JP2010179154A - Image processor, image processing system, image processing method, and storage medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processor capable of providing an excellent image after being processed with such a configuration as to appropriately suppress overshoot without losing information on a fine structure of an overshoot-suppressed region. <P>SOLUTION: First high frequency component acquiring means 112 acquires a first high frequency component from an object image. Region detecting means 113 detects an overshoot region of the object image based on the first high frequency component. Second high frequency component acquiring means 114 acquires a second high frequency component based on the overshoot region. Frequency component substituting means 115 substitutes the second high frequency component for the first high frequency component. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention is an image processing apparatus that is used in an apparatus or system that performs image processing such as gradation conversion processing, dynamic range change processing, and sharpening processing on an image captured by radiation such as X-rays. , An image processing system, an image processing method, a program for realizing the functions of the apparatus or system, a program for executing the processing steps of the method, and a storage medium in which these programs are stored in a computer-readable manner is there.

  In recent years, with the advancement of digital technology, an image (radiation image) obtained by radiography is digitized and acquired as a digital image, image processing is performed on the digital image, and the digital image after the image processing is processed. Are output on a CRT or the like, or output on a film.

  By the way, as for the radiographic image, in order to make the digital image suitable for display output or film output, gradation conversion processing and dynamic range change processing may be performed as the above-described image processing.

  However, even if the above-described image processing is performed, for example, an image obtained by photographing a chest with X-rays (an X-ray chest image) is an image of a lung field that easily transmits X-rays, and an X-ray. Is composed of an image of the mediastinum part that is very difficult to transmit, and therefore the range in which pixel values (values of pixels constituting a digital image) exist is very wide. For this reason, it has been considered very difficult to obtain an X-ray chest image capable of observing both partial images of the lung field and mediastinum simultaneously.

  Therefore, as a method for avoiding the above problem, for example, there is a method described in Non-Patent Document 1 (hereinafter referred to as “first method”).

The first method is the processed pixel value S _D , the original pixel value (input pixel value) S _org , the pixel value S _US of the low-frequency image of the original image (input image), and constants A, B, C (for example, , A = 3, B = 0.7)

It is represented by the following formula (1).

  Further, the first method is to change the weighting of the high frequency component (the first term in the above formula (1)) and the low frequency component (the second term in the above formula (2)). By setting = 3 and B = 0.7, the effect of enhancing the high-frequency component of the target image and compressing the dynamic range of the entire target image can be obtained.

  It should be noted that an evaluation that the first method is effective for diagnosis is obtained from five radiologists as compared with an unprocessed image.

Further, for example, in Patent Document 1 and the like, the processed pixel value S _D , the original pixel value (input pixel value) S _org , and the average profile Py and the X direction profile of the Y direction profile of the original image (input image) are described. I more than the average profile P _x,

A method represented by the following formula (2) (hereinafter referred to as “second method”) is described.

In the above formula (2), function F (x) in the "x> D _th", F (x) is "0", the "0 ≦ x ≦ D _th", F (x) is the intercept “E” and the slope “E / D _th ” monotonously decrease,

It is shown by the following formulas (3) to (5).

In the above formulas (4) and (5),
i = 1 to n
It is. “P _yi ” and “P _xi ” indicate profiles in the Y direction and the X direction. G (P _x , P _y ) is, for example,

It is shown by the following formula (6).

In the second method, the pixel value of the low frequency portion of the target image is compressed with a density range of “D _th ” or less.

  In addition, as a method similar to the second method, there are methods described in Non-Patent Document 2, Patent Document 2 and the like (hereinafter referred to as “third method”).

The third method is a pixel value S _D after processing, an original pixel value (input pixel value) S _org , and an average pixel value S when a moving average of the original image (input image) is taken with a mask size of M × M pixels. _{With US} and a monotonically decreasing function f (X),

It is represented by the following formulas (7) and (8).

  The third method is different from the second method represented by the above formula (2) in the method of creating a low frequency image. That is, in the above formula (2), a low frequency image is created from one-dimensional data, whereas in the above formulas (7) and (8), a low frequency image is created from two-dimensional data.

Similarly to the second method, the third method also compresses a density range equal to or lower than “D _th ” with the pixel value of the low-frequency portion of the target image.

  The dynamic range compression in the first to third methods as described above includes a function f1 () for converting a low-frequency image and a constant AX.

It is represented by the following formula (9).

  Patent Document 3 and the like describe the following method (hereinafter referred to as “fourth method”).

  The fourth method includes a gradation conversion function F1 (), a gradation conversion rate c (x, y), a pixel value fd (x, y) of the processed image, a first image f0 (x, y), The pixel value f1 (x, y) of the second image, the pixel value fus (x, y) of the smoothed (low frequency) image of the second image, and the coordinates (x, y) on the image,

A relationship represented by the formula

It is represented by the following formula (10).

  The image processing according to the fourth method can be easily combined with the gradation conversion processing, has the effect that it can be performed simultaneously with the gradation conversion processing, and has the effect that the high-frequency component can be adjusted simultaneously.

In addition to the image processing according to the first to fourth methods as described above, there is a so-called sharpening process for enabling easy observation of the fine structure of the image.
For example, the sharpening process includes an original image f (x, y), a processed image fp (x, y), a low-frequency image S _US 2 (x, y) of the original image, and a constant C. ,

There is a process represented by the following equation (11).

However, in the process represented by the above formula (11), a process for creating a high-frequency image by subtracting the low-frequency image S _US 2 (x, y) of the original image from the original image f (x, y) is performed. Is called. At this time, a high frequency component that causes overshoot or undershoot may occur in a portion of the high frequency image corresponding to the edge portion of the original image f (x, y). In this case, the high frequency component is added to the original image, the original image after gradation conversion, or the low frequency image in the above processing, and as a result, after the processing in which overshoot or undershoot occurs. Will be obtained.

