JPH11290303A - Image determining method, image determining apparatus, and storage medium - Google Patents

Abstract

(57) [Problem] To provide an image discriminating apparatus for accurately and efficiently discriminating a photographing position of a subject from a photographed image. SOLUTION: Profile creation means 113 creates a profile that crosses the highest density value of a first part (such as a lung part of the chest). The concave / convex portion acquiring means 114 acquires the concave and convex portions from the profile created by the profile creating means 113. The characteristic acquiring means 115 acquires the density values and the coordinates of the characteristic values of the concave portion (portion corresponding to the mediastinal portion of the chest) and the convex portion (portion corresponding to the lung of the chest). The determining unit 123 determines the photographing position from the density value acquired by the feature acquiring unit 115 and its coordinates.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image discriminating method and an image discriminating apparatus for automatically discriminating a category of an image by using a feature amount extracted from the image. The present invention relates to an image discriminating method and an image discriminating apparatus for discriminating a photographing position such as a lung front or a lung side from a shape characteristic of a region profile. The present invention also relates to a storage medium in which processing steps for performing the above-described image determination method are stored in a computer-readable manner.

[0002]

2. Description of the Related Art In recent years, with the advance of digital technology, for example, a radiation chest image has been digitized, and the digital image is subjected to image processing and displayed on a CRT or the like, or printed out. .

[0005] In radiographic imaging of the chest, it is general that the imaging position is different depending on the purpose of imaging. For example, depending on the purpose, the photographing position varies, such as photographing from the front of the lung or photographing from the side of the lung. Further, the density distribution of the image differs for each photographing position, and the width of the density distribution of the attention area also differs. For this reason, image processing for display output, print output, and the like is generally performed differently for each photographing position. Therefore, when performing image processing, it is necessary to input a photographing position to an apparatus (image processing apparatus) in order to perform processing suitable for the photographing position. However, the operation of inputting the photographing position one by one every time image processing is performed requires a great deal of labor, and input errors are likely to occur.

In order to solve this problem, there is an automatic discrimination method for automatically discriminating the photographing position using the feature amount of an image. This automatic discrimination method is disclosed in Japanese Patent Publication No. 6-775.
As described in No. 0 or the like, a method is used in which a cumulative histogram of an image is created, and the category of the image is determined based on the value of the rate of change at a substantially central portion of the cumulative histogram.

[0005]

However, in the conventional automatic discrimination method described in Japanese Patent Publication No. Hei 6-7750, a parameter for discriminating the category of an image is fixed in advance, and under a certain condition. Unless the image was taken, the category of the image could not be determined. Also,
According to this method, since the category of the image is determined based on the value of the change rate of the substantially central portion of the cumulative histogram, if the width between the maximum value and the minimum value of the density value changes, the determination accuracy decreases. There was a problem. Furthermore, in this method, when changing photographing conditions and the like, it is necessary to change parameters for determining the category of an image, and this change requires a skilled technique. Therefore, there is a problem that parameters cannot be easily changed.

SUMMARY OF THE INVENTION Accordingly, the present invention has been made to eliminate the above-mentioned drawbacks, and an image discriminating method capable of accurately and efficiently discriminating a photographing position of a subject from a photographed image, and an image discriminating method. It is an object of the present invention to provide a storage medium in which processing steps for performing a determination method are stored in a computer readable manner. Another object of the present invention is to provide an image discriminating apparatus for accurately and efficiently discriminating a photographing position of a subject from a photographed image.

[0007]

For such a purpose,
A first invention is an image discriminating method including a processing step of discriminating a photographing position of a subject from a photographed image of the subject having at least a first part and a second part. A comparing step of comparing the first density value of the photographed image area corresponding to the first part with the second density value of the photographed image area corresponding to the second part; And determining a photographing position.

