of executing the creating process on an average result of a plurality of lines.

In a thirty-second aspect based on the thirty-third aspect, the creating step obtains, as the image component of the line on which the creating process has not been executed, from a line near the line on which the creating process has been executed. Using the image component described above.

According to a thirty-third aspect, in the twentieth aspect, there is provided a cutting-out step of cutting out a partial image corresponding to a radiation irradiation field from the radiographic image, and the creating step comprises the step of cutting out the partial image obtained in the cutting-out step. Generating the image component of the image.

According to a thirty-fourth aspect, in the twentieth aspect, the method further comprises a detecting step of detecting the mounting of the grid, wherein the creating step executes the above-described processing based on a detection result of the detecting step. It is characterized by including.

A thirty-fifth invention is characterized in that, in the twentieth invention, an image storage step of storing the radiation image from which the image component has been removed by the creation step is included.

According to a thirty-sixth aspect, in the twentieth aspect, the solid-state imaging device having an image receiving surface having a size equal to or larger than the range of the spatial distribution to be acquired among the spatial distribution of the radiation intensity through the subject and the grid. And an imaging step of acquiring the radiation image.

A thirty-seventh aspect is characterized in that, in the twentieth aspect, an image component storing step of storing the image component created by the creating step is provided.

According to a thirty-eighth aspect, a program for causing a computer to realize the function of the radiation image processing apparatus according to any one of claims 1 to 18 or the function of the image processing system according to claim 19 is readable by a computer. It is recorded on a simple storage medium.

According to a thirty-ninth aspect, a program for causing a computer to execute the processing steps of the radiation processing method according to any one of claims 20 to 37 is recorded on a computer-readable storage medium.

A fortieth invention is a program for causing a computer to realize the function of the radiation image processing apparatus according to any one of claims 1 to 18 or the function of the image processing system according to claim 19. Features.

[0089] A forty-first invention is a program for causing a computer to execute the processing steps of the radiation processing method according to any one of claims 20 to 37.

[0090]

Embodiments of the present invention will be described below with reference to the drawings.

[Overview of First to Fifth Embodiments] Here, the first to fifth embodiments will be exemplified as embodiments of the present invention. Prior to a specific description of the first to fifth embodiments, an outline thereof will be described.

Here, X-rays are used as an example of radiation, and X-ray images obtained by X-ray photography are processed. The configuration described below is one configuration example to which the present invention is applied, and the present invention is not limited to this.

First, the present invention relates to, for example, JP-A-3-127.
A problem when grid fringes are removed by a conventional filtering process (simple filtering process) as described in No. 85 or the like has been solved by the following configuration.

That is, the signal of the X-ray image to be processed (hereinafter, also simply referred to as “target image signal” or “target image”)
, Which is supposed to exist as a stable stripe pattern over the entire image, and is estimated and obtained (hereinafter, simply referred to as “grid stripe component”). Then, grid fringe components are removed from the target image. For example, when the target image signal is a logarithmic image signal, the obtained grid fringe information is subtracted from the target image signal. This makes it possible to remove grid stripe information stably without affecting the target image signal.

More specifically, for example, a component substantially including the grid stripe component is separated from the frequency indicated by the grid stripe component, and the separated component is processed based on characteristic information that would be indicated by the grid stripe. The processed information is regarded as a grid stripe component, and is removed from the target image signal.

The grid fringe component has a considerably strong component according to the spatial spectrum expression, and if the spatial frequency of the grid fringe is appropriately selected, the Nyquist frequency at the time of sampling (corresponding to 1/2 of the sampling frequency) (Spatial frequency). As a result, a state in which the grid stripe component does not overlap with the main component of the normal image signal as shown in FIGS. 24A to 24D can be easily obtained.

Here, only when there is a steep fluctuation component with respect to the target image signal, the above-mentioned FIGS.
As described with reference to (d), it is difficult to separate the grid stripe component from the target image signal.

In some cases, the target image may include a region where no grid stripe exists. This occurs when an X-ray image of a subject including a portion that blocks X-rays almost completely is obtained, or when a strong X-ray exceeding the dynamic range of the sensor reaches a partial area of the sensor. In this case, there is no grid stripe component in the region due to saturation.

Usually, when an X-ray image is acquired, the amount of X-ray transmitted through the inside of the subject is emphasized.
Dose. In general, it is meaningless to extend the dynamic range of the sensor or the sensor amplifier to the outside of the subject without information. In the present region, the grid stripe component does not exist or the contrast is reduced.

Therefore, in the present invention, in the target image,
An artefact is generated by adaptively coping with both a case where there is a steep variation (for example, an edge portion) that makes it difficult to separate the grid fringe component and the target image signal, and a case where there is a region where the grid fringe itself does not exist. This eliminates grid fringe components without performing. Hereinafter, the outlines of the first to fifth embodiments as specific examples will be described.

In the first embodiment, one frame of a target image (an image obtained by X-ray photography) is analyzed. Then, for pixel defects in the direction orthogonal to the grid stripes, a linear prediction type pixel defect correction process is performed. The grid fringe component is once extracted by low-order FIR filtering based on the above analysis result, and then the envelope of the former filtering result is calculated by vector amplitude calculation of the filtering result and another FIR filtering result. Get information. Then, based on the envelope information, a non-stationary part is extracted from the grid stripe component, and the non-stationary part is repaired by using the surrounding stationary part, so that the grid stripe component as a whole is Convert to a stationary signal sequence. Further, in order to more appropriately extract only the grid fringe component, a filtering process for selectively extracting a spatial frequency corresponding to the grid fringe component is executed, and the result is used as a grid fringe component. The grid stripe component obtained as described above is subtracted from the target image signal. As a result, an image after removing the grid stripe component is obtained.

In the second embodiment, a target image is analyzed, and based on the result, pixel defect correction is performed on the image from which grid stripe components have been removed. As the pixel defect correction here, for example, pixel defect correction based on an average of surrounding pixel values can be applied.

In the third embodiment, a detecting means for detecting the grid mounting is provided, and when the grid mounting is confirmed by the detecting means, the target image is analyzed, and a prediction type pixel defect is determined based on the result. Correction and removal of grid stripe components are performed.

In the fourth embodiment, when the irradiation field of X-ray irradiation is narrowed down so as to correspond to the imaging region of the subject, only the partial image corresponding to the irradiation field is set as the target image. , The processing in the first embodiment is executed.

In the fifth embodiment, when grid stripes do not exist in a partial area of the target image, a method is used in which grid stripe removal processing is not performed on an image of a portion where no grid stripes exist. Specifically, regarding the extracted grid fringe component, information of a portion where no grid fringe component exists is replaced with zero (“0”) data, and the grid fringe removal process is not performed on the portion.

In the first to fifth embodiments, for example, since the extracted grid stripe component is image information, the image may be held. Thereby, even when the grid stripe component is removed from the target image, the original target image, that is, the image before the grid stripe is removed, is thereafter used by using the held image information of the grid stripe component. May be recoverable.

Hereinafter, the process of removing grid stripe components in the first to fifth embodiments will be described in detail. still,
In the following description, the first to fifth embodiments are collectively referred to as “the present embodiment”.

The grid fringe component removal processing is mainly processing including the following first to third processing steps.

First processing step: Line data in the direction orthogonal to the grid stripes is sampled from the target image obtained including the grid stripe components, and the spatial frequency of the grid stripes is determined.

Second processing step: Grid fringe components are sequentially extracted from the target image, and the result (grid fringe component) is subtracted from the target image. At this time, considering the occurrence of the artifact, even if an artifact occurs, The component mainly composed of grid stripes is extracted by FIR filtering of a relatively small span whose influence range becomes small.

Third processing step: Based on the spatial frequency of the grid stripe obtained in the first processing step, the envelope of the component mainly composed of the grid stripe obtained in the second processing step is expressed by
It is determined by calculating the vector amplitude of the component and a component whose phase is shifted by 90 ° by another FIR filtering.

The grid stripe component removal processing including the above-described first to third processing steps will be described more specifically. First, the envelope information obtained in the third processing step must be a positive value. And its features are:
It has the following features (1) and (2). (1) For a steeply changing portion (for example, an edge portion),
It shows a very large value. (2) For a portion where there is no grid stripe, a small value that is almost “0” is shown.

In the present embodiment, based on the above-mentioned envelope information, a value portion indicated by the feature (1) (a very large value portion) and a value portion indicated by the feature (2) (an extremely large value portion) By repairing the grid fringe component with respect to a portion having a smaller value, a more stable creation of the grid fringe component is realized.

As a method of repairing the grid fringe component, for example, the above-mentioned feature (1) is replaced by a component predicted from a stable grid fringe portion around the feature.
There is a method of making a grid stripe component stable as a whole.

For components obtained mainly as described above and mainly composed of grid stripes in which only stable components exist, ordinary filtering is performed on the entire line.
At this time, filtering is performed to extract only a spatial frequency in a narrower range around the spatial frequency of the grid stripe.

Results of the above filtering (extracted components)
Is a grid fringe component in the target image. If the envelope information has the feature (2), that is, if there is a portion where the grid fringe component does not exist in a component mainly composed of grid fringes, the grid Since no fringe exists, the component mainly including the grid fringe in the corresponding portion is “0”.
Replace with

When performing filtering processing on a target image, a fast Fourier transform algorithm may be used in order to perform steep filtering processing stably at high speed. In this case, the number of data points is “2”. (N is a positive integer). For this reason, "0" is padded around the normal data to match the score. The data in the range where “0” is reduced may be considered to be data corresponding to the portion where the envelope information indicates the feature (2).

Note that the spatial frequency effective as the spatial frequency of the grid itself is herein described as Japanese Patent Application No. 2000-02.
30% or more of the sampling frequency (reciprocal of the spatial sampling pitch) as proposed in No. 8161, etc.
% Or less (6 of the Nyquist frequency).
(0% or more and 80% or less) is effective.
The reason for this is that the main components of the image are generally concentrated below 30% of the sampling frequency and 40% of the sampling frequency.
The strong grid fringe component having a spatial frequency of 60% or less appears to have caused another periodic amplitude fluctuation when an interpolation process similar to linear interpolation is performed after sampling, and the grid fringe itself is stabilized. This is because they lack sex.