Here, the overshoot and undershoot have an excessively large absolute value among the high-frequency components obtained by subtracting the low-frequency image S _US 2 (x, y) of the original image from the original image f (x, y). Means a component, with an absolute value being positive overshoots and a negative absolute value being undershoots. Hereinafter, when simply referred to as overshoot, in principle, overshoot and undershoot are collectively referred to.

  Therefore, as a method for suppressing the above-described overshoot, for example, in Patent Document 4 and the like, the high-frequency component of the target image is further subjected to gradation conversion, thereby being a component having a large absolute value of the high-frequency component and corresponding to overshoot. A method for suppressing the portion has been proposed. In addition, a method for suppressing the above-described overshoot by creating a low frequency component using an intermediate value filter or a morphological filter has been proposed.

Japanese Patent No. 2509503 Japanese Patent No. 2663189 JP-A-10-272283 JP 11-1224749 A

SPIE Vol. 626 Medicine XIV / PACS (1986) Journal of Japanese Society of Radiological Technology, Vol. 45, No. 8, August 1989, p. 1030, Anan et al.

  However, in the conventional image processing method for suppressing the overshoot as described above, for example, in the image processing method described in Patent Document 4 or the like, an overshoot with a relatively small amplitude can be appropriately suppressed. As for overshoot, the amount of information of the high frequency component itself is lost when the overshoot is suppressed. For this reason, although the overshoot is suppressed, there is a problem that it is difficult to see the fine structure of the image in the region where the overshoot is suppressed.

  In addition, in the image processing method using an intermediate value filter or a morphological filter, a low frequency component is created by the filter, so that overshoot hardly occurs, but much processing time (calculation time) is required. In addition, there is a problem that the restoration of the fine structure is deteriorated depending on the region where the low frequency component is created.

  Therefore, the present invention was made to eliminate the above-described drawbacks, and is a configuration that realizes that information on the fine structure of the region where overshoot is suppressed is not lost while appropriately suppressing overshoot. , An image processing apparatus, an image processing system, an image processing method, a program for realizing the functions of the apparatus or system, a program for executing the processing steps of the method, and those An object of the present invention is to provide a storage medium that stores the program in a computer-readable manner.

  Under such an object, the first invention provides a first high-frequency component acquisition means for acquiring a first high-frequency component from the target image, and a first high-frequency component obtained by the first high-frequency component acquisition means. Based on the area detection means for detecting the overshoot area of the target image, the second high frequency component acquisition means for acquiring the second high frequency component based on the overshoot area obtained by the area detection means, Frequency component replacement means for replacing the first high-frequency component obtained by the first high-frequency component acquisition means with the second high-frequency component obtained by the second high-frequency component acquisition means. .

  The second invention is based on the first high-frequency component acquisition means for acquiring the first high-frequency component from the target image and the first high-frequency component obtained by the first high-frequency component acquisition means. Area detection means for detecting an overshoot area, second high frequency component acquisition means for acquiring a second high frequency component based on the overshoot area obtained by the area detection means, and acquisition of the first high frequency component The first high frequency component obtained by the means is replaced with the second high frequency component obtained by the second high frequency component acquisition means, and the high frequency component substituted by the frequency component substitution means. And post-processing image acquisition means for acquiring a target image for output.

  According to a third aspect, in the second aspect, the post-processing image acquisition unit performs a predetermined conversion on the high frequency component replaced by the frequency component replacement unit and adds the result to the target image. It is acquired as a target image.

  In a fourth aspect based on the third aspect, the target image is an input image, an image after gradation conversion of the input image, a low-frequency image obtained from the input image, and gradation conversion of the low-frequency image. It includes at least one of the subsequent images.

  According to a fifth invention, in the first or second invention, the second high-frequency component acquisition means acquires a high-frequency component according to a higher frequency region than the first high-frequency component as the second high-frequency component. It is characterized by that.

  In a sixth aspect based on the first or second aspect, the second high-frequency component acquisition means compresses the predetermined low-frequency component included in at least the first high-frequency component to compress the second high-frequency component. It is obtained as a component.

  In a sixth aspect based on the sixth aspect, the second high frequency component acquisition means compresses the first high frequency component by dynamic range.

  In an eighth aspect based on the first aspect or the second aspect, the second high frequency component acquisition means uses a smoothed image obtained by using at least one of an intermediate value filter and a morphological filter. The second high-frequency component is obtained.

  According to a ninth invention, in the first or second invention, the second high-frequency component acquisition means acquires the second high-frequency component only for the overshoot region obtained by the region detection means, The frequency component replacement means converts the first high frequency component obtained by the first high frequency component acquisition means corresponding to the overshoot region into the second high frequency component obtained by the second high frequency component acquisition means. It is characterized by replacing.

  According to a tenth aspect of the present invention, at least the first high-frequency component acquisition unit that acquires the first high-frequency component from the target image and the first high-frequency component obtained by the first high-frequency component acquisition unit are predetermined. And a second high-frequency component acquisition unit that obtains a second high-frequency component by performing a predetermined process including a process of reducing the low-frequency component.

  In an eleventh aspect of the present invention, a first high-frequency component acquisition unit that acquires a first high-frequency component from a target image and a first high-frequency component obtained by the first high-frequency component acquisition unit are at least predetermined Based on second high frequency component acquisition means for obtaining a second high frequency component by performing predetermined processing including processing for reducing low frequency components, and on the high frequency component obtained by the second high frequency component acquisition means, And a post-processing image acquisition means for acquiring a target image for output.

  In a twelfth aspect based on the eleventh aspect, the post-processing image acquisition unit performs a predetermined conversion on the high-frequency component replaced by the second high-frequency component acquisition unit and adds the result to the target image. It is acquired as the target image for output.

  In a thirteenth aspect based on the twelfth aspect, the target image includes an input image, an image after gradation conversion of the input image, a low frequency image obtained from the input image, and gradation conversion of the low frequency image. It includes at least one of the subsequent images.

