JP2001086409A - Device and method for image processing - Google Patents

Abstract

PROBLEM TO BE SOLVED: To output an image with stable density and/or contrast by deciding an image processing condition so as to reproduce a signal value obtained by analyzing an image signal in a recognized prescribed area in constant density and/or contrast, deciding a correction condition on the basis of an index extracted on the basis of the signal distribution of the image signal and performing image processing. SOLUTION: An area recognizing means 23a recognizes a prescribed area corresponding to the prescribed structure of a subject. An image processing condition deciding means 23b analyzes an image signal in the recognized area and decides an image processing condition for reproducing the image signal into fixed density and/or contrast which are preliminarily defined. An individual difference extracting means 23c extracts an index of the individual difference of the subject on the basis of the signal distribution of the image signal in the area, and a correction condition deciding means 23d decides a correction condition for the image processing condition decided by the means 23b on the basis of the index. An output data storing means 25 stores image data that has been subjected to image processing by an image processor 23.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to an image processing apparatus and an image processing method, and more particularly to an image processing apparatus and an image processing method for a radiation image.

[0002]

2. Description of the Related Art In recent years, a stimulable phosphor or FPD (Flat Panel Detect) has been used without using a silver halide film.
A radiation image generation method for directly extracting a radiation image as a digital signal from a radiation detector such as tor) has been used. In addition, various image processings are being performed for the purpose of making the radiation image obtained by the radiation image generation method more suitable for diagnosis.

FIG. 8 is a conceptual diagram of the configuration of a radiographic image capturing apparatus. In FIG. 8, radiation emitted from a radiation generator 50 passes through a subject 5 and irradiates an imaging panel mounted on the front of a radiation image reader 40. The radiation image reader 40 reads an image recorded on the panel, and the radiation image processing device 41 further performs predetermined image processing.

Japanese Patent Application Laid-Open Nos. 55-116340 and 2-276484 disclose an image of an anatomical human body that is important for diagnosis because of image processing for the purpose of finishing gradation suitable for diagnosis. A technique is described in which signal values based on a predetermined structure are obtained, and gradation processing for outputting the signal values at a predetermined density is performed.

[0005] For example, FIG. 9 is an explanatory diagram of gradation processing according to a conventional technique. FIG. 9A shows a read image and FIG.
FIG. 9B shows a histogram obtained by analyzing the image of FIG. The histogram H101 showing the signal distribution of the image F101 indicates the anatomical human body structure A.
(10) As signal values based on A (20), a high signal value S (101,10) corresponding to the human body structure A (10) and a low signal value S (101) corresponding to the human body structure A (20). ,
20) is detected. This high signal value S (101, 1
0) and the low signal value S (101, 20) as a reference, gradation processing is performed on the signal value S for each pixel in accordance with certain image processing conditions. Further, when performing dynamic range compression described in JP-A-2-292679, processing is performed using a fixed correction coefficient, and when performing frequency processing described in Japanese Patent Publication No. 62-62373, processing is performed using a fixed enhancement coefficient. .

[0006]

However, the amount of radiation transmitted through the subject 5 and the state of scattered radiation vary depending on the patient's physique, the presence or absence of casting, and the position (positioning) at the time of imaging. When image processing is performed under certain image processing conditions as described above, the following problem occurs.

For example, when frequency processing is to be performed with a predetermined degree of emphasis for each part and body position at the time of imaging, for example, the trunk of a patient K2 who is fat and thick, or a patient who wears a cast, FIG. 10 shows an image of a limb bone.
As shown in (B), a decrease in contrast occurs due to the effects of scattered radiation, image enlargement, and the like, and the patient K1 is thin and thin.
In contrast, as shown in FIG. 10A, the influence of scattered radiation, enlargement of an image, and the like is reduced, and the contrast is hardly reduced.

When the dynamic range compression is to be performed with the shape and the strength of the function of the enhancement coefficient being constant, for example, the bones near the side of the body of a thin patient K1 are compared with those of a patient with a thicker meat. In comparison, since the amount of X-ray transmission changes rapidly, the bone signal distribution extends over a wide range, making it difficult to keep the density range appropriate for diagnosis.
A problem arises in that a stable density cannot be expected after image processing.

In the case where gradation processing for adjusting the contrast is to be performed so that both the high signal and the low signal reference signal detected by analyzing the signal distribution of the image signal have a constant density, For example, in the case of a radiation image of a patient K2 whose subject is fat, if gradation processing for automatically adjusting the contrast is performed so that the signal values S (1) and S (2) have a constant density, the contrast is extremely increased. Too, signal value S
If a certain density is set based on one of (1) and S (2), the contrast may be too low.

Accordingly, the present invention provides a stable density and stable image output when imaging an image of a human body structure important for diagnosis, regardless of the patient's physique, positioning, or presence or absence of a device such as a cast. An object of the present invention is to provide an image processing apparatus and an image processing method capable of outputting in contrast.