The spatial frequency of the grid itself is represented by “fg [cy
c / mm] ”and the sampling pitch of the sensor is“ T ”, the spatial frequency fm of the grid stripe is

[0120]

(Equation 1)

It is expressed by the following equation (1).

In the present embodiment, in consideration of the fact that a stripe pattern corresponding to the spatial frequency fm represented by the above equation (1) exists in the target image as a grid stripe component, the first processing step described above. , Accurately extract grid stripes. That is, since the spatial frequency fg of the grid stripes is known in advance, the spatial frequency fm of the target image is searched for around the spatial frequency fm of the target image from the sampled line data. Therefore, it is regarded as the spatial frequency fm of the grid stripe in the target image.

In the above-described second processing step, the spatial frequency fm is subjected to FIR filtering with a span as small as possible around the spatial frequency fm, so that effective image components are almost completely removed. In a state where the influence of an artifact due to a steep change (for example, an edge portion) is within a narrow range, a grid stripe component is roughly extracted.

At this time, for example, the coefficient series of the FIR filter is made an even function in order to eliminate the phase fluctuation, and three or five points of FI are used to satisfy the above narrow range.
An R filter is desirable.

Specifically, for example, assuming that a symmetrical three-point FIR filter is used and its coefficients are (a1, b1, a1),
In order to obtain the coefficients (a1, b1, a1), there are two conditions, that is, the condition that the response at the spatial frequency fm is “1” and the condition that the DC component that is the center value of the image information is “0”. Conditions can be used.

That is, the coefficient calculation is a simultaneous equation represented by the following equation: 2 ^* a1 + b1 = 0 2 ^* a1 ^* cos (2πfmT) + b1 = 1,

[0127]

(Equation 2)

A solution represented by the following equation (2) is obtained.

In the above FIR filtering, the response is “1” at the spatial frequency fm, but the response gradually increases as the spatial frequency becomes higher. Generally, since there is no image component in this portion, grid stripes can be sufficiently extracted even with the FIR filtering.

Further, assuming that FIR filtering of five symmetrical points is used and the coefficients are (a2, b2, c2, b2, a2), the space (F1, B2, c2, b2, a2) must be In addition to the condition that the response at the frequency fm is “1” and the condition that the DC component that is the center value of the image information is set to “0”, the differential value of the response at the spatial frequency fm is “0”. "(Peak) may also be used.

That is, the coefficient calculation is performed according to the above equation (2).
By performing a simple operation from the solution represented by, (−a1 ² , 2a1 (1-b1), 1-2a1 ² − (1-
b1) ² , 2a1 (1-b1), −a1 ² ) are obtained.

As a method of obtaining a symmetric 5-point FIR filtering filter, for example, first, a filter having the coefficients (a1, b1, a1) represented by the above equation (2) is used.
When the shape is subtracted from “1”, the filter has a zero at a space fraction fm. Considering the process of performing the filtering by this filter twice, the spatial frequency fm still has a zero point, but the phase (sign) is not inverted. Such a filter is a symmetric 5-point FIR filtering filter. By subtracting this filter from “1”, a filter having a peak at a target spatial frequency fm can be configured.

FIG. 1 shows a three-point FI as described above.
It shows an example of the shapes (spatial frequency characteristics) of R filtering (hereinafter also referred to as “three-point FIR filtering”) and symmetric five-point FIR filtering (hereinafter also referred to as “five-point FIR filtering”).

In most cases, the result of filtering by the FIR filter shown in FIG. 1 is a result of extracting only grid fringe components. This is because, as is clear from FIG. 1, most of the effective image components mainly composed of low frequency components are removed.

However, as described above, it is true that a considerable amount of effective image components are included in components extracted by FIR filtering. initially,
I would like to perform filtering with a filter having a steep selection characteristic centered on the spatial frequency fm, but even if this is done, the frequency components that make up the rapidly changing portion included in the target image may be included. No change.

Therefore, in order to solve the above problem, in the present embodiment, in the above-described third step, a local envelope of the grid fringe component is obtained, and from the fluctuation, an artifact other than the grid fringe component is detected. By detecting a portion including a component that may generate a streak, only a grid stripe component is stably extracted (created).

Although the envelope of a general signal must be based on the Hilbert transform, the envelope of a single sine wave has a response amplitude of “1” at that frequency, and has a 90% response amplitude.
A spatial filter that causes a phase change of ° (π / 2) is applied, and a vector amplitude (square root of the sum of squares) of the result and the original signal is obtained.

When filtering is performed on discrete data by an FIR filter whose phase fluctuates by 90 °, the coefficients of the FIR filter are point-symmetric (odd function). For example, this coefficient is (-a3, 0, a3)
Then, in order to make the response “1” at the spatial frequency fm, the coefficient (−a3, 0, a3) needs to satisfy the following expression: 2 ^* a3 ^* sin (2πfmT) = 1,

[0139]

(Equation 3)

A solution represented by the following equation (3) is obtained.

The coefficient (−a) of the solution represented by the above equation (3)
The amplitude of the signal sequence obtained by the FIR filter having (3, 0, a3) and the original signal sequence are obtained.

For example, FIG. 2A shows the state shown in FIG.
This shows the result of applying the filtering having the solution coefficients (a1, b1, a1) shown in the above equation (2) to the image signal shown in FIG. As is clear from FIG. 2A, most of the grid stripe components are extracted.

FIG. 3A shows that the image signal shown in FIG. 25A is subjected to filtering having the solution coefficients (a1, b1, a1) shown in the above equation (2). This shows the result of the measurement.

The waveforms shown in FIGS. 2 (b) and 3 (b) (thick waveforms) are shown in FIGS. 2 (a) and 3 (b), respectively.
The result of performing filtering having the solution coefficient (−a3, 0, a3) represented by the above equation (3) on the original image signal shown in (a), and the sum of squares of the result of the filtering with the original image signal This shows an envelope taking the square root of.

In particular, in the envelope shown in FIG. 3B, when attention is paid to the dent portion shown in FIG. 3C, a clearly unsteady component exists in the dent portion. This is because the extracted grid fringe component is abnormally extracted by simple filtering (including the edge component of the target image, etc.), and if subtracted from the target image signal, an artifact occurs in the target image signal after this processing. Means that.

Therefore, in the present embodiment, a range indicating an abnormal numerical value is specified from the envelope obtained as described above, and the grid stripe component in the range is estimated by an estimated value from a numerical sequence around the grid stripe component. Correct (replace). In other words, the grid stripe component is formed (created) by utilizing the characteristic of having a steady component over the entire range, which is a characteristic of the inherent grid stripe component.

The estimated value (predicted value) used in the correction is obtained from the statistical properties of the data around the range showing the abnormal numerical value. For example, since the spatial frequency fm of the grid stripe component is known, the spatial frequency fm can be used as a statistical property.

For example, with the spatial frequency fm of the grid stripe and the phase φ,

[0149]

(Equation 4)

Using the sine wave represented by the following equation (4), a grid fringe component in an unsteady part is formed.

For example, as the simplest method, there is a method of obtaining two coefficients A and φ at a specific frequency from peripheral pixels using Fourier transform (Fourier series expansion).

However, normal Fourier transform cannot be used due to a problem such as a defect (unsteady part) in data. Therefore, here, the Fourier transform is generalized, and the amplitude and phase information is obtained in the sense of the least square. Therefore, the above equation (4) is

[0153]

(Equation 5)

This is transformed into the following equation (5).

In this case, when the data at the sampling point xi is “yi” (the number of data points n) (｛xi, y
i; i = 0 to n−1})

[0156]

(Equation 6)

This is represented by the following equation (6).

At this time, it should be noted that, as the component “xi, yi” used here, only data determined to be a stationary part from the above-described verification of the envelope component is selected. is there. Then, parameters R and I for minimizing the square error ε are obtained as follows.

First,

[0160]

(Equation 7)

Equation (7) becomes:

[0162]

(Equation 8)

This is expressed by the following equation (7 ′). By solving the simultaneous equations of the above equation (7 ′), the parameters R, I
Is obtained, and the phase φ and the amplitude A can be simultaneously estimated.

Here, the data series is k / (2 · fm)
If the section of (k is a positive integer) is equally divided into m at equal intervals, the above equation (7 ′) becomes

[0165]

(Equation 9)

As represented by the following equation (8), a discrete Fourier transform (Fourier series expansion) for obtaining a coefficient of a specific frequency is obtained.

According to the above equation (7 ′) or equation (8), using appropriate appropriate data around the non-stationary part,
By calculating the values of the parameters R and I, repair (replacement) of the unsteady portion removed as inappropriate is performed.

As another repair method, there is a method in which a linear prediction model is considered and repair is performed by sequentially predicting by a linear prediction algorithm without specifying the spatial frequency of grid stripes.

The signal waveform obtained by the above-described repair is generally a steady sine wave, and is a component that expresses the grid fringe component very well.

However, the above signal waveform (the signal waveform of the grid fringe component) originally has a narrow span FIR based on the coefficients (a1, b1, a1) expressed by the above equation (2).
This is a result of filtering, which has the response characteristics of the filter as shown in FIG. 1 and contains a considerably large amount of image components other than grid stripe components.

Therefore, in the present embodiment, the above signal waveform is further subjected to filtering for extracting only a component near the spatial frequency fm of the grid stripe. This filtering is performed on the component in which the non-stationary component has been repaired by the above-described operation.
Artifacts such as ringing due to the filtering do not occur.

When the above-mentioned envelope is formed for the signal of the grid fringe component after the above filtering, if there is a portion where a very small value (a value close to “0”) is observed, The portion is a portion where the grid fringe component was originally not observed for some reason (for example, X-ray is completely blocked or the sensor is in saturation), and a portion where the grid fringe component is not originally present. It is. Therefore, with respect to this portion, the information is recorded and replaced with “0” after the subsequent filtering. The result of this processing is subtracted from the target image signal as a grid stripe component.