  In a fourteenth aspect based on the tenth or eleventh aspect, the second high frequency component acquisition means is at least a predetermined low frequency of the first high frequency component obtained by the first high frequency component acquisition means. A second high frequency component is obtained by performing predetermined processing including processing for reducing low frequency components exceeding a predetermined threshold among frequency components.

  In a fifteenth aspect based on the tenth or eleventh aspect, the second high-frequency component acquisition means is at least a predetermined low frequency of the first high-frequency component obtained by the first high-frequency component acquisition means. A second high frequency component is obtained by performing predetermined processing including processing for reducing a low frequency component exceeding a predetermined threshold among frequency components and processing for adjusting the predetermined high frequency component.

  A sixteenth aspect of the invention is an image processing system in which a plurality of devices are communicably connected to each other, and at least one of the plurality of devices is any one of the first to fifteenth aspects of the invention. It has the function of the image processing apparatus.

  According to a seventeenth aspect of the present invention, a first high-frequency component acquisition step for acquiring a first high-frequency component from the target image, and a first high-frequency component obtained by the first high-frequency component acquisition step, An area detection step for detecting an overshoot area, a second high-frequency component acquisition step for acquiring a second high-frequency component based on the overshoot area obtained by the area detection step, and the first high-frequency component acquisition And a frequency component replacement step of replacing the first high frequency component obtained by the step with the second high frequency component obtained by the second high frequency component acquisition step.

  An eighteenth aspect of the invention is characterized in that, in the seventeenth aspect of the invention, a post-processing image acquisition step of acquiring a target image for output based on the high frequency component replaced by the frequency component replacement step.

  According to a nineteenth aspect of the present invention, there is provided a first high-frequency component creating step for creating a first high-frequency component of a target image and an overshoot region of the target image by analyzing the high-frequency component created by the first high-frequency component creating step. Detecting a region, a second high-frequency component generating step for generating a second high-frequency component of the target image, and a first high-frequency component generating step corresponding to the overshoot region detected by the region detecting step And a frequency component replacement step of replacing the first high frequency component created by the step with the second high frequency component created by the second high frequency component creation step.

  According to a twentieth aspect of the present invention, a first high-frequency component creating step for creating a first high-frequency component of a target image, and an overshoot region of the target image by analyzing the high-frequency component created by the first high-frequency component creating step Detecting a region, a second high-frequency component generating step for generating a second high-frequency component of the target image, and a first high-frequency component generating step corresponding to the overshoot region detected by the region detecting step The frequency component replacement step for replacing the first high-frequency component created by the above-mentioned procedure with the second high-frequency component created by the second high-frequency component creation step, and the high-frequency component replaced by the frequency component substitution step are predetermined. And a post-processing image acquisition step of acquiring the post-processing image by performing the above conversion and adding it to the target image. To.

  In a twenty-first aspect, in the nineteenth or twentieth aspect, the target image is an input image, an image after gradation conversion of the input image, a low-frequency image obtained from the input image, and a rank of the low-frequency image. It includes at least one of the images after the tone conversion.

  In a twenty-second aspect based on the nineteenth or twentieth aspect, in the second high-frequency component creating step, the second high-frequency component is the first high-frequency component created in the first high-frequency component creating step. The method includes a step of setting a high frequency component depending on a higher frequency region.

  According to a twenty-third aspect, in the nineteenth or twentieth aspect, the second high-frequency component creating step is configured such that the second high-frequency component is the first high-frequency component created by the first high-frequency component creating step. Including a step of using a high frequency component subjected to dynamic range compression.

  In a twenty-fourth aspect based on the twenty-third aspect, the second high-frequency component creating step is a step of creating a smoothed image used when performing the dynamic range compression by an intermediate value filter or a morphological filter. It is characterized by including.

  In a twenty-fifth aspect based on the nineteenth or twentieth aspect, in the second high-frequency component creating step, the second high-frequency component is generated from a smoothed image obtained by an intermediate value filter or a morphological filter. It is characterized by including the step to create.

  In a twenty-sixth aspect based on the nineteenth, twenty-fourth, twenty-fourth, or twenty-fifth aspect, the second high-frequency component creating step includes the second high-frequency component only in the overshoot region detected by the region detecting step. It is characterized by including the step to create.

  In a twenty-seventh aspect of the present invention, at least the first high-frequency component acquisition step for acquiring the first high-frequency component from the target image and the first high-frequency component obtained by the first high-frequency component acquisition step And a second high-frequency component acquisition step for obtaining a second high-frequency component by performing predetermined processing including processing for reducing the low-frequency component.

  In a twenty-eighth aspect of the present invention, at least a first high-frequency component acquisition step for acquiring a first high-frequency component from a target image and the first high-frequency component obtained by the first high-frequency component acquisition step Based on the second high-frequency component acquisition step for obtaining the second high-frequency component by performing predetermined processing including processing for reducing the low-frequency component, and the high-frequency component obtained by the second high-frequency component acquisition step, And a post-processing image acquisition step of acquiring a target image for output.

  In a twenty-ninth aspect based on the twenty-eighth aspect, the post-processing image acquisition step performs a predetermined conversion on the high-frequency component acquired in the second high-frequency component acquisition step and adds the result to the target image. It is acquired as the target image for output.

  In a thirtieth aspect based on the twenty-ninth aspect, the target image is an input image, an image after gradation conversion of the input image, a low-frequency image obtained from the input image, and gradation conversion of the low-frequency image. It includes at least one of the subsequent images.

  In a thirty-first aspect, in the twenty-seventh or twenty-eighth aspect, the second high-frequency component acquisition step has at least a predetermined low frequency with respect to the first high-frequency component obtained in the first high-frequency component acquisition step. A second high frequency component is obtained by performing predetermined processing including processing for reducing low frequency components exceeding a predetermined threshold among frequency components.

  In a thirty-second invention, in the twenty-seventh or twenty-eighth invention, the second high-frequency component acquisition step has at least a predetermined low frequency relative to the first high-frequency component obtained in the first high-frequency component acquisition step. A second high frequency component is obtained by performing predetermined processing including processing for reducing a low frequency component exceeding a predetermined threshold among frequency components and processing for adjusting the predetermined high frequency component.