[0011]

SUMMARY OF THE INVENTION The object of the present invention is as set forth in claim 1.
The problem can be solved by the image processing device described in (1). That is, the image processing apparatus according to the first aspect is an image processing apparatus that processes a radiation image generated based on radiation transmitted through a subject, and recognizes a specific area corresponding to a predetermined structure of the subject. Region recognition means, and image processing condition determination means for determining image processing conditions so that a signal value obtained by analyzing the image signal of the area is reproduced at a predetermined constant density and / or contrast, An individual difference extracting unit that extracts an index of an individual difference of the subject based on a signal distribution of the image signal in the region; and a correction condition of the image processing condition determined by the image processing condition determining unit based on the index. And an image processing apparatus for performing image processing on the radiation image based on image processing conditions corrected by the correction conditions.

According to the image processing apparatus of the first aspect,
Determine the correction conditions of the image processing conditions based on the index of the individual difference of the subject, and use the determined correction conditions, regardless of the patient's physique, positioning, presence or absence of equipment such as casting, etc. It is possible to output an image with a stable density and / or contrast when outputting an image obtained by shooting.

In this image processing apparatus, the image processing condition determining means detects a reference signal value corresponding to a predetermined structure of a subject based on a signal distribution of the image signal. Preferably, the correction condition is determined based on a reference signal value.

According to this image processing apparatus, since the correction conditions are determined, it is possible to more accurately reflect the individual difference of the subject and to achieve stable density and contrast.

Further, these image processing devices may detect a difference between the reference signal values as the index.

According to this image processing apparatus, the correction condition can be determined according to the degree of individual difference based on the difference between the reference signal values.

In these image processing apparatuses, it is preferable that the image processing is a gradation processing for the radiographic image, and the correction condition is a density and / or a contrast of the gradation processing.

In this image processing apparatus, the degree of correction varies depending on the size of the physique and the wearing object such as a cast, so that it is possible to prevent the contrast from being too high depending on the condition of the patient and to output a stable image. Becomes

In these image processing apparatuses, it is preferable that the image processing is frequency processing on the radiation image, and the correction condition is a value of an enhancement coefficient used in the frequency processing.

According to this image processing apparatus, since the degree of correction varies depending on the degree of the physique and the wearing object such as a cast, the contrast is prevented from being lowered depending on the condition of the patient, and the stable image output is achieved. Becomes possible.

In these image processing apparatuses, it is preferable that the image processing is a dynamic range compression processing for the radiation image, and the correction condition is a correction coefficient used for the dynamic range compression processing.

According to this image processing apparatus, since the degree of correction changes depending on the degree of the physique and the wearing object such as a cast, the degree of the dynamic range compression processing is changed depending on the condition of the patient to obtain a stable density. Can be output.

The object of the present invention can be solved by an image processing apparatus according to the present invention. That is, the image processing apparatus according to claim 7 is an image processing apparatus that processes a radiation image generated based on radiation transmitted through a subject,
An image processing condition for analyzing an image signal, recognizing a signal value corresponding to a predetermined structure of a subject, and determining an image processing condition so that the signal value is reproduced at a predetermined constant density and / or contrast. Determining means, an individual difference extracting means for extracting an index of an individual difference of the subject based on a signal value corresponding to a predetermined structure of the subject, and an image determined by the image processing condition determining means based on the index. A correction condition determining means for determining a correction condition of the processing condition, and performing image processing on the radiation image based on the image processing condition corrected by the correction condition.

According to the image processing apparatus of the seventh aspect,
Determine the correction conditions of the image processing conditions based on the index of the individual difference of the subject, and use the determined correction conditions, regardless of the patient's physique, positioning, presence or absence of equipment such as casting, etc. It is possible to output an image with a stable density and / or contrast when outputting an image obtained by shooting.

In this image processing apparatus, the image processing condition determining means detects a reference signal value corresponding to a predetermined structure of a subject based on a signal distribution of the image signal. Preferably, the correction condition is determined based on a reference signal value.

According to this image processing apparatus, since the correction condition is determined, individual differences of the subject can be more accurately reflected, and stable density and contrast can be obtained.

The image processing apparatus may detect a difference between the reference signal values as the index.

According to this image processing apparatus, the correction condition can be determined according to the degree of individual difference based on the difference between the reference signal values.

In this image processing apparatus, it is preferable that the image processing is a gradation processing for the radiation image, and the correction condition is a density and / or a contrast of the gradation processing.

According to this image processing apparatus, the degree of correction varies depending on the degree of the physique and the wearing object such as a cast, so that it is possible to prevent the contrast from being too high depending on the condition of the patient and to output a stable image. Becomes possible.

In this image processing apparatus, the image processing is frequency processing for the radiation image, and the correction condition is a value of an enhancement coefficient used for the frequency processing.

According to this image processing apparatus, the degree of correction varies depending on the degree of the physique and the wearing object such as a cast, so that the contrast is prevented from lowering depending on the condition of the patient, and a stable image output is achieved. Becomes possible.