[0008] In a second aspect based on the first aspect,
The processing step acquires a maximum density value of a captured image area corresponding to the first part as the first density value,
A density value acquiring step of acquiring a minimum density value of a photographed image area corresponding to the second region as the second density value, wherein the comparing step includes the steps of: acquiring the density values obtained in the density value acquiring step; The method includes the step of comparing.

A third invention is an image discriminating method including a process step of discriminating a photographing position of a subject from a photographed image of the subject having at least a first portion and a second portion. Comparing the distance between the first coordinate serving as a feature in the captured image area corresponding to the first part and the second coordinate serving as a feature in the captured image area corresponding to the second part. A comparison step; and a determination step of determining the photographing position from the comparison result of the comparison step.

In a fourth aspect based on the third aspect,
The processing step acquires the coordinates of the midpoint of the photographed image area corresponding to the first part as the first coordinates, and calculates the coordinates of the midpoint of the photographed image area corresponding to the second part as the first coordinate. A coordinate acquisition step of acquiring the coordinates as 2 coordinates,
The comparing step includes a step of comparing distances obtained by the respective coordinates obtained in the coordinate obtaining step.

A fifth invention is an image discriminating method including a process step of discriminating a photographing position of a subject from a photographed image of the subject having at least a first portion and a second portion. A density value obtaining step of obtaining a maximum density value of a captured image area corresponding to the first region;
In the captured image, a profile creation step for creating a profile of an image density value passing through the coordinates of the highest density value obtained in the density value acquisition step, and a concave portion and a convex portion from the profile created in the profile creation step An uneven part obtaining step to obtain; a density obtaining step of obtaining a characteristic value and coordinates of each part from the concave part and the convex part obtained in the uneven part obtaining step; and a density value of each part obtained in the above characteristic obtaining step. And a determining step of determining the photographing position from the coordinates.

According to a sixth aspect of the present invention, in the fifth aspect,
The processing step includes a pass-through deletion step of deleting a pass-through area and an area adjacent to the pass-through area from the captured image within a predetermined width, and the density value obtaining step is performed in the pass-through deletion step. A step of performing a process of obtaining the maximum density value on a captured image.

[0013] In a seventh aspect based on the fifth aspect,
The feature obtaining step includes a step of obtaining a highest density value of the convex portion and a step of obtaining a lowest density value of the concave portion, and the determining step includes determining a highest density value of the convex portion obtained in the feature obtaining step. A first comparison step of comparing the lowest density value of the concave portion with the first density step, and a first determination step of determining a photographing body position based on a comparison result of the first comparison step.

According to an eighth aspect based on the fifth aspect,
The feature acquiring step includes a step of obtaining coordinates of a highest density value of the convex portion, and a process of obtaining coordinates of a lowest density value of the concave portion, and the determining step includes calculating the coordinates of the convex portion obtained in the feature obtaining step. A second comparing step of comparing the coordinates of the highest density value with the coordinates of the lowest density value of the concave portion; and a second determining step of determining a photographing position based on a result of the comparison in the second comparing step. It is characterized by.

According to a ninth aspect, in the fifth aspect,
The determining step includes a third determining step of determining a photographing position based on whether or not a concave portion is obtained in the uneven portion acquiring step.

According to a tenth aspect, in any one of the first, third, and fifth aspects, the photographed image to be processed is a smoothed image.

In an eleventh aspect based on any one of the first, third and fifth aspects, the photographed image is a radiation chest image, the first part is a lung, and the second part is a lung part. Is characterized by being a mediastinum.

A twelfth invention is an image discriminating apparatus for discriminating a photographing position of a subject from a photographed image of the subject having at least a first portion and a second portion, and corresponds to the first portion. Comparing means for comparing the first density value of the photographic image area with the second density value of the photographic image area corresponding to the second portion; and determining the photographic position from the comparison result of the comparing means. Means.

In a thirteenth aspect based on the twelfth aspect, a maximum density value of a photographed image area corresponding to the first part is obtained as the first density value, and the maximum density value corresponding to the second part is obtained. The image processing apparatus further includes a density value obtaining unit that obtains the lowest density value of the captured image area as the second density value, wherein the comparing unit compares the respective density values obtained by the density value obtaining unit. .