  According to a thirty-third aspect, there is provided a computer program for causing a computer to realize the function of the image processing apparatus according to any one of the first to twenty-fifth aspects or the function of the image processing system according to the sixteenth aspect. It is recorded on a readable storage medium.

  A thirty-fourth invention is characterized in that a program for causing a computer to execute the processing steps of the image processing method according to any of the seventeenth to thirty-second inventions is recorded on a computer-readable storage medium.

  A thirty-fifth aspect of the invention is a program for causing a computer to realize the function of the image processing apparatus according to any one of the first to twenty-fifth aspects of the invention or the function of the image processing system according to the sixteenth aspect of the invention. It is characterized by that.

  A thirty-sixth aspect of the invention is a program for causing a computer to execute the processing steps of the image processing method according to any of the seventeenth to thirty-second aspects of the invention.

  As described above, in the present invention, the first high-frequency component is replaced with the second high-frequency component acquired based on the overshoot region in the target image (such as an image obtained by radiography). Specifically, for example, the second high-frequency component is acquired as a high-frequency component that is unlikely to cause overshoot based on the overshoot region, and the first high-frequency component corresponding to the overshoot region is obtained using the second high-frequency component. Is replaced. As a result, it is possible to provide a good processed image (display output, film output image, etc.) of the target image with suppressed overshoot.

  Further, the second high-frequency component is applied only to the overshoot region, the high-frequency component (a high-frequency component obtained by reducing a predetermined low-frequency component from the first high-frequency component, etc.) When configured to do so, overshoot can be suppressed without degrading the restoration of the fine structure in the target image.

Further, when the second high-frequency component is configured to be a component obtained by dynamic range compression of the first high-frequency component, overshoot can be suppressed in a state in which the fine structure is restored in the target image. .
In addition, as a compression of the first high frequency component, when a predetermined low frequency component in the first high frequency component is compressed and a dynamic range compression capable of adjusting the predetermined high frequency component is used, overshoot is suppressed. And a good post-processing image can be obtained.

  In addition, when the processing for obtaining the second high-frequency component and replacing the first high-frequency component is performed only for the overshoot region, an excessive increase in processing time can be avoided.

  Further, the smoothed image used for acquiring the second high-frequency component is acquired by using a filter that hardly causes overshoot, such as an intermediate value filter (intermediate value calculation) or a morphological filter (morphological calculation). In this case, it is possible to obtain a good post-processing image in which overshoot is suppressed.

  Further, the processing for obtaining the second high-frequency component and replacing the first high-frequency component is performed only for the overshoot region, and the smoothed image used for obtaining the second high-frequency component is intermediate. Overshoot suppression without excessively increasing processing time when configured to use a filter that does not easily generate overshoot, such as value filter (intermediate value calculation) or morphological filter (morphology calculation) A good post-processed image can be obtained.

In 1st Embodiment, it is a block diagram which shows the structure of the X-ray imaging apparatus to which this invention is applied. It is a flowchart for demonstrating operation | movement of the said X-ray imaging apparatus. It is a flowchart for demonstrating operation | movement of the said X-ray imaging apparatus in 2nd Embodiment.

[First embodiment]
The present invention is applied to, for example, an X-ray imaging apparatus 100 as shown in FIG.
In particular, the X-ray imaging apparatus 100 of the present embodiment appropriately prevents overshoot and the like by changing the high-frequency component of the target image by the configuration and operation described below.

<Configuration of X-ray imaging apparatus 100>
As shown in FIG. 1, the X-ray imaging apparatus 100 is a two-dimensional X-ray on which an X-ray generation circuit (in FIG. 1) 101 that generates X-rays and X-ray light that has passed through a subject 103 is imaged. A sensor 104, a data acquisition circuit (... in the drawing) 105 that collects captured images output from the two-dimensional X-ray sensor 104, and a preprocessing circuit that performs preprocessing on the captured images collected by the data acquisition circuit 105 (In the drawing, part) 106, a main memory 109 for storing various information such as a captured image after the pre-processing in the pre-processing circuit 106 and a processing program for executing various processing, an instruction for executing X-ray imaging, An operation panel 110 for performing various settings on the apparatus, an image processing circuit 111 for performing image processing on a captured image after preprocessing by the preprocessing circuit 106, and a processing after processing by the image processing circuit 111 Display images Output circuit 118 for outputting power or film, and CPU 108 for controlling the operation of the entire apparatus. A data collection circuit 105, a preprocessing circuit 106, an image processing circuit 111, an output circuit 118, a CPU 108, a main memory 109 and the operation panel 110 are connected to the CPU bus 107 and configured to exchange data with each other.

  The image processing circuit 111 includes a first high-frequency component creation circuit 112 that creates a first high-frequency component from the image that has been pre-processed by the pre-processing circuit 106, and a first high-frequency component creation circuit 112 that obtains the first high-frequency component creation circuit 112. An area detection circuit 113 that detects an area where overshoot occurs from the absolute value of the high-frequency component of the second, and a second high-frequency component generation that generates a second high-frequency component corresponding to the detection area obtained by the area detection circuit 113 Among the first high-frequency components obtained by the circuit 114 and the first high-frequency component creation circuit 112, the second high-frequency component creation circuit 114 obtains a high-frequency component corresponding to the detection region obtained by the region detection circuit 113. The frequency component replacement circuit 115 that replaces the second high frequency component and the high frequency component after replacement obtained by the frequency component replacement circuit 115 are added to the original image, the low frequency image, or the image after gradation conversion. To and a processed image circuit 116 to obtain the processed image.

<Operation of X-ray Imaging Apparatus 100>
The main memory 109 stores data and processing programs necessary for executing various control processes in the CPU 108 in advance, and includes a work memory for working the CPU 108. Here, for example, a processing program according to the flowchart of FIG. 2 is used as a processing program stored in the main memory 109, in particular, a processing program for image processing.
Therefore, the CPU 108 reads out and executes the processing program shown in FIG. 2 from the main memory 109, thereby performing the operation control of the entire apparatus 100 as described below according to the operation from the operation panel 110. Do.