In this image processing apparatus, the image processing is a dynamic range compression processing for the radiation image, and the correction condition is a correction coefficient used for the dynamic range compression processing.

According to this image processing apparatus, since the degree of correction changes depending on the degree of the physique and the wearing such as a cast, the degree of the dynamic range compression processing is changed depending on the condition of the patient to obtain a stable density. Can be output.

The object of the present invention can be solved by an image processing method according to the present invention. 14. The image processing method according to claim 13, wherein the image processing method is a method of processing a radiation image generated based on radiation transmitted through a subject, and recognizes a specific region corresponding to a predetermined structure of the subject. An image processing condition is determined so that a signal value obtained by analyzing the image signal of the area is reproduced at a predetermined constant density and / or contrast, and based on a signal distribution of the image signal in the area. Extracting an index of the individual difference of the subject, determining a correction condition of the image processing condition based on the index, and performing image processing on the radiation image based on the image processing condition corrected by the correction condition. It is characterized by.

According to the image processing method of the present invention, the correction condition of the image processing condition is determined based on the index of the individual difference of the subject, and the patient's physique, positioning, cast, etc. are determined by the determined correction condition. Regardless of the presence or absence of the instrument, it is possible to output in a stable tone when outputting an image obtained by imaging a human body structure important for diagnosis.

In this image processing method, a reference signal value corresponding to a predetermined structure of a subject is detected based on a signal distribution of the image signal, and the correction condition is determined based on the detected reference signal value. .

According to this image processing apparatus, since the correction condition is determined, the individual difference of the subject can be more accurately reflected, and the density and contrast can be stabilized.

In these image processing methods, a difference between the reference signal values may be detected as the index.

According to this image processing apparatus, the correction condition can be determined according to the degree of individual difference based on the difference between the reference signal values.

In these image processing methods, it is preferable that the image processing is gradation processing for the radiation image, and the correction condition is a density and / or contrast of the gradation processing.

In this image processing method, the degree of correction changes according to the degree of physique, so that it is possible to prevent the contrast from being too high depending on the patient's physique and to output a stable image. .

In these image processing apparatuses, it is preferable that the image processing is frequency processing on the radiation image, and the correction condition is a value of an enhancement coefficient used in the frequency processing.

According to this image processing apparatus, since the degree of correction changes according to the degree of physique, the contrast is prevented from being lowered by the patient's physique, and a stable image can be output. .

In these image processing apparatuses, the image processing is a dynamic range compression process for the radiation image, and the correction condition is a correction coefficient used for the dynamic range compression process.

According to this image processing apparatus, the degree of correction changes in accordance with the degree of physique, so that the degree of dynamic range compression processing is changed depending on the patient's body type, so that an image with a stable density can be output. It becomes possible.

The object of the present invention can be solved by an image processing method according to claim 19. 20. The image processing method according to claim 19, wherein the image processing method is a method of processing a radiation image generated based on radiation transmitted through an object, wherein the image signal is analyzed and a signal value corresponding to a predetermined structure of the object is analyzed. And determines image processing conditions so that this signal value is reproduced at a predetermined constant density and / or contrast. Based on the distribution of signal values corresponding to a predetermined structure of the subject, Extracting an index of the individual difference, determining a correction condition of the image processing condition based on the index, and performing image processing on the radiation image based on the image processing condition corrected by the correction condition. I do.

According to the image processing method of the present invention, the correction condition of the image processing condition is determined based on the index of the individual difference of the subject, and the patient's physique, positioning, cast, etc. are determined by the determined correction condition. Regardless of the presence or absence of the instrument, it is possible to output in a stable tone when outputting an image obtained by imaging a human body structure important for diagnosis.

In this image processing method, it is preferable that a reference signal value corresponding to a predetermined structure of a subject is detected based on a signal distribution of the image signal, and the correction condition is determined based on the detected reference signal value. .

According to this image processing apparatus, since the correction condition is determined, the individual difference of the subject is more accurately reflected,
Stable density and contrast can be obtained.

In these image processing methods, it is preferable to detect a difference between the reference signals as the index.

According to this image processing apparatus, it is possible to determine the correction condition according to the degree of individual difference based on the difference between the reference signal values.

In these image processing methods, it is preferable that the image processing is a gradation processing for the radiation image, and the correction condition is a density and / or a contrast of the gradation processing.

In this image processing method, the degree of correction changes in accordance with the degree of the physique, so that it is possible to prevent the contrast from being too high depending on the patient's physique and to output a stable image. .

In these image processing apparatuses, it is preferable that the image processing is frequency processing on the radiation image, and the correction condition is a value of an enhancement coefficient used in the frequency processing.

According to this image processing apparatus, since the degree of correction changes according to the degree of physique, the contrast is prevented from being lowered by the patient's physique, and a stable image can be output. .