In this embodiment, the extraction processing of the grid fringe component is performed while sequentially taking out the line data of interest from the target image signal one line at a time. It is also possible to extract the grid fringe component after averaging the data, that is, weakening the image component or enhancing the grid fringe component.

The above is because the grid stripes and the sensor are usually arranged almost in parallel, and the components of the grid stripes of a certain line and the grid stripes of the lines in the vicinity thereof are very similar. It is.

Therefore, in another embodiment of the present invention, the lines to be processed are thinned out in order to reduce the number of calculation processes for extracting grid fringe components by utilizing the above-mentioned feature of being very similar. A grid stripe component extracted from a certain line is set as a grid stripe component of a line near the grid line component. That is, the line near the line from which the grid stripe component is extracted is not subjected to the extraction process of the grid stripe component, and the grid stripe component extracted from a certain line is used to perform a subtraction process from the line data near the line. Can also be performed.

In order to determine whether or not the above-mentioned thinning-out is possible, in the above-mentioned first processing step, when measuring the spatial frequency of the grid stripes, the lines before and after the sampling or the lines before and after the sampling were measured. The phase difference between the grid stripes of the lines is checked to confirm that the grid stripes are not inclined with respect to the sensor.

In another embodiment of the present invention,
As described above, when removing the grid fringe component from the target image signal, by detecting the signal intensity and the like, the irradiation field in the target image signal is recognized, and only for the image data inside the irradiation field, The above-described grid stripe component removal processing may be performed.

By the way, for example, X
When acquiring a line image, there is a problem of a defective pixel as a problem peculiar to the solid-state imaging device in which a plurality of pixels are arranged. Due to the redundancy of the image information (spatially having a low-frequency component as a main component), if the number of defective pixels is small, in most cases, the defective pixels are repaired by average interpolation from the peripheral pixel values ( Correction) is possible.

However, prediction is generally required due to the statistical properties around the defective pixel. For example, as in the present embodiment, when grid stripes are 50% or more of the Nyquist frequency, the prediction is reversed in the mean interpolation.

FIG. 4 shows a response function as filtering in a case where an arbitrary point is regarded as a defective pixel in one dimension and interpolation is performed using an average of two pixels on both sides thereof. In FIG. 4, the spatial frequency is represented by the horizontal axis. As shown in FIG. 4, if the spatial frequency is low and 50% or less of the Nyquist frequency, the response is "positive", that is, the phase is not inverted. In contrast, the Nyquist frequency of 5
If it exceeds 0%, the phase is inverted, and the expected interpolation result cannot be obtained.

FIG. 5 shows an example of defective pixel interpolation. In FIG. 5, black dots represent pixel values obtained from normal pixels, and points indicated by arrows (“defective pixel positions”) represent defective pixels for which no data has been obtained. FIG. 5 shows a state in which each pixel data (data indicated by a black dot) vibrates finely because grid stripes are reflected. A white dot indicated by “A: interpolated value by average” in FIG. 5 is a pixel value obtained by conventional average interpolation, and is indicated by “B: ideal interpolated value” in FIG. A certain white dot is an interpolation value in consideration of grid stripes.

In the present embodiment, the following two methods are performed to obtain an ideal interpolation value indicated by “B: ideal interpolation value” in FIG. (Method 1: Method of Linear Prediction) In FIG. 5, the data at the defective pixel position is obtained by linear prediction from the surrounding pixels. (Method 2) By performing the above-described grid fringe removal processing, grid fringe components are removed from the original image signal, so that the main component of 50% or less of the Nyquist frequency is eliminated, and interpolation processing by averaging as in the conventional method is performed. Perform

The outline of the linear prediction method (method 1) will be described below.

[0184] First, as the image data to be processed (pixel data), the data sequence _{{X n, X n-1} , X n-2, ···, X np} is given, the data in the "n" X _n But,

[0185]

(Equation 10)

The following equation (9) is used. In the above equation (9), “ε _n ” represents a white noise sequence, and “ai ｛i = 1,..., P｝” represents a linear prediction coefficient. Such a series is called "autoregressive process (AR process)
_Xn ".

When the above equation (9) is rewritten using the delay operator Z ⁻¹ ,

[0188]

[Equation 11]

Expression (10) is obtained.

However, the above equation (10) is

[0191]

(Equation 12)

Since it is expressed by the following equation (10 ′),
About X_nIs the pulse transfer function 1 / A (z_{- 1}Linear with
Filter input ε_nDefined as output for
Torr estimation).

The above equation (9) is expressed by the linear prediction coefficient ai
If {i = 1,..., P} is obtained from a reliable data series, it indicates that the pixel data at the n-th point can be predicted from the pixel data at the (n−1) -th point. .

Linear prediction coefficient ai {i = 1,..., P}
Can be predicted by using the maximum likelihood estimation (least squares estimation) in the process that the device or system is stationary. In other words, it is possible to find one that minimizes the power (variance) of ε _n . Since ε _n is an error obtained by the least squares estimation, it does not have the following correlation components corresponding to the prediction order. Therefore, the prediction error ε _n obtained by being predicted with the necessary and sufficient order p becomes white noise as defined by Expression 9.

Since the variance of the prediction error ε _n is the mean square (mean value “0”), the function E [*] representing the mean is

[0196]

(Equation 13)

This is represented by the following equation (11).

In the above equation (11), R (τ) = E [X _m X _{m +} τ] (covariance function: covar
iance), and in order to obtain the minimum value, if both sides are differentiated by a coefficient a _k and set to “0”,

[0199]

[Equation 14]

The following simultaneous equation (12) is obtained. This is called a normal equation or Yule-Walker equation.

In practice, the calculation of the autocorrelation R (*) is not performed at all pixel points, but uses an estimated value calculated with a limited (given) number of pixels. For example, the Levinson algorithm, which is a fast algorithm, is used.
An algorithm may be used. This Burg
The algorithm is an algorithm based on a maximum entropy method that can be obtained without directly calculating a covariance (autocorrelation) with data having a smaller number of pixels. In these algorithms, if the prediction error has a normal distribution and the number of pixels is large, they match mathematically, but if the number of pixels is small, the Burg algorithm (an algorithm based on the maximum entropy method)
Is advantageous.

The coefficient a _k obtained by the above calculation
Is used to obtain pixel data expected from pixels before and after the defective pixel.

[First Embodiment] The present invention is applied to, for example, an X-ray image acquiring apparatus 100 as shown in FIG.

<Overall Configuration and Operation of X-Ray Image Acquisition Apparatus 100> The X-ray image acquisition apparatus 100 of the present embodiment is an apparatus for acquiring an X-ray image for medical use (for image diagnosis or the like). As shown in FIG. 6, an X-ray generation unit 101 that generates X-rays on a subject 102 (here, a human body) and a grid 1 for removing scattered X-rays from the subject 102
03, a planar X-ray sensor 104 for detecting the distribution of the X-ray dose transmitted through the subject 102, a controller 105 (CONT) of the X-ray generation unit 101, and an electric signal output from the X-ray sensor 104 as digital data. An analog / digital (A / D) converter 106 for converting the image data into digital data, a memory 107 for temporarily storing digital data output from the A / D converter 106 as target image data, A memory 108 for storing the acquired digital data, an arithmetic unit 109 for performing arithmetic processing on the target image data in the memory 107 using the data in the memory 108, and a target image data after the arithmetic processing in the arithmetic unit 109 Conversion table (reference table: Loo)
k Up Table (hereinafter also referred to as “LUT”) 1
10, a memory 111 for storing gain pattern data for correcting a variation in gain of each pixel constituting the X-ray sensor 104, and a gain pattern data of the memory 111 for the converted target image data output from the LUT 110. Arithmetic unit 112 for performing arithmetic processing using
A memory 113 for temporarily storing the target image data after the calculation result in 12, and a memory 11 for storing information (defective pixel position information and the like) on defective pixels unique to the X-ray sensor 104.
4 and a correction processing unit 1 that performs correction processing on the target image data stored in the memory 113 using the information in the memory 114.
15, a grid fringe detecting unit 116 for detecting information about grid fringes in the target image data after the correcting process in the memory 113, and a correcting process in the memory 113 based on the information obtained by the grid fringe detecting unit 116. A grid fringe component extracting unit 117 for extracting a grid fringe component from the subsequent target image data, a memory 118 for temporarily storing the grid fringe component extracted by the grid fringe component extracting unit 117, and a correction process in the memory 113 Target image data to memory 1
An arithmetic unit 119 for subtracting the grid fringe component from the memory unit 18; a memory 120 for temporarily storing the operation result (target image data after the grid fringe component removal) in the arithmetic unit 119;
And an image processing unit 121 that performs image processing on the target image data within 20 and outputs the processed image data.

In the X-ray image acquiring apparatus 100 as described above, the controller 105 of the X-ray generation unit 101, when a generation trigger is applied by an operator from an operation unit (not shown), the X-ray generation by the X-ray generation unit 101 Start radiation. Thereby, the X-ray generation unit 101 emits X-rays to the subject 102 which is a human body.

The X-rays emitted from the X-ray generator 101
An X-ray sensor 104 passes through a grid 103 that transmits the subject 102 and removes scattered X-rays from the subject 102.
To reach.

The X-ray sensor 104 places the X-ray on the surface (image receiving surface) for detecting the distribution of the X-ray dose transmitted through the subject 102.
A plurality of detectors (pixels) for detecting the line intensity are arranged in a matrix, and an electric power corresponding to the X-ray intensity obtained by the plurality of detectors (pixels) arranged in the matrix is provided. Output a signal.

For example, the following sensors (1) and (2) can be applied as the X-ray sensor 104. Sensor (1): A sensor which once converts X-ray intensity into fluorescence, and detects the fluorescence by photoelectrically converting the fluorescence with a plurality of detectors arranged in a matrix. Sensor (2): X-rays emitted to a specific object are photoelectrically converted in the object, attracting free electrons released therefrom to form a charge distribution by a uniform electric field, and forming the charge distribution in a matrix. A sensor adapted to convert to an electric signal by a plurality of arranged charge detectors (capacitors).