Step S200:
First, the X-ray generation circuit 101 emits an X-ray beam 102 to a subject (inspected object) 103. The X-ray beam 102 emitted from the X-ray generation circuit 101 passes through the object 103 while being attenuated, reaches the two-dimensional X-ray sensor 104, and is output as an X-ray image by the two-dimensional X-ray sensor 104. Is done.
Examples of the X-ray image output from the two-dimensional X-ray sensor 104 include a human body image such as a chest image.

Next, the data acquisition circuit 105 converts the X-ray image output from the two-dimensional X-ray sensor 104 into an electrical signal and supplies it to the preprocessing circuit 106.
The preprocessing circuit 106 performs preprocessing such as offset correction processing and gain correction processing on the signal (X-ray image signal) from the data acquisition circuit 105.
The X-ray image signal preprocessed by the preprocessing circuit 106 is transferred as input image information to the main memory 109 and the image processing circuit 111 via the CPU bus 115 under the control of the CPU 108.

Step S201:
In the image processing circuit 111, a gradation conversion circuit (not shown) uses a gradation conversion function F1 () for the image f (x, y) processed by the preprocessing circuit 106.

Gradation conversion processing is performed according to the following expression (12), and an original image (original image) f0 (x, y) is acquired.
In the above formula (12), “x” and “y” represent the X coordinate and the Y coordinate, respectively.

Step S202:
The first high-frequency component creation circuit 112

As shown by the following equations (13) and (14), the first image f (x, y) obtained in step S201 is subtracted from the low-frequency image f _us (x, y) to obtain the first The high frequency component h1 (x, y) is acquired.

The low-frequency image f _us (x, y) in the equation (13) is created by moving average the original image f (x, y) as shown in the equation (14). “D1” in Expression (14) is simply referred to as a mask size hereinafter.

Step S203:
The region detection circuit 113 detects a region where the absolute value of the high frequency component h1 (x, y) obtained by the first high frequency component creation circuit 112 is equal to or greater than a certain threshold Th, and the detection region (overshoot is detected). Coordinates (x, y) of a region that is likely to occur (hereinafter also simply referred to as “overshoot region”) are stored in the main memory 109.
As described above, the region where the absolute value of the high frequency component h1 (x, y) is equal to or greater than the predetermined threshold Th is detected as the overshoot region. The high frequency component causing overshoot has an absolute value higher than that of the normal high frequency component. This is due to the largeness.

Step S204:
The second high frequency component creation circuit 114 is based on the coordinates of the overshoot area stored in the main memory 109.

The second high frequency component h2 (x, y) is acquired by the following equations (15) and (16).

Here, “d2” in the equation (16) is hereinafter simply referred to as a mask size. The mask size d2 is equal to the mask size d1 in the above equation (14).
d2 <d1
Have the relationship As a result, the second high frequency component h2 (x, y) has a higher frequency component than the first high frequency component h1 (x, y). In addition, due to this relationship, the second high frequency component h2 (x, y) is a component that causes an overshoot less than the first high frequency component h1 (x, y).

Note that if the mask size d2 is set too small, the restoration of the fine structure in the target image may be reduced. Therefore, the mask size d2 is determined in a range in which overshoot hardly occurs and the restoration of the fine structure is not reduced so much. It is desirable to do.
In this case, the recoverability of the medium frequency is deteriorated as compared with the case where the mask size d1 having a normal size is used, but the overshoot can be suppressed without reducing the recoverability of the fine structure.

_{Further, "f us 2 (x,} y)" of the above formula (15) is a low-frequency image (smoothed image) represented by the above formula (16), the low-frequency image f _us 2 (x , Y) Other acquisition methods include, for example, the following methods (1) and (2).

Method (1): Method Using Morphological Filter (Morphological Operation) In this method (1), a morphological filter (morphological operation) represented by the following equations (17) to (20) is used. A low frequency image f _us 2 (x, y) is generated.

  In the above formulas (17) to (20), “D (x, y)” is a disk-like filter represented by the following formula (21), and “r1” is an arbitrary constant, depending on the target image. Is selected.

Method (2): Method Using Intermediate Value Filter In this method (2), the low frequency image f _us 2 (x, y) is generated using the intermediate value filter.
Specifically, the original image f (x, y) within the fixed distance d2 is sorted with the coordinates (x, y) as the center, and the intermediate pixel value is set to _fus2 (x, y).

Low-frequency image f _us 2 (x, y) obtained using the method as described above (1) or method (2) is to store the edge structure. Then, as shown in the above equation (15), the second high frequency component h2 (x, y) obtained using this low frequency image f _us 2 (x, y) is subject to overshoot. The component which becomes is hard to come out. Further, by generating a low-frequency image f _us 2 (x, y) acquired using a morphological filter or an intermediate value filter only for a region where overshoot is likely to occur, the entire target image is generated. The processing time can be shortened as compared with the case where a morphological filter or an intermediate value filter is applied.

Step S205:
The frequency component replacement circuit 115 replaces the first high frequency component h1 (x, y) obtained by the first high frequency component creation circuit 112 in step S202 with

Based on the relational expression (22), the second high-frequency component h2 (x, y) is substituted.
As represented by the relational expression (22), among the first high frequency components h1 (x, y), the components having an absolute value equal to or greater than a certain threshold Th are changed to the second high frequency components h2 (x, y). Replaced.

Step S206:
The post-processing image creation circuit 116

According to the following equation (23), the first high-frequency component h1 (x, y) after replacement obtained by the frequency component replacement circuit 115 is added to, for example, the original image f0 (x, y), thereby performing post-processing. An image fd (x, y) is acquired.

  In step S206, the target image to which the first high-frequency component h1 (x, y) after replacement obtained by the frequency component replacement circuit 115 is added is limited to the original image f0 (x, y). For example, a pre-processed image, a low-frequency image, or an image obtained by gradation conversion of a low-frequency image may be used.