In these image processing apparatuses, it is preferable that the image processing is a dynamic range compression processing for the radiation image, and the correction condition is a correction coefficient used for the dynamic range compression processing.

According to this image processing apparatus, the degree of correction changes in accordance with the degree of physique, so that the degree of dynamic range compression processing is changed according to the patient's body type, so that an image with a stable density can be output. It becomes possible.

[0059]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a first embodiment of the present invention will be described. FIG. 1 is a block diagram showing a hardware configuration of an example of a system using the image processing device of the present invention. In this embodiment, a case is shown in which the present invention is applied to radiation imaging of a human body for medical use.

The radiation source 1 is controlled by a radiation controller 2 to emit radiation (generally X-rays) toward a subject (such as a human chest) M. The radiation image reader 3 is provided with a radiation image conversion panel 4 having a stimulable phosphor at a position facing the radiation source 1 across the subject M.

The radiation image conversion panel 4 accumulates energy in the built-in stimulable phosphor in accordance with the radiation transmittance distribution of the subject M with respect to the amount of irradiation radiation from the radiation source 1, and forms a latent image of the subject, ie, a radiation image. To form The radiation image conversion panel 4 includes a stimulable phosphor layer on a support,
It is provided by vapor phase deposition of a stimulable phosphor or application of a stimulable phosphor. Examples of the material for the stimulable phosphor include, for example, JP-A-61-72091 and JP-A-59-7
Materials such as those disclosed in JP-A-5200 are used.

The laser beam from the laser beam generator 5 shines on the radiation image conversion panel 4 via a scanner (for example, an optical deflector such as a polygon mirror) 6 via various optical systems and a reflecting mirror 7. It is guided as scanning light (excitation light) for exhaust excitation. The light guide 8 converts photostimulated light emitted from the radiation image conversion panel 4 scanned by the laser beam into a photomultiplier tube (hereinafter, also referred to as a photomultiplier).
Light is guided to 10.

The guided light is photoelectrically converted by the photomultiplier 10 as a photoelectric conversion means into a current signal corresponding to the incident light (the transmitted dose). The output current from the photomultiplier 10 is converted into a voltage signal by the current / voltage converter 11, amplified by the amplifier 12, and then converted into digital signal image data by the A / D converter 13. The A / D converter 1
The image data output from 3 is a digital signal of 12 bits (0 to 4095) in the present embodiment.

Therefore, the radiation image reader 3 causes the radiation image information stored and recorded in the stimulable phosphor to be stimulated by scanning the radiation image conversion panel 4 with the excitation light, and the stimulated emission is performed. This is a means for obtaining image data by photoelectrically reading. Note that reading means other than the above-described one using a laser beam may be used.

It should be noted that digital image data can be obtained by using an FPD (flat panel detector) instead of the radiation image reader 3 using a stimulable phosphor.

The data output from the radiation image reader 3 is output to the image processing device 23 in the radiation image display 20, where it is subjected to appropriate image processing and displayed. The radiation image display 20 includes an image data storage unit 21, an image processing device 23, an output data storage unit 25, and an output unit 2.
6. These means 21 to 26 are controlled by control means (not shown).

The image processing device 23 may be inside or outside the radiation image display 20.

The image data storage means 21 is a storage means for storing the image data read by the radiation image reader 3. The image data stored in the image data storage means 21 is subjected to image processing by the image processing device 23, converted into density and gradation suitable for diagnosis, and outputted to the output means 26.
The image is displayed on. To store a plurality of image data, a plurality of image data storage means 21 may be provided, an image data storage means capable of storing a plurality of image data may be used, or, of course, these may be combined. .

The image processing apparatus 23 further includes an area recognizing unit 23a, an image processing condition determining unit 23b, an individual difference extracting unit 23c, and a correction condition determining unit 23d.
The area recognizing means 23a is means for recognizing a specific area corresponding to a predetermined structure of the subject, the image processing condition determining means 23b
Means for analyzing an image signal of a recognized area to determine image processing conditions for reproducing a predetermined constant density and / or contrast; individual difference extracting means 2
3c is based on the signal distribution of the image signal in the area,
The means for extracting an index of the individual difference of the subject and the correction condition determining means 23d are means for determining a correction condition of the image processing condition determined by the image processing condition determining means 23b based on the index. The details will be described later.

The output data storage means 25 is a storage means for storing image data processed by the image processing device 23, that is, display image data.

The output means 26 includes an output data storage means 25
C that displays an image based on the data stored in
An RT display, a printer capable of outputting to film, a recording device capable of recording on a recording medium such as a tape disk, and the like.

Here, the image processing apparatus of the present embodiment executes the dynamic range compression processing, the frequency processing, and the gradation processing on the image data. It will be described using FIG.
FIG. 2 is a conceptual diagram illustrating the concept of the dynamic range compression processing.