The A / D converter 106 is provided for the X-ray sensor 104
The electric signal output from is digitized and output.
Specifically, the A / D converter 106 includes the X-ray generation unit 101
The electric signals output from the X-ray sensor 104 are sequentially converted into digital data and output in synchronization with the emission of the X-rays or the driving of the X-ray sensor 104.

In FIG. 6, one A / D converter 1
06 is provided, for example, a plurality of A / D
Converters may be provided and configured to operate in parallel. As a result, the digital conversion speed can be increased, and the processing can proceed efficiently.

The digital data output from the A / D converter 106 is temporarily stored in the memory 107 as target image data. Therefore, the memory 107 stores digital image data (target image data) which is a set of a plurality of pixel data corresponding to a plurality of pixels constituting the X-ray sensor 104.

[0212] In the memory 108, digital data obtained by imaging in a state where no X-rays are emitted is stored in advance. The digital data is data for removing fixed pattern noise that exists in an offset manner and is unique to the X-ray sensor 104 from the target image data stored in the memory 107. Therefore, in the X-ray image acquisition apparatus 100, imaging is performed in a state where the X-ray generation unit 101 does not emit X-rays, and the digital data acquired thereby is stored in the memory 108 as image data.

The arithmetic unit 109 is stored in the memory 108 for each of a plurality of pixel data constituting the target image data (image data obtained by X-rays transmitted through the subject 102) stored in the memory 107. A process of subtracting pixel data at a corresponding position from a plurality of pieces of pixel data constituting image data (fixed pattern noise image data obtained by imaging without X-rays) is executed.

The LUT 110 converts the target image data processed by the arithmetic unit 109 into a value proportional to the logarithm thereof and outputs it.

[0215] The memory 111 stores gain pattern data for correcting the variation in the gain of each pixel constituting the X-ray sensor 104 with respect to the target image data converted by the LUT 110. For this reason, in the X-ray image acquisition apparatus 100, X-ray imaging is performed in the absence of the subject 102, and the fixed pattern noise is removed from the image data obtained by using the digital data stored in the memory 108. Then, the data obtained by converting the data into a value proportional to the logarithmic value by the LUT 110 is
The data is stored in the memory 111 as gain pattern data.

The arithmetic unit 112 subtracts the gain pattern data of the memory 111 from the target image data output from the LUT 110 (equivalent to division if logarithmic conversion has not been performed).
And output. The target image data subjected to the subtraction processing by the arithmetic unit 112 is temporarily stored in the memory 113.

When image data used for the gain pattern data to be stored in the memory 111 is obtained and photographing is performed with the grid 103 attached, grid stripes appear in the obtained gain pattern data itself. It will be crowded. It is expected that the arithmetic unit 112
When the gain pattern data is subtracted from the target image data, the grid fringe component reflected in the subject 102 is close to the fluctuation of the gain, so that the grid fringe component may be removed. However, it is unlikely that gain pattern data obtained by shooting without the subject 102 will be obtained every time an image is obtained (every actual shooting). In most cases, gain pattern data is obtained once a day or even less frequently. Since the positional relationship between the X-ray generation unit 101 and the X-ray sensor 104 may change for each imaging, the grid fringe component is not removed by the above subtraction. Even if the positional relationship is unchanged, the contrast of grid stripes differs between the shooting with the subject 102 and the shooting without the subject 102 because the X-ray scattered dose and the ray quality are generally different. Stripe components are not removed by subtraction. In any case, if the directions of the grids 103 match, the spatial frequency of the grid stripes does not change. Preferably, when obtaining the gain pattern data, the grid 103 itself should be removed so that grid stripes are not included in the gain pattern data.

The memory 114 stores information on defective pixels unique to the X-ray sensor 104 (defect pixel position information, etc.). Specifically, for example, a generally planar X-ray sensor is manufactured by a semiconductor manufacturing technique, but the yield is not 100%, and for some reason in the manufacturing process,
Some of the detectors (pixels) are defective pixels that have no meaning as detectors, ie, whose outputs have no meaning. Here, the X-ray sensor 104 is inspected in advance in the manufacturing process or by means (not shown), and the position information of the defective pixel obtained as a result is stored in the memory 11.
4 is stored.

The correction processing unit 115 corrects defective pixel data among a plurality of pixel data constituting the target image data stored in the memory 113 based on the position information of the defective pixel stored in the memory 114, and performs the correction. The subsequent pixel data is stored again in the corresponding position of the memory 113.

The grid fringe detecting section 116 has a memory 113
Of the target image data (the image data after the correction processing by the correction processing unit 115) is analyzed for grid stripes,
Spatial frequency fm of grid stripe and angle θ of grid stripe
Is detected and output.

The grid fringe component extraction unit 117
13, the target image data (image data after the correction processing by the correction processing unit 115) is read out, and the spatial frequency fm of the grid stripe obtained by the grid stripe detection unit 116 and the angle θ of the grid stripe are obtained from the read image data. Based on this, a grid stripe component is extracted. The grid fringe component obtained by the grid fringe component extracting unit 117 is temporarily stored in the memory 118.

The arithmetic unit 119 subtracts the grid fringe component stored in the memory 118 from the target image data (the image data after the correction processing by the correction processing unit 115) in the memory 113. The target image data from which the grid fringe component has been subtracted by the arithmetic unit 119 is temporarily stored in the memory 120.

The image processing section 121 performs image processing on the target image data in the memory 120 so that the observer can easily observe the image data. Examples of the image processing here include the following processing.・ Process of removing random noise from the target image. When the target image is displayed, the gradation is converted or the detailed part is emphasized so that the density value becomes easy for the observer to see.・ Cut out unnecessary parts for the observer from the target image,
The information amount of the target image is reduced, or the target image information is compressed.

The target image data processed by the image processing unit 121 is displayed on a display unit or stored in a storage unit or storage medium externally or in the X-ray image acquisition apparatus 100 by means not shown. Recording on a recording medium or analysis processing is performed.

<Specific Configuration and Operation of X-ray Image Acquisition Apparatus 100>
In the following, the following components that seem to require a specific description will be specifically described. (1) Image data of grid fringe components stored in the memory 118 (2) Correction processing of defective pixels by the correction processing unit 115 (3) Detection and detection of grid fringe components by the grid fringe detecting unit 116 and the grid fringe component extracting unit 117 Extraction processing

(1) Image Data of Grid Stripe Component Stored in Memory 118 The image data of the grid stripe component stored in the memory 118 is data to be subtracted from the target image data on which the grid stripe component is superimposed. However, if the data stored in the memory 118 is separately stored in association with the target image data after the subtraction as in the present embodiment, the original image data from which grid stripes have been removed can be reconstructed from the original image data. The target image data on which grid stripes are superimposed can be reproduced. This allows
For example, in the grid removal processing, even if the target image data is damaged due to some trouble, it is possible to return to the original target image data by the above-described reproduction processing.

(2) Correction Processing of Defective Pixel by Correction Processing Unit 115 The correction processing unit 115 executes the processing described below by, for example, software using a microprocessor.

FIGS. 7 to 10 show examples of the distribution of pixel defects in the X-ray sensor 104. FIG. here,
It is assumed that a pixel defect basically exists only with a width of one pixel. This is because an X-ray sensor having a plurality of pixel defects adjacent to each other in a large lump is generally not used because it is difficult to repair defective pixels.

In FIGS. 7 to 10, each square represents a pixel, and a black square represents a defective pixel. The lower part of the figure shows the direction (vertical direction) of the grid stripes of the grid 103.

First, the defective pixel shown in FIG. 7 is a basic pixel defect. As shown in FIG. 7, eight adjacent pixel components a1 to a8 are formed around the defective pixel (black square). Existing.

FIG. 11 schematically shows the signal distribution on the spatial frequency axis of the grid stripe component in the target image in which the defective pixel shown in FIG. 7 exists and the grid stripe component exists. In FIG. 11, the horizontal axis represents the spatial frequency axis u in the horizontal direction of the target image, the vertical axis represents the vertical spatial frequency axis v of the target image, and both the spatial frequency axis u and the spatial frequency axis v. With respect to the axis, “sampling frequency” which is the reciprocal of the pixel pitch and “Nyquist frequency” which is a half value thereof are shown.

Since the grid stripes oscillate in the horizontal direction of the target image and are constant in the vertical direction, the components of the grid stripes must exist on the spatial frequency axis u as shown in FIG. (See white circles in the figure).

In a normal image, the main component is distributed in a spatial frequency region which is equal to or smaller than a half value of the Nyquist frequency, and if there is no grid stripe component, the average of the pixel values on both sides of the defective pixel is determined. Can be interpolated. this is,
This is because the influence of the interpolation on the spatial spectrum indicates, for example, the response function (characteristic) of the filtering shown in FIG.

Therefore, in the case of the correction of the defective pixel shown in FIG. 7, since there is no grid stripe component in the vertical direction, the average of the pixels in the vertical direction, that is, the average of the pixel component a2 and the pixel component a6, or any one of them. By one pixel component,
Almost satisfactory correction is possible.

The defective pixel shown in FIG. 8 is a pixel defect connected in a horizontally long manner (see a black square in FIG. 8). In the case of a defective pixel having such a shape, similarly to the defective pixel shown in FIG. 7, since the vertical direction of each pixel is a direction parallel to the grid stripe, the pixel components above and below the target defective pixel are used.
Almost satisfactory correction is possible.

The defective pixel shown in FIG. 9 is a pixel defect connected in a vertically long manner (see a black square in FIG. 9). In the case of a defective pixel having such a shape, there is no pixel having a reliable value above and below the target pixel except for the defective pixel at the upper end or the defective pixel at the lower end of the connected defective pixel. By performing defect correction on the defective pixel in such a state by simple averaging in the horizontal direction or the like, as described with reference to FIG.
A correction result of an unexpected value is obtained.