Step S207:
As described above, the processed image fd (x, y) obtained by the image processing circuit 111 is subjected to display output or film output by the output circuit 118.

  As described above, in this embodiment, an overshoot region is detected by analyzing the first high-frequency component acquired from the original image, and overshoot occurs in the first high-frequency component in the overshoot region. Since it is configured to replace the difficult high frequency component (second high frequency component), it is possible to provide a good processed image with suppressed overshoot.

  Further, only the overshoot region is configured so that the first high-frequency component is replaced with a high-frequency component (a second high-frequency component with a smaller mask size d2) due to the higher-frequency region. Overshoot can be suppressed without degrading the restoration of the structure.

  Furthermore, as a method for generating the second high-frequency component, a method using a morphological filter or an intermediate value filter is used, and the processing is performed only on the overshoot region, thereby further increasing the overshoot region. In addition to expecting suppression of shoots, the processing time can be shortened compared to a case where a morphological filter or an intermediate value filter is applied to the entire target image.

[Second Embodiment]
In the present embodiment, the operation of the X-ray imaging apparatus 100 shown in FIG. 1 is, for example, an operation according to the flowchart of FIG.

  In the flowchart of FIG. 3, the same reference numerals are given to the steps for executing the same process as in the flowchart of FIG. 2, and the detailed description thereof will be omitted.

Step S200 to Step S203:
As in the first embodiment, an image obtained by X-ray imaging is processed by the preprocessing circuit 106, and the processed image f (x, y) is transferred to the image processing circuit 111. (Step S200).
The image processing circuit 111 obtains an original image (original image) f0 (x, y) by performing gradation conversion processing on the image f (x, y) (step S201), and generates a first high-frequency component. The circuit 112 acquires the first high-frequency component h1 (x, y) from the original image f (x, y) (step S202), and the region detection circuit 113 uses the first high-frequency component h1 (x, y). An overshoot region is detected (step S203).

  Note that step S203 is not essential, and is necessary when a low-frequency image is created only in an area corresponding to an overshoot area as described later, but is not necessarily required in other cases.

Step S301:
The second high frequency component creation circuit 114 acquires the result of dynamic range compression of the first high frequency component h1 (x, y) as the second high frequency component h2 (x, y).

Although any method can be applied as the above dynamic range compression processing method, in the present embodiment, as an example, the method described in Patent Document 3 described above is used.
Therefore, here, the dynamic range compression processing is performed so that the absolute value of the first high-frequency component h1 (x, y) is equal to or greater than the predetermined threshold Th, that is, the dynamic range only in the overshoot region. Although the compression rate at this time is arbitrary, in the present embodiment, as an example, the maximum compression rate is used. As a result, the first high-frequency component h1 (x, y) equal to or greater than the predetermined threshold Th is compressed as much as possible. For example, the first high-frequency component h1 (x, y) is subjected to dynamic range compression so that a predetermined low-frequency component in the first high-frequency component h1 (x, y) is clipped to be approximately equal to or less than a predetermined threshold Th.

Here, as the acquisition method for the low-frequency image (smoothed image) f _us 3 (x, y) used for dynamic range compression, for example, the following methods (1) to (3) may be mentioned. .

Method (1): Method based on the coordinates of the overshoot area In this method (1), based on the coordinates of the overshoot area stored in the main memory 109,

The low-frequency image _fus 3 (x, y) is acquired by the following expression (24).
“D2” in the equation (24) is a mask size, and with respect to the mask size d1 in the equation (14),
d2 <d1
Have the relationship

Method (2): Method Using Morphological Filter In this method (2), the low-frequency image f _us 3 (x) is obtained by the morphological filter represented by the following equations (25) to (28). , Y).

  In the above formulas (25) to (28), “D (x, y)” is a disk-like filter represented by the following formula (29), and “r1” is an arbitrary constant, depending on the target image. Is selected.

Method (3): Method Using Intermediate Value Filter In this method (3), the low frequency image f _us 3 (x, y) is generated using the intermediate value filter.
Specifically, the original image f (x, y) within the fixed distance d2 is sorted with the coordinate (x, y) as the center, and the pixel value in the middle is set to fus3 (x, y).

The low-frequency image _fus 3 (x, y) obtained by using the method (2) or the method (3) as described above preserves the edge structure, and the low-frequency image _fus 3 (x, y, The second high-frequency component h2 (x, y) obtained by using y) is difficult to produce a component that causes overshoot. Further, by generating a low-frequency image f _us 3 (x, y) acquired using a morphological filter or an intermediate value filter only for a region where overshoot is likely to occur, the entire target image is generated. The processing time can be shortened as compared with the case where a morphological filter or an intermediate value filter is applied.

Step S302:
The post-processing image creation circuit 116

The processed image fd (x, y) is acquired by the following expressions (30) to (32).

  In the above formulas (30) to (32), “h1 (x, y)” is the first high-frequency component (second high-frequency component) after dynamic range compression by the second high-frequency component creation circuit 114. It is. Also, the above formulas (30) and (31) show the sharpening process, and the above formula (32) shows the dynamic range compression process. In the above equation (32), “F2 ()” represents a function for gradation conversion of a low-frequency image.

Step S207:
As described above, the processed image fd (x, y) obtained by the image processing circuit 111 is subjected to display output or film output by the output circuit 118.

  As described above, in the present embodiment, a component obtained by dynamic range compression of the first high frequency component is used as the second high frequency component. That is, since the predetermined low-frequency component in the first high-frequency component corresponding to the region where overshoot is likely to occur is compressed, overshoot can be suppressed without impairing the restoration of the fine structure. In addition, since the first high frequency component is compressed by using a dynamic range compression method capable of compressing a predetermined low frequency component and adjusting the predetermined high frequency component, the first high frequency component can be freely changed. Thus, the second high-frequency component capable of suppressing overshoot without impairing the recoverability of the fine structure can be obtained. Furthermore, overshoot can be effectively suppressed if it is configured to create a low-frequency image used for dynamic range compression using a morphological filter or intermediate value filter, etc. only in areas where overshoot is likely to occur. In addition, the processing time can be shortened as compared with the case where a morphological filter or an intermediate value filter is used for the entire first high-frequency component image.