A fourteenth invention is an image discriminating apparatus for discriminating a photographing position of a subject from a photographed image of the subject having at least a first part and a second part, and corresponds to the first part. Comparison means for comparing the distance between the first coordinate serving as a feature in the captured image area and the second coordinate serving as a feature in the captured image area corresponding to the second part, and a comparison between the comparison means Determining means for determining the photographing position from the result.

In a fifteenth aspect based on the fourteenth aspect, the coordinates of the midpoint of the photographed image area corresponding to the first part are obtained as the first coordinates, and the coordinates corresponding to the second part are obtained. The image processing apparatus further includes a coordinate acquiring unit that acquires coordinates of a middle point of the captured image area as the second coordinates, wherein the comparing unit includes:
The distance obtained by each coordinate obtained by the coordinate obtaining means is compared.

A sixteenth invention is an image discriminating apparatus for discriminating a photographing position of a subject from a photographed image of the subject having at least a first part and a second part, and corresponds to the first part. Density value obtaining means for obtaining a maximum density value of a captured image area; and profile creation means for creating a profile of an image density value passing through coordinates of the maximum density value obtained by the density value obtaining means in the captured image; A concave / convex portion acquiring unit that obtains concave portions and convex portions from the profile created by the profile creating unit, and a feature acquisition that obtains a density value and a coordinate of a characteristic value of each portion from the concave and convex portions obtained by the concave / convex portion acquiring unit. Means for determining the photographing position from the density values and the coordinates of each part obtained by the feature acquiring means.

In a seventeenth aspect based on the sixteenth aspect, the image processing apparatus further comprises a pass-through deletion means for deleting a pass-through region and a region in contact with the pass-through region from the photographed image within a predetermined width. The means performs processing for obtaining the maximum density value on the photographed image processed by the pass-through deletion means.

In an eighteenth aspect based on the sixteenth aspect, the feature acquiring means acquires a maximum density value of the convex portion, a minimum density value of the concave portion, and the discriminating means comprises
The highest density value of the convex portion obtained by the feature obtaining means,
First comparing means for comparing with the lowest concentration value of the concave portion,
A first determining unit that determines a photographing position based on a comparison result of the first comparing unit.

In a nineteenth aspect based on the sixteenth aspect, the feature acquiring means obtains the coordinates of the highest density value of the projection and the coordinates of the lowest density value of the recess, and A second comparing means for comparing the coordinates of the highest density value of the convex portion obtained by the feature acquiring means with the coordinates of the lowest density value of the concave portion, and a comparison result of the second comparing means. A second determination unit that determines a photographing position.

In a twentieth aspect based on the sixteenth aspect, the discriminating means includes a third discriminating means for discriminating a photographing position based on whether or not a concave portion is obtained by the uneven portion acquiring means. It is characterized by.

According to a twenty-first aspect, in any one of the twelfth, fourteenth, and sixteenth aspects, the photographed image to be processed is a smoothed image.

According to a twenty-second invention, in any one of the twelfth, fourteenth, and sixteenth inventions, the photographed image is a radiation chest image, the first part is a lung, and the second part is a lung part. Is characterized by being a mediastinum.

A twenty-third aspect of the present invention is a storage medium storing a computer readable processing step of determining a photographing position of a subject from a photographed image of the subject having at least a first part and a second part. The processing step includes a processing step of the image determining method according to any one of claims 1 to 11.

[0030]

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

(First Embodiment) The present invention provides, for example,
It is applied to an image discriminating apparatus 100 as shown in FIG.

As shown in FIG. 1, the image discriminating apparatus 100 includes a feature amount extracting unit 110 and a feature amount extracting unit 110.
The image discriminating unit 120 to which the output of the image is supplied, and the control unit 13 that controls each of the feature amount extracting unit 110 and the image discriminating unit 120
0 and a program memory 140 accessed by the control unit 130.