The dynamic range compression processing (equalization processing) compresses the dynamic range of an image signal based on the unsharp image signal Sus, so that even an image having a wide dynamic range is expressed in a density range in which the entire image can be easily viewed. is there. The dynamic range compression processing is represented by the following equation.

S = Sorg−β (Sus−A) (1) where β = βL (Sus ≦ A), βH (Sus> A) S: Display image data (processing result) Sorg: Image data (Original image signal) Sus: Unsharp image signal β: Correction coefficient A: Correction boundary signal By executing the dynamic range compression processing, the human body structures A (1) and A (2) which are important for diagnosis have low density area and high density area. This is an image processing that is particularly effective when both are included.

The frequency processing (edge enhancement processing) adjusts the spatial frequency characteristics of the image to output the human body structure A (n) more sharply. The frequency processing is represented by the following arithmetic expression.

S = Sorg + β (Sorg−Sus) (2) where β = βL (Sus ≦ A), βH (Sus> A) S: display image data (processing result) Sorg: image data (original image signal) Sus : Unsharp image signal β: Enhancement coefficient By executing this image processing, image processing is particularly effective in that trabecular bones, fractures, outlines of organs, and the like can be clearly described.

The gradation processing is an image processing that is particularly effective for determining an optimum gradation processing condition for each image, performing image processing, and outputting an image having an appropriate gradation. The image processing apparatus according to the present embodiment performs the following automatic gradation processing. First, the obtained image data is analyzed to obtain a region of interest (ROI).
And the reference signal value are detected, and the gradation conversion table (LU
T) and the reference output density are set, the gradation processing conditions are determined, and gradation processing is performed. Next, the gradation processing will be described with reference to a conceptual diagram illustrating the detection of the region of interest in FIG. 3 and FIG.

First, detection of the ROI will be described. RO
Upon detecting I, a profile corresponding to the image data is obtained. The profile represents the change of the pixel value on an arbitrary line segment by taking the position on the line segment on the horizontal axis and the pixel value on the vertical axis. And the horizontal profile P1
And scanning the vertical profile P2 to find the minimum value, maximum value, inflection point, intersection with the threshold value, and the like, and calculate the ROI based on these.

When the ROI is obtained, the reference signal value S is determined by the image processing condition determining means 23b as follows.

FIG. 4 is a graph showing an example of a histogram obtained from image data. Image processing condition determining means 23
b performs statistical processing on the image data of the ROI (in this example, the chest region 512 in FIG. 3) to obtain a histogram H1 which is a frequency distribution of the image data. Then, the frequency distribution is analyzed to detect the reference signal value S (m, n). In FIG. 4, a plurality of reference signal values S (m, n) are used as reference signal values S (m, n).
An example in which (1,1) and S (1,2) are detected is shown.
The analysis of the frequency distribution can be realized by a method using differential processing, a method using integration processing, or another method.

In the following description, the reference signal value S (m,
In the notation of n), with respect to the numerical value in parentheses, m is an image for each photographing, that is, a division of the subject Km, and a signal value corresponding to the image F1 obtained by photographing the subject K1 is a reference signal value S
(1, n), a signal value corresponding to the image F2 obtained by photographing the subject K2 is represented as a representative signal value S (2, n). Also, n is a classification of the human body structure, and the human body structure A
The signal value corresponding to (1) is the representative signal value S (m, 1), and the signal value corresponding to the human body structure A (2) is the representative signal value S.
(M, 2).

Recognition of ROI and reference signal value S
Is detected, the histogram is normalized based on this signal value, and gradation processing is performed through a gradation conversion table. As the image processing, individual difference extracting means 23 for applying this gradation processing, further frequency processing, and dynamic range compression
c, an example of the correction condition determining means 23d for the image processing condition is shown below.

An example of application to frequency processing, for example, in the case of photographing a trunk, will be described. FIG. 5 shows the subject K
5A is an example of a histogram according to m, FIG. 5A shows a histogram H2 corresponding to a thin subject K2, and FIG.
Shows an example of the histogram H3 corresponding to the thick subject K3.

For the image signal, the reference signal value S (m, 1) of the low signal region indicating the bone portion and the reference signal value of the high signal region indicating the soft portion are obtained by signal value analysis such as histogram analysis as described above. Find S (m, 2).

Next, from the plurality of reference signal values Sa, a reference amount Sa representing a feature of a subject (patient), which is an example of the index of the present invention, is determined by the individual difference extracting means 23c as follows. First, two reference signal values S (m, 1), S
X = S using the function z = f (x, y) from (m, 2)
(M, 1), y = S (m, 2), and a reference amount Sa representing the characteristics of the patient is calculated as in equation (2).

Sa = f (S (m, 1), S (m, 2)) (2) This function z = f (x, y) is, for example, a histogram width of a signal distribution when expressed by Expression (3). Represents This is because the fater the patient, the smaller the difference between the signal values of the bone and the soft part, so that Sa becomes smaller.