  Further, by configuring the low-frequency image mask size d2 used for dynamic range compression to be smaller than d1, overshoot can be suppressed.

  Further, the first high-frequency component is replaced with a high-frequency component (second high-frequency component with a smaller mask size d2) in the higher-frequency region in a region where overshoot is likely to occur, thereby providing a fine structure. Overshoot can be suppressed without degrading the resilience.

It is to be noted that an object of the present invention is to supply a storage medium storing software program codes for realizing the functions of the host and terminal according to the first and second embodiments to a system or apparatus, and the computer of the system or apparatus Needless to say, this can also be achieved by (or CPU or MPU) reading and executing the program code stored in the storage medium.
In this case, the program code itself read from the storage medium realizes the functions of the first and second 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 floppy (registered trademark) 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 may be used. it can.
Further, by executing the program code read by the computer, not only the functions of the first and second embodiments are realized, but also an OS running on the computer based on the instruction of the program code. Needless to say, the present invention includes a case where the functions of the first and second 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 the memory provided in the extension function board inserted in the computer or the function extension unit connected to the computer, the function extension is performed based on the instruction of the program code. It goes without saying that the CPU or the like provided in the board or the function expansion unit performs part or all of the actual processing and the functions of the first and second embodiments are realized by the processing.

DESCRIPTION OF SYMBOLS 100 X-ray imaging apparatus 101 X-ray generation circuit 102 X-ray beam 103 Subject 104 Two-dimensional X-ray sensor 105 Data acquisition circuit 106 Pre-processing circuit 107 CPU bus 108 CPU
109 Main Memory 110 Operation Panel 111 Image Processing Circuit 112 First High Frequency Component Creation Circuit 113 Area Detection Circuit 114 Second High Frequency Component Creation Circuit 115 Frequency Component Replacement Circuit 116 Post-Processing Image Creation Circuit 118 Output Circuit

Claims (36)