Therefore, in the present embodiment, a coefficient a _{k is calculated} using a simultaneous equation as shown in the above equation (12) by using a normal pixel component connected to the right or left side of the target defective pixel.
(K = 1 to P) are obtained from the left and right sides of the target defective pixel. At this time, for example, the number of pixels to be used is set to about 20, and the order k is set to about 5.

Then, using the coefficient a _k , the above equation (9)
, The target defective pixel value _Xn is predicted, and the average of all the defective pixel values _Xn obtained as a result is obtained. As a result, "B: ideal interpolation value" as shown in FIG. 5 is obtained.

In this embodiment, the coefficient a _k (k = 1 to P) is obtained by using the above equation (12). However, the present invention is not limited to this. For example, the maximum entropy Alternatively, an algorithm called a method may be used.

The defective pixel shown in FIG. 10 is a defective pixel obtained by superimposing the state of the defective pixel shown in FIG. 8 and the state of the defective pixel shown in FIG. Among the defective pixels in this state, the problematic pixel is a pixel at a portion where a vertically connected defective pixel intersects with a horizontally connected defective pixel (a pixel overlapping a cross), that is, a pixel component a
This is a defective pixel surrounded by 4, a12, a18, and a24.

The correction of the defective pixel in the state shown in FIG. 10 is performed by correcting the average of the upper and lower pixel components for the linear defective pixel connected in the horizontal direction, and by correcting the linear defective pixel connected in the vertical direction. Is corrected using the simultaneous equations described above. Specifically, the following three methods can be mentioned, but almost the same result can be obtained by using any of the methods.

The pixel components a4, a12, a18, a24
Is corrected using an average value of the upper and lower defect correction values (corrected defective pixel values). The correction of the defective pixel surrounded by the pixel components a4, a12, a18, and a24 is performed by solving the simultaneous equations of the above equation (12) using the values of the left and right pixels that are the defect correction values. Correction of a defective pixel surrounded by the pixel components a4, a12, a18, and a24 is performed using an average value of the result and the result.

By performing the above-described processing, defective pixel data in a plurality of pixel data constituting the target image data stored in the memory 113 is corrected.

(3) Grid fringe component detection and extraction processing by grid fringe detector 116 and grid fringe component extractor 117

The grid fringe detecting section 116 has a memory 113
A part of the target image data stored in the target image data is read, the spectrum of the grid stripe included in the target image data is checked by the read data, and the spatial frequency f of the grid stripe is checked.
m and the angle θ are detected. The information on the spatial frequency fm and the angle θ (hereinafter, also referred to as “angle η”) of the grid stripes is used in a grid stripe component extraction process in a subsequent grid stripe component extraction unit 117.

FIGS. 12A to 12D are diagrams for explaining the detection processing of the spatial frequency fm and the angle θ of the grid stripe.

FIG. 12A shows an image of the entire target image, and “L1” to “L6” indicate the line positions from the top of the target image. The grid fringe detector 116 measures the spatial frequency fm of the grid fringes based on the result of Fourier transform of the lines L1 to L6.
At this time, when detecting the spectrum of the grid fringe, the grid fringe detecting unit 116 performs
The average (or the average of the spectrum) of several lines before and after each of the lines L1 to L6 may be used.

FIGS. 12B to 12D respectively show the results of the Fourier transform on the line L1. That is, FIG. 12B shows the amplitude spectrum (or power spectrum), and FIG. 12C shows the value of the real part which is the coefficient of the cosine wave of the Fourier transform result.
FIG. 4D shows the value of the imaginary part which is the coefficient of the sine wave.

FIG. 13 is a flowchart showing the above-described processing of the grid fringe detecting section 116.

First, the grid fringe detecting section 116 clears a variable “cumulation” for obtaining the average of the spectrum (step S201). Further, the grid fringe detecting unit 116 clears a counter (variable) n of the number of lines to be obtained when calculating the average of the spectrum (step S202). The grid fringe detecting unit 116 initializes a variable i for selecting a processing target line (selected line) from the lines L1 to L6 shown in FIG. 12A to "1" (step S203). Thus, in the first processing, the line L1 is selected and processed as the target line.

Then, the grid stripe detecting section 116 determines whether or not the processing of the next steps S205 to S215 has been executed for all the lines L1 to L6 of the target image (step S204). As a result of this determination, only when the processing is completed, a step S2 described later
Proceeding to step S16, if the processing has not been completed yet, the processing from the next step S205 is executed.

If the result of determination in step S204 is that processing has not been completed, first, the grid stripe detection unit 116 selects a line Li indicated by a variable i from the lines L1 to L6 of the target image and stores the data ( Line data Li) is obtained (step S205).

Next, the grid fringe detector 116 performs a Fourier transform process such as a fast Fourier transform on the line data Li acquired in step S205 (step S205).
206).

Next, the grid fringe detecting section 116 determines the power spectrum (or amplitude spectrum) from the Fourier transform result (data in the spatial frequency domain) in step S206.
Is acquired (step S207).

Next, the grid fringe detecting section 116 determines whether or not a significant spectrum (peak value) indicating grid fringes exists in the power spectrum obtained in step S207 (step S208).

Specifically, for example, first, since the absolute spatial frequency of the grid lead that causes grid stripes is known at the stage when the grid 103 is installed, the frequency is used as “fg”. Thus, the determination processing in step S208 can be performed accurately.

That is, assuming that the sampling pitch of the X-ray sensor 104 is “Ts”, the rough spatial frequency fm at which grid stripes occur is

[0258]

(Equation 15)

This can be specified by the following conditional expression (13).

At this time, in the above equation (13), when the condition shown in the above is satisfied, the spatial frequency fm 'obtained in is used, and when the condition shown in the above is not satisfied, "J ← J + 1 ”.

The accurate spatial frequency fm of the grid fringes should exist near “fm ′” obtained by the above equation (13), and whether or not the peak value (significant spectrum showing grid fringes) exists. When judging whether or not there is a different peak value due to the influence of an image component, a noise component, or the like, a peak that is a significant spectrum indicating grid stripes is not affected even if a different peak value exists due to the influence of an image component, a noise component, or the like. The value can be detected.

The frequency fg of the grid lead of the grid 103 is manufactured quite accurately. However, if the distance between the grid 103 and the X-ray sensor 104 is longer than an arbitrary distance, the X Since the X-ray beam from the ray generation unit 101 has a cone beam shape, the X-ray beam is expanded and reaches the X-ray sensor 104. For this reason, an accurate spatial frequency fm differs, and cannot be easily predicted. Therefore, of the frequencies near “fm ′” obtained by the above equation (13),
The frequency indicating the peak value is obtained as the spatial frequency fm.

However, in the above case, it is necessary to determine whether or not the peak value obtained is a significant peak value that is actually stably present. This determination is usually made based on the noise level. The noise level may be measured in advance, or an average value of components other than the peak value of the high band of the spectrum may be used. For example, if the ratio of the total (or average) of the power spectra of neighboring spectral values indicating the peak value to the noise level is about 10 or more, it is usually determined to be a significant and stable peak value.

If no significant peak value exists as a result of the determination in step S208 as described above, the grid fringe detecting section 116 counts up the variable i in order to process the next line (step S217). Then, the process returns to step S204, and the subsequent processing steps are repeatedly executed.

On the other hand, if the result of determination in step S208 is that there is a significant peak value, the grid fringe detector 116
Sets the spatial frequency indicating the peak value to Pi (step S209) and adds this to the variable cumulation (step Ss210).

The grid fringe detector 116 obtains the phase of the grid fringe, sets the phase and the line position indicated by the variable i for the variable θn and the variable Mn (steps S 211 to S 214), n is incremented (step S215). After that, the grid fringe detecting unit 116 sets the variable i to process the next line.
Is counted up (step S217), the process returns to step S204, and the subsequent processing steps are repeatedly executed.

As described above, all the lines L1 to L
When the processing of steps S205 to S215 is completed for No. 6, the grid pattern detection unit 116 divides the current value of the variable by the current value of the variable n (the number of significant peak values). To determine the spatial frequency fm of the average grid stripe (step S21)
6).

After executing the processing in FIG. 13, the grid fringe detecting unit 116 obtains the variable θ obtained in step S213.
i (i = 0 to n-1) is used to determine an average grid stripe angle η (angle θ) from the line position Mi (i = 0 to n-1) in step S214.

That is, the grid fringe detecting section 116 sets i
A phase theta _i of th line Li, the phase difference between the phase theta _{i + 1} of the (i + 1) th line L (i + 1) (θ i -θ i + 1)
Of the position of the grid 103 is calculated by the following equation using the spatial frequency fm of the grid stripe, and the line of the line Li and the line L (i + 1) is obtained by the following equation: {(θ _i −θ _{i + 1} ) / 2π} / fm The difference is determined by the equation (Mi + 1) −Mi, and based on these results, the angle η of the grid stripe is calculated as

[0270]

(Equation 16)

It is determined by the following equation (14).

If the angle η obtained by the above equation (14) is not abnormally inclined, the grid stripe detecting section 116 executes a process of extracting grid stripes every few lines, If the angle η is abnormally inclined,
The processing of extracting grid stripes is executed for each line.

In the above-described processing executed by the grid fringe detecting section 116, it is assumed that the grid fringe direction (vertical or horizontal direction) is known. For example, when the grid fringe direction is unknown. For example, the same process is performed in advance in both the vertical direction and the horizontal direction, and the direction in which a significant peak value is detected is set as a direction substantially orthogonal to the grid stripes.

The processing described above is executed by the grid fringe detecting unit 116, so that the spatial frequency f
m and the angle θ (angle η) are obtained. The grid fringe component extracting unit 117 uses the spatial frequency fm and the angle θ (angle η) of the grid fringe obtained by the grid fringe detecting unit 116 to store the grid fringe component that is actually stored in the memory 113. Grid stripe components are extracted from the image data, and the grid stripe components are stored in the memory 118.