F (x, y) = (xy) (3) Therefore, this Sa serves as a guideline indicating the characteristic of the patient's physique such as fat or thin. Sa makes it possible to represent individual differences among patients, such as differences in body thickness and the presence or absence of wearing objects such as casts. 5 (A) and 5 (B), reference amounts Sa2 and Sa3 are shown by a histogram H2 with a thin patient as a subject and a histogram H3 with a fat patient as a subject, respectively.

Next, based on the reference amount Sa obtained in this manner, the emphasis count of the frequency processing is corrected. The sharper the patient is, the lower the sharpness becomes due to the influence of scattered rays and enlargement of a geometric image, etc. Therefore, β2 in which the enhancement coefficient β1 of the frequency processing is corrected according to Sa as shown in equation (4) Ask for. The greater the β2, the stronger the frequency emphasis is applied.

Β2 = β1 × k / Sa (4) where k is a constant The calculation of the equation (4) is a setting of the correction condition of the image processing condition of the present invention, and this calculation is executed by the correction condition determining means 23d. You.

According to the equation (4), as Sa representing the histogram width becomes smaller, the emphasis coefficient β2 becomes larger, and it is possible to compensate for the reduced sharpness. Therefore, sharpness can be compensated for in a fat patient in which Sa tends to be small. Moreover, since Sa becomes smaller as the patient becomes thicker, the degree of correction changes following the degree of physique, and the output becomes stable.

In addition to the frequency processing, an example of application to a dynamic range compression processing in which an entire image having a wide dynamic range is included in a density range that is easy to view without lowering the contrast of a fine structure portion of a subject will be described.

In the same manner as in the application to the frequency processing, the reference signal value S (m, 1) in the low signal area and the reference signal value S (m, 2) in the high signal area are obtained, and the reference amount Sa is further obtained. .

The correction for the correction coefficient of the dynamic range compression processing is performed based on the reference amount Sa thus obtained.

For example, bones near the side of the body of a thin patient such as the pelvis and the side of the lumbar vertebrae have an abrupt change in the amount of X-ray transmission as compared with a patient with a thicker flesh, so that the signal distribution of the bone covers a wide concentration range. .

Therefore, β4 obtained by correcting the correction coefficient β3 of the dynamic range compression in accordance with the reference amount Sa is obtained as in equation (5). The larger the β, the stronger the correction by the dynamic range compression.

Β4 = β3 × k × Sa (5) where k is a constant The calculation of the equation (5) is a setting of the correction condition of the image processing condition of the present invention, and this calculation is executed by the correction condition determining means 23d. You.

According to the equation (5), the correction coefficient β4 increases as the Sa representing the histogram width increases, and it becomes possible to compress the dynamic range with a higher intensity for a wide signal distribution. Therefore, the dynamic range compression processing can be performed with a higher strength for a leaner patient who tends to have a larger Sa. Moreover, since Sa becomes larger as the patient becomes thinner, the degree of correction changes following the degree of physique, and the output becomes stable.

Another example of application to gradation processing in chest imaging and the like will be described. In the same way as the application example to frequency processing,
The reference signal value S (m, 1) in the low signal region and the reference signal value S (m, 2) in the high signal region are obtained, and further, the reference amount Sa is obtained.

The gradation processing parameter is corrected based on the reference amount Sa thus obtained. Reference signal value S (m, 1) in low signal region, reference signal value S in high signal region
When gradation processing is performed so that the density of this portion is constant with reference to (m, 2), in the case of a patient with a narrow histogram (a small reference amount Sa), such as a fat patient, this floor is used. The contrast is too high due to the tone processing. Therefore, when Sa becomes equal to or less than the fixed value k, the contrast is limited as in the equations (6-1) and (6-2).

When k> Sa S (m, 1) '= S (m, 1) + (k-Sa) × (1-a) (6-1) S (m, 2)' = S (m , 2) − (k−Sa) × a (6-2) where 0 <a <1 The calculation of Expressions (6-1) and (6-2) is the setting of the correction condition of the image processing condition of the present invention. This calculation is executed by the correction condition determining means 23d.

According to the equations (6-1) and (6-2), Sa becomes smaller as the patient becomes thicker, so that the degree of correction changes in accordance with the degree of physique, and the patient's physique Depending on the situation, the output may be stabilized while avoiding excessive contrast.

As described above, according to the image processing apparatus of the present embodiment, for example, when performing gradation processing as image processing, it is possible to prevent the contrast from being excessively high. The sharpness can be prevented from lowering, and when performing dynamic range compression processing, it is possible to stabilize the image to be output according to the image processing to be performed on the image data, such as to achieve a stable density. It becomes possible.

Next, a second embodiment of the present invention will be described with reference to FIG. FIG. 6 is a block diagram illustrating the image processing apparatus 203 taken out from the radiation image processing apparatus according to the second embodiment.

The second embodiment differs from the first embodiment in the part of the image processing device 203, and the other parts are the same. Therefore, the description of the overlapping part will be omitted.