First high-frequency component acquisition means for acquiring a first high-frequency component from the target image;
Area detection means for detecting an overshoot area of the target image based on the first high frequency component obtained by the first high frequency component acquisition means;
Second high-frequency component acquisition means for acquiring a second high-frequency component based on the overshoot region obtained by the region detection means;
Frequency component replacement means for replacing the first high frequency component obtained by the first high frequency component acquisition means with the second high frequency component obtained by the second high frequency component acquisition means, An image processing apparatus. First high-frequency component acquisition means for acquiring a first high-frequency component from the target image;
Area detection means for detecting an overshoot area of the target image based on the first high frequency component obtained by the first high frequency component acquisition means;
Second high-frequency component acquisition means for acquiring a second high-frequency component based on the overshoot region obtained by the region detection means;
Frequency component replacement means for replacing the first high-frequency component obtained by the first high-frequency component acquisition means with the second high-frequency component obtained by the second high-frequency component acquisition means;
An image processing apparatus comprising: a post-processing image acquisition unit that acquires a target image for output based on the high frequency component replaced by the frequency component replacement unit.   The post-processing image acquisition unit acquires a result obtained by performing predetermined conversion on the high-frequency component replaced by the frequency component replacement unit and adding the result to the target image as the target image for output. 2. The image processing apparatus according to 2.   The target image includes at least one of an input image, an image after gradation conversion of the input image, a low-frequency image obtained from the input image, and an image after gradation conversion of the low-frequency image. The image processing apparatus according to claim 3.   3. The image processing apparatus according to claim 1, wherein the second high-frequency component acquisition unit acquires a high-frequency component according to a higher frequency region than the first high-frequency component as the second high-frequency component. .   3. The second high-frequency component acquisition means acquires the second high-frequency component by compressing at least a predetermined low-frequency component included in the first high-frequency component. Image processing apparatus.   7. The image processing apparatus according to claim 6, wherein the second high frequency component acquisition means compresses the first high frequency component by dynamic range.   The second high-frequency component acquisition means acquires the second high-frequency component using a smoothed image obtained using at least one of an intermediate value filter and a morphological filter. The image processing apparatus according to claim 1 or 2. The second high-frequency component acquisition means acquires the second high-frequency component only for the overshoot region obtained by the region detection means,
The frequency component replacement means replaces the first high frequency component obtained by the first high frequency component acquisition means corresponding to the overshoot region with the second high frequency component obtained by the second high frequency component acquisition means. The image processing apparatus according to claim 1, wherein the image processing apparatus is replaced with First high-frequency component acquisition means for acquiring a first high-frequency component from the target image;
A second high-frequency component is obtained by performing a predetermined process including a process of reducing at least the predetermined low-frequency component on the first high-frequency component obtained by the first high-frequency component acquisition means. An image processing apparatus comprising: a high-frequency component acquisition unit. First high-frequency component acquisition means for acquiring a first high-frequency component from the target image;
A second high-frequency component is obtained by performing a predetermined process including a process of reducing at least the predetermined low-frequency component on the first high-frequency component obtained by the first high-frequency component acquisition means. High-frequency component acquisition means;
An image processing apparatus comprising: a post-processing image acquisition unit that acquires a target image for output based on the high frequency component obtained by the second high frequency component acquisition unit.   The post-processing image acquisition unit acquires a result obtained by performing a predetermined conversion on the high-frequency component replaced by the second high-frequency component acquisition unit and adding the result to the target image as the target image for output. The image processing apparatus according to claim 11.   The target image includes at least one of an input image, an image after gradation conversion of the input image, a low-frequency image obtained from the input image, and an image after gradation conversion of the low-frequency image. The image processing apparatus according to claim 12.   The second high-frequency component acquisition unit is configured to reduce a low-frequency component exceeding a predetermined threshold among at least the predetermined low-frequency component with respect to the first high-frequency component obtained by the first high-frequency component acquisition unit. The image processing apparatus according to claim 10, wherein the second high-frequency component is obtained by performing a predetermined process including:   The second high-frequency component acquisition unit is configured to reduce a low-frequency component exceeding a predetermined threshold among at least the predetermined low-frequency component with respect to the first high-frequency component obtained by the first high-frequency component acquisition unit. 12. The image processing apparatus according to claim 10, wherein the second high-frequency component is obtained by performing predetermined processing including a step of adjusting the predetermined high-frequency component. An image processing system in which a plurality of devices are communicably connected to each other,
The image processing system according to claim 1, wherein at least one of the plurality of devices has the function of the image processing apparatus according to claim 1. A first high-frequency component acquisition step of acquiring a first high-frequency component from the target image;
An area detection step for detecting an overshoot area of the target image based on the first high frequency component obtained by the first high frequency component acquisition step;
A second high-frequency component acquisition step for acquiring a second high-frequency component based on the overshoot region obtained by the region detection step;
A frequency component replacement step of replacing the first high-frequency component obtained in the first high-frequency component acquisition step with the second high-frequency component obtained in the second high-frequency component acquisition step. Image processing method.   18. The image processing method according to claim 17, further comprising a post-processing image acquisition step of acquiring a target image for output based on the high frequency component replaced by the frequency component replacement step. A first high-frequency component creating step of creating a first high-frequency component of the target image;
A region detecting step of analyzing the high frequency component created by the first high frequency component creating step to detect an overshoot region of the target image;
A second high-frequency component creating step for creating a second high-frequency component of the target image;
The first high-frequency component created by the first high-frequency component creation step corresponding to the overshoot region detected by the region detection step is used as the second high-frequency component created by the second high-frequency component creation step. And a frequency component replacement step for replacing the frequency component. A first high-frequency component creating step of creating a first high-frequency component of the target image;
A region detecting step of analyzing the high frequency component created by the first high frequency component creating step to detect an overshoot region of the target image;
A second high-frequency component creating step for creating a second high-frequency component of the target image;
The first high-frequency component created by the first high-frequency component creation step corresponding to the overshoot region detected by the region detection step is used as the second high-frequency component created by the second high-frequency component creation step. A frequency component replacement step to replace with,
And a post-processing image acquisition step of acquiring a post-processing image by performing predetermined conversion on the high-frequency component replaced in the frequency component replacement step and adding it to the target image.   The target image includes at least one of an input image, an image after gradation conversion of the input image, a low-frequency image obtained from the input image, and an image after gradation conversion of the low-frequency image. The image processing method according to claim 19 or 20.   The second high frequency component creating step includes a step of making the second high frequency component a high frequency component depending on a higher frequency region than the first high frequency component created by the first high frequency component creating step. The image processing method according to claim 19 or 20, wherein the image processing method is characterized in that:   The second high-frequency component creation step includes a step of making the second high-frequency component a high-frequency component obtained by dynamic range compression of the first high-frequency component created in the first high-frequency component creation step. The image processing method according to claim 19 or 20.   24. The image according to claim 23, wherein the second high-frequency component creating step includes a step of creating a smoothed image used when the dynamic range compression is performed by an intermediate value filter or a morphological filter. Processing method.   21. The step of creating the second high-frequency component includes a step of creating the second high-frequency component from a smoothed image obtained by an intermediate value filter or a morphological filter. The image processing method as described.   26. The step of creating the second high-frequency component includes the step of creating the second high-frequency component only for the overshoot region detected by the region detection step. The image processing method as described. A first high-frequency component acquisition step of acquiring a first high-frequency component from the target image;
A second high-frequency component is obtained by performing a predetermined process including a process of reducing at least the predetermined low-frequency component on the first high-frequency component obtained in the first high-frequency component acquisition step. An image processing method comprising: a high-frequency component acquisition step. A first high-frequency component acquisition step of acquiring a first high-frequency component from the target image;
A second high-frequency component is obtained by performing a predetermined process including a process of reducing at least the predetermined low-frequency component on the first high-frequency component obtained in the first high-frequency component acquisition step. A high-frequency component acquisition step;
An image processing method comprising: a post-processing image acquisition step of acquiring a target image for output based on the high frequency component obtained by the second high frequency component acquisition step.   The post-processing image acquisition step is characterized in that a result obtained by performing a predetermined conversion on the high-frequency component acquired in the second high-frequency component acquisition step and adding the result to the target image is acquired as the target image for output. The image processing method according to claim 28.   The target image includes at least one of an input image, an image after gradation conversion of the input image, a low-frequency image obtained from the input image, and an image after gradation conversion of the low-frequency image. 30. The image processing method according to claim 29.   The second high frequency component acquisition step is a process of reducing a low frequency component exceeding a predetermined threshold among at least the predetermined low frequency component with respect to the first high frequency component obtained in the first high frequency component acquisition step. 29. The image processing method according to claim 27 or 28, wherein the second high-frequency component is obtained by performing a predetermined process including:   The second high frequency component acquisition step is a process of reducing a low frequency component exceeding a predetermined threshold among at least the predetermined low frequency component with respect to the first high frequency component obtained in the first high frequency component acquisition step. 29. The image processing method according to claim 27 or 28, wherein a second high-frequency component is obtained by performing a predetermined process including a process for adjusting the predetermined high-frequency component.   A computer-readable storage medium storing a program for causing a computer to realize the function of the image processing apparatus according to any one of claims 1 to 15 or the function of the image processing system according to claim 16.   A computer-readable storage medium storing a program for causing a computer to execute the processing steps of the image processing method according to any one of claims 17 to 32.   The program for making a computer implement | achieve the function of the image processing apparatus of any one of Claims 1-15, or the function of the image processing system of Claim 16.   A program for causing a computer to execute the processing steps of the image processing method according to any one of claims 17 to 32.

Images