FIG. 14 is a flowchart showing the grid stripe component extraction processing in the grid stripe component extraction unit 117.

First, the grid fringe component extracting unit 117 receives the grid fringe detecting unit 116 as a processing parameter.
The angle θ of the grid stripe and the spatial frequency fm of the grid stripe are obtained. Grid stripe component extraction unit 117
Executes the processes of steps S300 to S321 described below based on the above processing parameters (grid stripe angle θ and grid stripe spatial frequency fm).

When the value of the spatial frequency fm of the grid stripe given to the grid stripe component extraction unit 117 is "0", for example, no grid stripe exists on the target image, that is, the grid 103 is used. In this case, the grid stripe component extraction unit 117
Does not execute the processing shown in FIG. Also, for example, the memory 118 stores “0” data as a grid stripe component in this case. Alternatively, the arithmetic unit 119
Does not function, the target image data stored in the memory 113 is stored in the memory 120 as it is.

When the grid stripe component angle θ and the grid stripe spatial frequency fm are given from the grid stripe detection unit 116 to the grid stripe component extraction unit 117, first, the grid stripe component extraction unit 117 calculates the equation (2) ) Are used to determine coefficients a1 and a2 (or coefficients (a2, b2, c2, b2, a2) of a symmetric five-point filter) from the spatial frequency fm (step S300). The FIR filter corresponding to the coefficient obtained here is referred to as “FIR1”.

Next, the grid fringe component extraction unit 117 obtains a coefficient a3 from the spatial frequency fm using the above equation (3) (step S301). The FIR filter corresponding to the coefficient obtained here is referred to as “FIR2”.

Next, the grid stripe component extraction section 117 generates a window function centered on the spatial frequency fm in order to perform FIR filtering in the region of the spatial frequency fm (step S302). As the window function here, for example, a Gaussian distribution function centered on the spatial frequency fm can be applied.

Next, the grid fringe component extracting unit 117 determines a range of lines for which grid fringe extraction processing is performed on line data constituting the target image, based on the grid fringe angle θ (step S303). ~ Step S30
5).

More specifically, for example, the grid fringe component extracting unit 117 sets the reference value of the angle θ to “0.1 degrees”, and sets the grid fringe angle θ obtained by the grid fringe detecting unit 116 to the reference value of 0.1. It is determined whether it is larger or smaller than once (step S303). If the result of this determination is that the angle θ is smaller than the reference value of 0.1 degree, the grid stripe component extraction unit 1
17 sets “5” for the variable “skip”, thereby deciding to omit the process for every five lines (step S304). On the other hand, when the angle θ is equal to or larger than the reference value of 0.1 degrees, the grid stripe component extraction unit 117 sets the variable skip to “0” to execute the process on all the lines. Determine (Step S30)
5).

Steps S303 to S305
In, for example, not only the setting for the variable skip, but also a more detailed setting may be performed based on the angle θ.

After the processing of step S304 or S305, the grid stripe component extraction unit 117 sets the variable count
The variable count is initialized by setting the value of the variable skip set in step S304 or step S305 to t (step S30).
6).

The grid fringe component extraction unit 117
It is determined whether or not the current value of the variable “count” is equal to or larger than the value of the variable “skip”, that is, whether or not it is necessary to execute the grid stripe extraction processing for the target line data (step S307). As a result of this determination, if the process is to be executed, the process proceeds to step S310; otherwise, the process proceeds to step S308. However, when the step S307 is first performed, the process is always determined to be executed, and the process proceeds to the next step S310.

If the result of determination in step S 307 is that processing is to be executed (skip ≦ count), first, the grid stripe component extraction unit 117 extracts one line of data to be processed from the target image data stored in the memory 113 (step S 307). The target line data) is obtained (step S310).

In step S310, the target line data may be obtained as it is from the target image data. For example, the average value (moving average value) of several lines before and after the target line data is obtained by actual processing. You may make it acquire as target line data.

Next, the grid fringe component extraction unit 117 converts the target line data acquired in step S310 into
Filtering is performed using the FIR1 for which the coefficient has been determined in step S300, and stored in the line buffer 1 (not shown) (step S311). By this processing, the data of the image component including the grid stripe component is stored in the line buffer 1.

Next, the grid stripe component extraction unit 117 performs filtering using the FIR2 for which the coefficients have been determined in step S301 on the line data stored in the line buffer 1, and sends the result to the line buffer 2 (not shown). It is stored (step S312). As a result of this processing, data for obtaining the envelope of the grid stripe component is stored in the line buffer 2.

Next, the grid fringe component extracting unit 117 obtains the envelope of the grid fringe component (step S313).
That is, the grid fringe component extraction unit 117 calculates the amplitude (that is, the square root of the sum of squares) of the vector having the data in the line buffer 1 and the data in the line buffer 2 as components.
And stores the result in a line buffer 3 (not shown). As the calculation here, even if the calculation does not take the square root, it can be applied due to the monotonic increase of the square root, and the same effect can be obtained.

Next, the grid fringe component extraction unit 117 examines the data in the line buffer 3, ie, the envelope data, and determines the upper limit value th1 and the lower limit value th2 of the threshold value for detecting the abnormal data. Determine (Step S3
14). Various methods can be applied as a method of determining the upper limit value th1 and the lower limit value th2. For example, an average value and a standard deviation value are determined, and the standard value is multiplied by n times (n
Is, for example, a value of about “3”) or more.
h1 and the lower limit value th2, the line buffer 3
A method of obtaining a histogram of the envelope data in the above, and determining the upper limit value th1 and the lower limit value th2 centering on the mode value.

Next, the grid fringe component extraction unit 117 converts the data having a value greater than or equal to the value th1 or less than or equal to the value th2 in the envelope data in the line buffer 3 into abnormal data (data due to a sharp change in image data or the like). ), The data of the grid fringe component in the line buffer 1 corresponding to the abnormal data is estimated and rewritten from data around the abnormal data (step S315). At this time, data rewriting is performed so that the entire grid fringe component is in a stable state indicating a periodic fluctuation pattern. still,
Since the processing in step S315 has been described in the description of the equation (7 ′) and the like, the details are omitted here.

Next, the grid stripe component extraction unit 117 performs a Fourier transform process on the data of the grid stripe component in the line buffer 1 that has become entirely stable by the processing of step S315, and The spatial frequency domain data of the component is obtained (step S316). still,
The transformation process in step S316 is not limited to the Fourier transform, and for example, another orthogonal transform such as a cosine transform can be applied.

Next, the grid fringe component extraction unit 117 performs filtering on the data in the spatial frequency domain obtained in step S316 using a window function centered on the spatial frequency fm obtained in step S302 (step S317). . As a result, the data of the grid fringe component more selectively represents grid fringes.

Next, the grid fringe component extraction unit 117 performs an inverse conversion process of the conversion in step S316 on the data of the grid fringe component after the filtering process in step S317, and compares the result with the actual grid fringe component value. (Step S318).

Next, the grid fringe component extraction unit 117 converts the data of the grid fringe component obtained in step S318 into
The data is stored in the relevant position in the memory 118 (step S31).
9).

Then, the grid stripe component extracting section 117
"0" is set for the variable "count" (step S3).
20) It is determined whether or not the processing from step S307 has been performed on all the line data constituting the target image data in the memory 113 (step S321). As a result of this determination, this process is terminated only when the process is completed, and when the process is not completed yet, the process returns to step S307 again, and the subsequent process steps are repeatedly executed.

On the other hand, if the result of determination in step S307 above is that processing is not to be executed (skip> count),
The grid fringe component extraction unit 117 copies the data of the grid fringe component obtained in the preceding stage to the corresponding position in the memory 118 (step S308), and increments a variable count indicating the line to be copied (step S3).
09), returning to step S307 again, and repeatedly executing the subsequent processing steps.

[Second Embodiment] The present invention is applied to, for example, an X-ray image acquiring apparatus 400 as shown in FIG. X-ray image acquiring apparatus 400 of the present embodiment is different from X-ray image acquiring apparatus 100 shown in FIG. 6 in the following configuration.

The X-ray image acquisition device 400 shown in FIG.
In FIG. 6, components that operate in the same manner as the X-ray image acquiring apparatus 100 shown in FIG. 6 are given the same reference numerals, and detailed descriptions thereof will be omitted.

The X-ray image acquiring apparatus 100 shown in FIG.
In the above, the correction processing unit 115 is configured to perform correction processing of a defective pixel using data in the memory 114 on target image data in the memory 113. On the contrary,
In the X-ray image acquiring apparatus 400 according to the present embodiment, FIG.
As shown in FIG. 5, the correction processing unit 115 performs a process of correcting a defective pixel using the data in the memory 114 on the target image data in the memory 120, that is, the target image data after the grid stripe component has been removed. It was configured as follows.

Therefore, according to the X-ray image acquiring apparatus 400 of the present embodiment, it is not necessary to perform pixel defect correction in consideration of grid stripes, so that the average value of pixel values that are not peripheral defects as conventionally known is required. , A simple defective pixel correction, such as correcting a defective pixel, can be applied.

[Third Embodiment] The present invention is applied to, for example, an X-ray image acquiring apparatus 500 as shown in FIG. X-ray image acquiring apparatus 500 of the present embodiment is different from X-ray image acquiring apparatus 100 shown in FIG. 6 in the following configuration.

Incidentally, the X-ray image acquiring apparatus 500 shown in FIG.
In FIG. 6, components that operate in the same manner as the X-ray image acquiring apparatus 100 shown in FIG. 6 are given the same reference numerals, and detailed descriptions thereof are omitted.

An X-ray image acquiring apparatus 500 according to the present embodiment
As shown in FIG. 16, the configuration of the X-ray image acquiring apparatus 100 of FIG. 6 is further provided with a detection unit (switch) 122 for detecting the attachment of the grid 103.

The detection unit 122 supplies a detection result (grid mounting signal) of the mounting of the grid 103 to the correction processing unit 115 and the grid stripe detection unit 116, respectively.