The image processing device 203 includes an image processing condition determining unit 203b, an individual difference extracting unit 203c, and a correction condition determining unit 203d. The individual difference extracting unit 203c and the correction condition determining unit 203d are the same as the individual difference extracting unit 23c and the correction condition determining unit 23d, respectively. The image processing condition determination unit 203b analyzes the signal sent from the image data storage unit 21 and reproduces the signal at a predetermined density and / or contrast.

FIG. 7 is a graph showing an example of a histogram obtained from the signal sent from the image data storage means 21. In FIG. 7, the horizontal axis represents the signal value, and the vertical axis represents the appearance frequency of each signal value. When the analysis is performed based on the signal values of FIG.
Although more signals are included than in the example described with reference to FIG. 3 in which the signal analysis in the OI is performed, the frequency distribution is analyzed, and the reference signal value S (1,1) S (1,2) is similarly detected.

Image processing conditions for normalizing the reference signal values S (1,1) and S (1,2) corresponding to these predetermined structures so as to have a constant density and contrast are determined.

The processes such as the dynamic range compression process, the frequency process, and the gradation process after the reference signal value S is obtained are performed in the same manner as in the first embodiment, so that the description is omitted.

[0109]

According to the invention as set forth in claim 1, 7, 13 or 19, the index is based on the individual difference of the subject such as the difference in the body thickness depending on the physique of the patient and the presence or absence of a fixture such as a cast. Determine the correction conditions for the image processing conditions, and use the determined correction conditions to obtain images obtained by imaging human diagnostic structures that are important for diagnosis, regardless of the patient's physique, positioning, and the presence or absence of equipment such as casts. When outputting, it is possible to output an image with stable density and / or contrast.

According to the present invention, since the correction condition is determined based on the reference signal value, the individual difference of the subject is more accurately reflected, and the density and the contrast are stabilized. It can be.

According to the third, ninth, fifteenth, or twenty-first aspect of the present invention, it is possible to determine the correction condition according to the degree of the individual difference based on the difference between the reference signal values.

According to the invention as set forth in claim 4, 10, 16 or 22, the degree of correction changes according to the degree of physique, so that it is possible to prevent the contrast from becoming too high depending on the patient's body type. As a result, a stable image can be output.

According to the invention as set forth in claim 5, 11, 17 or 23, the degree of correction changes according to the degree of physique, so that sharpness is prevented from being lowered by the patient's physique. Thus, an image with a stable contrast can be output.

According to the sixth, twelfth, eighteenth, or twenty-fourth aspect of the present invention, the degree of correction changes following the degree of physique, so that the degree of dynamic range compression processing is changed according to the patient's body type. Thus, an image having a stable density can be output.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a hardware configuration of an image processing apparatus.

FIG. 2 is a conceptual diagram illustrating the concept of a dynamic range compression process.

FIG. 3 is a conceptual diagram illustrating detection of a region of interest.

FIG. 4 is a graph showing an example of a histogram obtained from image data.

FIG. 5 is an example of a histogram according to a subject.

FIG. 6 is a block diagram showing a part of the radiation image processing apparatus.

FIG. 7 is an example of a histogram of an image signal.

FIG. 8 is a conceptual diagram of a configuration of a radiographic image capturing apparatus according to a conventional technique.

FIG. 9 is an explanatory diagram of gradation processing according to a conventional technique.

FIG. 10 is an explanatory diagram illustrating the state of scattered radiation depending on the physique.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Radiation source 2 Radiation controller 3 Radiation image reader 20 Radiation image display 21 Image data storage means 23 Image processing device 23a Area recognition means 23b Image processing condition determination means 23c Individual difference extraction means 23d Correction condition determination means 25 Output data Storage means 26 Output means

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 2H013 AC06 4C093 AA26 CA04 EB17 FF01 FF08 FF19 FF28 5B057 AA08 BA03 CE03 CE11 DA17 DB02 DC23

Claims (24)