When the grid 103 is mounted according to the grid mounting signal from the detection unit 122, the correction processing unit 115 performs the defective pixel correction processing in consideration of the grid stripe as described in the first embodiment. Execute. If not, the defective pixel is corrected by the average value of the pixel values of the peripheral non-defective pixels.

Similarly, the grid fringe detecting section 116 also receives the grid 103 from the grid mounting signal from the detecting section 122.
When is mounted, the detection processing (analysis processing) of the grid stripe as described in the first embodiment is executed.
However, when the grid 103 is not mounted according to the grid mounting signal, the grid fringe detecting unit 116 immediately determines that there is no grid fringe without executing the grid fringe detecting process, and performs a process corresponding to the grid fringe. Execute.

As described above, in the X-ray image acquiring apparatus 500 according to the present embodiment, the detection unit 122 is provided, and the grid stripe is detected based on the detection result. The time required for the processing can be greatly reduced.

[Fourth Embodiment] The present invention is applied to, for example, an X-ray image acquiring apparatus 600 as shown in FIG. X-ray image acquiring apparatus 600 of the present embodiment is different from X-ray image acquiring apparatus 100 shown in FIG. 6 in the following configuration.

The X-ray image acquisition device 600 shown in FIG.
In FIG. 6, components that operate in the same manner as the X-ray image acquiring apparatus 100 shown in FIG. 6 are given the same reference numerals, and detailed descriptions thereof will be omitted.

[0312] X-ray image acquisition apparatus 600 of the present embodiment.
As shown in FIG. 17, the configuration of the X-ray image acquiring apparatus 100 shown in FIG. 6 is further provided with a memory 123 for storing X-ray irradiation area data.

In the memory 123, image data (irradiation area data) obtained by cutting out only the area irradiated with X-rays from the target image data stored in the memory 113 is stored. , The detection and extraction of grid fringe components are performed.

Specifically, first, in X-ray photography, generally,
In order to avoid exposure of the subject 102 (here, the human body) to a portion other than the target portion, an irradiation field aperture is provided at the exit of the X-ray generation tube of the X-ray generation unit 101. Thus, X-ray irradiation can be performed only on a necessary portion of the subject 102.

When the above-mentioned irradiation field stop function is used, in the image obtained by X-ray photography, not all the image signals obtained from the X-ray sensor 104 are effective, and the X-rays obtained by the irradiation field stop are not effective.
Only the partial image corresponding to the radiation field is valid.

Therefore, in the present embodiment, X-ray irradiation is performed by computer means (CPU or the like) (not shown) from the target image data in the memory 113 based on the X-ray intensity distribution, aperture shape, or other information. A valid partial image area (irradiation area) corresponding to the field is found, and only data (irradiation area data) of the irradiation area part is stored in the memory 123.

As described above, in the present embodiment, not the irradiation area data in the memory 123, that is, only the data of a necessary portion whose information amount has been reduced, but not the entire target image data, the processing time can be reduced. Can be realized.

In this embodiment, the irradiation area is cut out from the target image after the defective pixel correction.
For example, after cutting out the irradiation area, defective pixel correction may be performed on the irradiation area.

[Fifth Embodiment] In the first embodiment, in the X-ray image acquiring apparatus 100 shown in FIG. 6, the grid stripe component extraction processing in the grid stripe component extraction unit 117 is performed in the manner shown in FIG. The processing was performed according to the flowchart.
In the present embodiment, the grid stripe component extraction process in the grid stripe component extraction unit 117 is, for example, a process according to a flowchart shown in FIG.

In the grid stripe component extraction processing of FIG. 18, the same steps as those in the grid stripe component extraction processing of FIG. 14 are denoted by the same reference numerals, and detailed description thereof will be omitted.

First, before describing the grid stripe component extraction processing in the present embodiment, FIG. 19A shows a part of a target image including a grid stripe component. The portion indicated by “*” indicates a portion where no grid stripe component exists due to the presence of a substance that blocks X-rays.

FIG. 19 (b) shows the result of extracting grid stripe components from the target image of FIG. 19 (a) by filtering in the first embodiment. . As shown in FIG. 19B, in the grid fringe component, an artifact appears in a portion corresponding to a portion indicated by “*” in the target image, and therefore, as described in the first embodiment, the portion is removed. It is extracted from the envelope and estimated from the data before and after the envelope so that the whole becomes a stable grid stripe. FIG. 19 shows the result.
It is a figure of (c). With such a stable grid fringe component, a new artifact does not occur due to conversion processing such as Fourier transform.

In the first embodiment, FIG.
(A) is subtracted from the original image (target image) as shown in FIG. 3 (a) to remove the grid stripe component. It is conceivable that a new grid fringe component appears in a portion where no fringe component exists (a portion indicated by “*”).

Therefore, in this embodiment, FIG.
As shown in (d), in a grid fringe component as shown in (c) of the figure, a portion where a grid fringe component does not originally exist (a portion indicated by “*”) is set to “0”.

For this reason, in the present embodiment, the grid stripe component extraction unit 117 executes the grid stripe component extraction processing according to the flowchart of FIG. That is, after executing the processing of step S318, the grid fringe component extracting unit 117 replaces the portion of the stable grid fringe component data obtained by this processing, which has been guessed in step S315, with “0”. (Step S
700) Then, the process proceeds to the next step S319. As a result, it is possible to reliably prevent a new grid stripe component from being generated by the grid stripe removal process even in a portion where the grid stripe component does not originally exist.

In the first to fifth embodiments, hardware is used. However, the present invention can be realized by controlling the entire apparatus by software.

Further, an object of the present invention is to supply a storage medium storing program codes of software for realizing the functions of the host and the terminal according to the first to fifth embodiments to a system or an apparatus, and to provide the system or apparatus with the storage medium. By reading and executing the program code stored in the storage medium by the computer (or CPU or MPU) of the device,
Needless to say, this is achieved. In this case, the program code itself read from the storage medium implements the functions of the first to fifth embodiments, and the program code and the storage medium storing the program code constitute the present invention. It will be. As a storage medium for supplying the program code, a ROM, a flexible disk, a hard disk, an optical disk, a magneto-optical disk, a CD-ROM, a CD-R, a magnetic tape, a nonvolatile memory card, or the like can be used. The functions of the first to fifth embodiments are realized by executing the program code read by the computer, and the OS running on the computer based on the instruction of the program code. It goes without saying that the present invention includes a case in which the functions of the first to fifth embodiments are realized by performing part or all of the actual processing. Further, after the program code read from the storage medium is written to a memory provided in an extension function board inserted into the computer or a function extension unit connected to the computer, the function extension is performed based on the instruction of the program code. C provided on board and function expansion unit
It goes without saying that a PU or the like performs part or all of the actual processing, and the processing realizes the functions of the first to fifth embodiments.

FIG. 20 shows the functions 80 of the computer.
0 shows a configuration example. Computer function 80
0, as shown in FIG.
M802, RAM 803, and keyboard (KB) 80
9 and a CR of a CRT display (CRT) 810 as a display unit.
T controller (CRTC) 806, hard disk (HD) 811 and flexible disk (FD) 81
The second disk controller (DKC) 807 and a network interface controller (NIC) 808 for connection to the network 840 are connected to each other via a system bus 804 so that they can communicate with each other.

The CPU 801 has a ROM 802 or HD
By executing the software stored in 811 or the software supplied from the FD 812, each component connected to the system bus 804 is comprehensively controlled.
That is, the CPU 801 reads out a processing program according to a predetermined processing sequence from the ROM 802, the HD 811 or the FD 812 and executes the processing program, thereby performing control for realizing the operations in the first to fifth embodiments. .

[0330] The RAM 803 functions as a main memory or a work area of the CPU 801. KBC805
Controls an instruction input from the KB 809 or a pointing device (not shown). CRTC 806 uses C
The display of the RT 810 is controlled. The DKC 807 includes a boot program, various applications, editing files,
It controls access to the HD 811 and FD 812 that store a user file, a network management program, and a predetermined processing program in the first to sixth embodiments. The NIC 808 controls bidirectional data exchange with other devices or systems on the network 840.

[0331]

As described above, according to the present invention, it is possible to obtain a good radiographic image from which radiographic images obtained by radiography using a grid have been removed from which radiographic image components due to the grid have been removed. A processing device, an image processing system, a radiation image processing method, a storage medium, and a program can be provided.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a spatial frequency characteristic of a filter for extracting a grid fringe component from a target image in the first to fifth embodiments.

FIG. 2 is a diagram illustrating an example of a grid stripe component extracted from a target image by the filter.

FIG. 3 is a diagram for explaining another example of a grid stripe component extracted from a target image by the filter.

FIG. 4 is a diagram for explaining spatial frequency characteristics in defective pixel correction of a target image.

FIG. 5 is a diagram for explaining spatial frequency characteristics in defective pixel correction for a target image in which the grid stripe component exists.

FIG. 6 is a block diagram showing a configuration of an X-ray image acquiring apparatus to which the present invention is applied in the first embodiment.

FIG. 7 shows a target image in the X-ray image acquiring apparatus,
FIG. 7 is a diagram for explaining an example (example 1) of a state of a defective pixel.

FIG. 8 is a diagram for explaining an example (example 2) of the state of the defective pixel.

FIG. 9 is a diagram for explaining an example (example 3) of the state of the defective pixel.

FIG. 10 is a diagram for explaining an example (example 4) of the state of the defective pixel.

FIG. 11 is a diagram for explaining a spatial frequency distribution of grid stripe components in a target image in the X-ray image acquiring apparatus.

FIG. 12 is a diagram for describing detection (analysis) of a grid fringe component with respect to a target image in the X-ray image acquisition apparatus.

FIG. 13 is a flowchart for explaining a detection (analysis) process of the grid stripe component.

FIG. 14 is a flowchart illustrating a process of extracting a grid stripe component from a target image based on the result of the grid stripe component detection (analysis) process.

FIG. 15 is a block diagram illustrating a configuration of an X-ray image acquiring apparatus to which the present invention has been applied in the second embodiment.

FIG. 16 is a block diagram illustrating a configuration of an X-ray image acquiring apparatus to which the present invention has been applied in the third embodiment.

FIG. 17 is a block diagram illustrating a configuration of an X-ray image acquiring apparatus to which the present invention has been applied in the fourth embodiment.

FIG. 18 is a flowchart illustrating a process of extracting a grid stripe component from a target image based on the result of the grid stripe component detection (analysis) process in the fifth embodiment.

FIG. 19 is a diagram for describing a process of extracting the grid stripe component using a specific example.

FIG. 20 is a block diagram showing an example of a configuration for reading and executing a program from a computer-readable storage medium storing a program for causing a computer to realize the functions of the first to fifth embodiments.

FIG. 21 is a diagram for explaining radiation imaging using a grid.

FIG. 22 is a diagram for explaining an example of the effect of filtering on an image obtained by the radiography.

FIG. 23 is a diagram for explaining another example of the effect of filtering the image obtained by the radiography.

FIG. 24 is a diagram for explaining an example of an effect of filtering on an image on which grid stripe components are superimposed, obtained by the radiography.

FIG. 25 is a diagram for explaining an example of an effect of filtering on an image on which grid stripe components are superimposed, obtained by the radiography.

[Explanation of symbols]

 REFERENCE SIGNS LIST 100 X-ray image acquisition device 101 X-ray generator 102 Subject 103 Grid 104 X-ray sensor 105 Controller 106 Analog / Digital (A / D) converter 107 Memory 108 Memory 109 Computing unit 110 Conversion table 111 Memory 112 Computing unit 113 Memory 114 Memory 115 Correction processing unit 116 Grid fringe detecting unit 117 Grid fringe component extracting unit 118 Memory 119 Computing unit 120 Memory 121 Image processing unit

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G06T 1/00 290 G06T 1/00 400B 5C077 400 H04N 5/32 H04N 1/409 A61B 6/00 350N 5 / 32 H04N 1/40 101C F-term (reference) 2G088 EE01 GG19 JJ05 KK32 4C093 AA01 CA07 EB24 FD05 FD11 FF01 5B047 AA17 AB02 BA01 BB08 BC07 DA06 DC07 5B057 AA07 AA08 BA03 CA02 CA05 CA12 CB12 CB02CB08 CC12 CB02CB08 EX00 HX30 HX58 5C077 LL02 MP01 PP01 PP02 PQ12 PQ20 PQ24 SS01 TT10

Claims (41)

[Claims] 1. A radiographic image processing apparatus for processing a radiographic image obtained by radiography using a grid for removing scattered radiation from a subject, comprising: an image originating from the grid from data of the radiographic image. A radiation image processing apparatus comprising: a generation unit configured to generate a component. 2. The radiographic image according to claim 1, wherein said generating means generates the image component based on a feature that the image component superimposed on the radiographic image is stationary over the entire image. Processing equipment. 3. The apparatus according to claim 2, wherein said creating means has an analyzing means for obtaining at least the spatial frequency of the spatial frequency and the angle of the periodic pattern presented by the image component by analyzing the radiation image. The radiation image processing apparatus according to claim 1. 4. The extracting means for extracting a predetermined component including the image component from the radiation image based on an analysis result of the analyzing means, and processing the predetermined component obtained by the extracting means. 4. The radiation image processing apparatus according to claim 3, further comprising: a processing unit that obtains the image component; and a removing unit that removes the image component obtained by the processing unit from the radiation image. 5. The apparatus according to claim 4, wherein the extracting means performs filtering for extracting the component having the spatial frequency obtained by the analyzing means from the radiation image.
The radiation image processing apparatus according to claim 1. 6. The processing means for processing, as the processing of the predetermined component, a processing of estimating an unsteady portion of the predetermined component from a steady portion before and after the predetermined component. Item 5. A radiation image processing apparatus according to Item 4. 7. The radiation image processing apparatus according to claim 6, wherein the processing unit detects the non-stationary part based on envelope information of the predetermined component. 8. The processing means estimates an amplitude and a phase of a sine wave corresponding to the image component based on a stationary portion before and after the unsteady portion of the predetermined component and the spatial frequency. 7. The radiation image processing apparatus according to claim 6, wherein the unsteady portion is repaired based on the estimation result. 9. The radiation image processing apparatus according to claim 8, wherein the processing unit obtains the image component by performing further filtering on the predetermined component after the repair. 10. The radiation according to claim 6, wherein the processing means obtains the image component by replacing the unsteady portion satisfying a predetermined condition among the unsteady portions with a predetermined value. Image processing device. 11. The radiation image processing apparatus according to claim 1, wherein the creation unit executes the image component creation processing for a predetermined line selected from the radiation image. 12. The radiation image processing apparatus according to claim 11, wherein said creating means executes the creating process on an average result of a plurality of lines. 13. The image forming apparatus according to claim 1, wherein the creating unit uses, as the image component of the line on which the creating process has not been performed, the image component obtained from the line on which the creating process has been performed in the vicinity of the line. The radiation image processing apparatus according to claim 11. 14. A method according to claim 1, further comprising a cutting unit for cutting out a partial image corresponding to a radiation irradiation field from the radiation image, wherein the creating unit creates the image component of the partial image obtained by the cutting unit. The radiation image processing apparatus according to claim 1. 15. A detecting means for detecting the mounting of the grid, wherein the creating means is configured to detect the mounting of the grid based on a detection result of the detecting means.
The radiation image processing apparatus according to claim 1, wherein the processing is performed. 16. The radiation image processing apparatus according to claim 1, further comprising image storage means for storing the radiation image from which the image component has been removed by the creation means. 17. An imaging unit for acquiring the radiation image by a solid-state imaging device having an image receiving surface having a size equal to or larger than the range of the spatial distribution of an acquisition target among the spatial distribution of radiation intensity via the subject and the grid. The radiation image processing apparatus according to claim 1, wherein: 18. The radiation image processing apparatus according to claim 1, further comprising an image component storage unit that stores the image component created by the creation unit. 19. An image processing system in which a plurality of devices are communicably connected to each other, wherein at least one of the plurality of devices is the radiation image processing according to any one of claims 1 to 18. An image processing system having the function of an apparatus. 20. A radiographic image processing method for processing a radiographic image obtained by radiography using a grid for removing scattered radiation from a subject, the radiographic image processing method comprising: A radiographic image processing method, comprising a generating step of generating an image component to be processed. 21. The method according to claim 20, wherein the creating step includes creating the image component based on a feature that the image component superimposed on the radiation image is stationary over the entire image. Radiation image processing method. 22. The method according to claim 22, wherein the creating step includes an analyzing step of obtaining at least the spatial frequency of the spatial frequency and the angle of the periodic pattern represented by the image component by analyzing the radiation image. 20. The radiation image processing method according to item 20, 23. An extracting step of extracting a predetermined component including the image component from the radiographic image based on an analysis result of the analyzing step, and processing the predetermined component obtained by the extracting step. 23. The radiation image processing method according to claim 22, further comprising: a processing step of obtaining the image component; and a removing step of removing the image component obtained by the processing step from the radiation image. 24. The radiation image processing method according to claim 23, wherein the extracting step includes a step of performing filtering for extracting the component having the spatial frequency obtained in the analyzing step from the radiation image. 25. The processing step of performing a processing of estimating a non-stationary part of the predetermined component from a steady part before and after the predetermined part as the processing of the predetermined component. The radiation image processing method according to claim 23, wherein 26. The radiation image processing method according to claim 25, wherein the processing step includes a step of detecting the unsteady portion based on envelope information of the predetermined component. 27. The processing step comprising: estimating an amplitude and a phase of a sine wave corresponding to the image component based on a stationary portion before and after the unsteady portion of the predetermined component and the spatial frequency. 26. The radiation image processing method according to claim 25, further comprising a step of repairing the unsteady portion based on the estimation result. 28. The radiation image processing method according to claim 27, wherein the processing step includes a step of further performing filtering on the predetermined component after the repair to obtain the image component. 29. The image processing device according to claim 25, wherein the processing step includes a step of obtaining the image component by replacing the non-stationary portion satisfying a predetermined condition among the non-stationary portions with a predetermined value. The radiographic image processing method described in the above. 30. The radiation image processing method according to claim 20, wherein the creating step includes a step of executing the image component creating process for a predetermined line selected from the radiation image. 31. The radiation image processing method according to claim 30, wherein the creating step includes a step of executing the creating process on an average result of a plurality of lines. 32. The creating step includes a step of using, as the image component of the line on which the creating process has not been performed, the image component obtained from a line on which the creating process has been performed near the line. 31. The radiographic image processing method according to claim 30, wherein: 33. A clipping step of cutting out a partial image corresponding to a radiation irradiation field from the radiation image, and the creating step includes a step of creating the image component of the partial image obtained in the cutting step. 21. The radiation image processing method according to claim 20, wherein: 34. The method according to claim 20, further comprising a detecting step of detecting the mounting of the grid, wherein the creating step includes a step of executing the processing based on a detection result of the detecting step. Radiation image processing method. 35. The radiographic image processing method according to claim 20, further comprising an image storing step of storing the radiographic image from which the image components have been removed by the generating step. 36. An imaging step of acquiring the radiation image by a solid-state imaging device having an image receiving surface having a size equal to or larger than the range of the spatial distribution of the acquisition target among the spatial distribution of the radiation intensity via the subject and the grid. 21. The radiation image processing method according to claim 20, wherein: 37. The radiation image processing method according to claim 20, further comprising an image component storing step of storing the image component created in the creating step. 38. A computer-readable storage storing a program for causing a computer to realize the function of the radiation image processing apparatus according to claim 1 or the function of the image processing system according to claim 19. Medium. 39. A computer-readable storage medium storing a program for causing a computer to execute the processing steps of the radiation processing method according to claim 20. 40. A program for causing a computer to realize the function of the radiation image processing apparatus according to claim 1 or the function of the image processing system according to claim 19. A program for causing a computer to execute the processing steps of the radiation processing method according to any one of claims 20 to 37.