[Claims] 1. An image processing apparatus for processing a radiation image generated based on radiation transmitted through a subject, comprising: a region recognizing unit configured to recognize a specific region corresponding to a predetermined structure of the subject; Image processing condition determining means for determining image processing conditions so that a signal value obtained by analyzing the image signal is reproduced at a predetermined constant density and / or contrast; and An individual difference extracting unit that extracts an index of an individual difference of the subject based on the signal distribution; and a correction condition determining unit that determines a correction condition of the image processing condition determined by the image processing condition determining unit based on the index. An image processing apparatus that performs image processing on the radiation image based on the image processing condition corrected by the correction condition. 2. The image processing condition determining means detects a reference signal value corresponding to a predetermined structure of a subject based on a signal distribution of the image signal, and the correction condition determining means detects a reference signal value based on the reference signal value. The image processing apparatus according to claim 1, wherein the correction condition is determined. 3. The image processing apparatus according to claim 1, wherein a difference between the reference signal values is detected as the index. 4. The method according to claim 1, wherein the image processing is a gradation processing for the radiation image, and the correction condition is a density and / or a contrast of the gradation processing.
4. The image processing device according to 2 or 3. 5. The image processing according to claim 1, wherein the image processing is frequency processing on the radiation image, and the correction condition is a value of an enhancement coefficient used for the frequency processing. apparatus. 6. The image processing apparatus according to claim 1, wherein the image processing is a dynamic range compression processing for the radiation image, and the correction condition is a correction coefficient used for the dynamic range compression processing. Image processing device. 7. An image processing apparatus for processing a radiation image generated based on radiation transmitted through a subject, analyzing an image signal, recognizing a signal value corresponding to a predetermined structure of the subject, and recognizing the signal. Image processing condition determining means for determining an image processing condition so that a value is reproduced at a predetermined constant density and / or contrast; and an individual of the subject based on a signal value corresponding to a predetermined structure of the subject. An individual difference extracting unit for extracting an index of a difference, a correction condition determining unit for determining a correction condition of the image processing condition determined by the image processing condition determining unit based on the index, and an image corrected by the correction condition An image processing apparatus that performs image processing on the radiation image based on processing conditions. 8. The image processing condition determining means detects a reference signal value corresponding to a predetermined structure of a subject based on a signal distribution of the image signal, and the correction condition determining means detects a reference signal value based on the reference signal value. The image processing apparatus according to claim 7, wherein the correction condition is determined. 9. The image processing apparatus according to claim 7, wherein a difference between the reference signal values is detected as the index. 10. The image processing apparatus according to claim 7, wherein the image processing is gradation processing for the radiation image, and the correction condition is a density and / or contrast of the gradation processing. Image processing device. 11. The image processing apparatus according to claim 7, wherein the image processing is frequency processing on the radiation image, and the correction condition is a value of an enhancement coefficient used in the frequency processing.
Or the image processing apparatus according to 9. 12. The image processing apparatus according to claim 7, wherein the image processing is a dynamic range compression processing for the radiation image, and the correction condition is a correction coefficient used for the dynamic range compression processing. Image processing device. 13. An image processing method for processing a radiation image generated based on radiation transmitted through a subject,
Image processing conditions are set such that a specific area corresponding to a predetermined structure of a subject is recognized, and a signal value obtained by analyzing an image signal of the area is reproduced at a predetermined constant density and / or contrast. Is determined, based on the signal distribution of the image signal in the region, an index of the individual difference of the subject is extracted, and a correction condition of the image processing condition is determined based on the index, and the correction condition is corrected by the correction condition. Performing image processing on the radiation image based on the image processing conditions. 14. The method according to claim 1, wherein a reference signal value corresponding to a predetermined structure of the subject is detected based on a signal distribution of the image signal, and the correction condition is determined based on the detected reference signal value. 14. The image processing method according to 13. 15. The image processing method according to claim 13, wherein a difference between the reference signal values is detected as the index. 16. The method according to claim 1, wherein the image processing is gradation processing for the radiation image, and the correction condition is a density and / or contrast of the gradation processing.
The image processing method according to 3, 14, or 15. 17. The method according to claim 13, wherein the image processing is frequency processing on the radiation image, and the correction condition is a value of an enhancement coefficient used for the frequency processing.
16. The image processing method according to 14 or 15. 18. The image processing apparatus according to claim 13, wherein the image processing is a dynamic range compression processing for the radiation image, and the correction condition is a correction coefficient used for the dynamic range compression processing. Image processing method. 19. An image processing method for processing a radiation image generated based on radiation transmitted through a subject,
The image signal is analyzed, a signal value corresponding to a predetermined structure of the subject is recognized, and image processing conditions are determined so that the signal value is reproduced at a predetermined constant density and / or contrast. Based on the distribution of signal values corresponding to the predetermined structure, an index of the individual difference of the subject is extracted, a correction condition of the image processing condition is determined based on the index, and the image processing corrected by the correction condition is performed. An image processing method, wherein image processing is performed on the radiation image based on a condition. 20. A method according to claim 20, wherein a reference signal value corresponding to a predetermined structure of the subject is detected based on a signal distribution of the image signal, and the correction condition is determined based on the detected reference signal value. 20. The image processing method according to 19. 21. The image processing method according to claim 19, wherein a difference between the reference signals is detected as the index. 22. The image processing apparatus according to claim 1, wherein the image processing is gradation processing for the radiation image, and the correction condition is a density and / or contrast of the gradation processing.
22. The image processing method according to 9, 20, or 21. 23. The image processing apparatus according to claim 19, wherein the image processing is frequency processing on the radiation image, and the correction condition is a value of an enhancement coefficient used for the frequency processing.
22. The image processing method according to 20 or 21. 24. The method according to claim 19, wherein the image processing is a dynamic range compression processing for the radiation image, and the correction condition is a correction coefficient used for the dynamic range compression processing. Image processing method.