Various processing programs including, for example, processing programs in accordance with the flowchart shown in FIG. 2 are stored in the program memory 140 in advance, and these processing programs are read out and executed by the control unit 130. Thus, operation control of the feature amount extraction unit 110 and the image determination unit 120, which will be described later, is performed.

For example, as shown in FIG. 3, the feature quantity extraction unit 110 passes through the input image data (here, the radiation chest image data f (x, y) as shown in FIG. 4). A pass-through deletion circuit 11 for deleting a region (a shaded portion in FIG. 4) and a body region in contact with the pass-through region at a fixed width.
1, a maximum value calculation circuit 112 for calculating the maximum density value and its coordinates from the area not deleted by the pass-through deletion circuit 111, and a coordinate of the maximum density value calculated by the maximum value calculation circuit 112 Creating circuit 113 for creating a profile passing through, an unevenness calculating circuit 114 for calculating an unevenness area from the profile created by the profile creating circuit 113, and a density of the unevenness calculated from the unevenness area calculated by the unevenness calculating circuit 114. It has a minimum value and a minimum value and a minimum value calculation circuit 115 for calculating its coordinates.

The image discriminating section 120 determines a photographing position from the uneven portion area calculated by the uneven portion calculating circuit 114, and the highest density value calculated by the density value calculating circuit 115. A first determination circuit 123 that determines the photographing position from the values and the minimum value, a second determination circuit 122 that determines the photographing position from the coordinates of the highest value and the lowest value of the density value calculated by the density value calculation circuit 115, It has.

The image discriminating apparatus 100 as described above operates as follows by reading and executing the processing program of FIG. 2 stored in the program memory 140 by the control unit 130.

First, the pass-through deletion circuit 111 determines, in the input image data f (x, y), a pass-through region in the irradiation region and a body region which is in contact with the pass-through region with a certain width, for example, Replace with a 0 "pixel (step S201).
Specifically, for the input image data f (x, y),

[0038]

(Equation 1)

The conversion of the image represented by the following equation (1) is performed. In this equation (1), “f1 (x, y)” is an image obtained by deleting a through region and a body region in contact with the through region with a certain width from the image data f (x, y). Show data. “Sgn (x, y)” has constants Th1 determined by experiments and the like and constants d1 and d2 that determine the width of deleting the body region.

[0040]

(Equation 2)

Is represented by the following equation (2).

Next, the maximum value calculating circuit 112 calculates the maximum value max of the density value and its coordinates in the image data f1 (x, y) obtained by the through deletion circuit 111 (step S202).

Next, the profile creation circuit 113 calculates the maximum value max of the density value calculated by the maximum value calculation circuit 112.
, A vertical profile is created on the body side (step S203). Here, as shown in FIG. 4, it is assumed that a profile in the horizontal axis a direction passing through the maximum value max is created. The profile creation circuit 113 creates a profile as shown in FIGS. 5A to 5F, for example. In FIG. 5, (a) is a standard profile in which the imaging position is in front of the lung, and (b) to (e), the chest is directed to the sensor surface and the left body side is in contact with the sensor surface. Lift the right body side with the sensor face and body direction at 15 °, 30 °.
It shows the profile at each photographing position when the angle is gradually changed to, for example, °, 45 °, and 60 °. Also,
(F) shows a standard profile in which the photographing position is the state of the lung side surface.

Next, the uneven portion calculating circuit 114 calculates an uneven portion region from the profile created by the profile creating circuit 113 (step S204). In particular,

[0045]

(Equation 3)

The x region represented by the following equation (3) is calculated. In this equation (3), “y1” indicates the coordinate position on the y-axis of the highest value of the density value calculated by the highest value calculation circuit 112, and “d” indicates a constant determined by an experiment or the like. . Then, the concavo-convex portion calculation circuit 114 sets the x region calculated by the above equation (3) as a concave region and the other region as a convex region.

Next, the third determination circuit 121 determines whether or not the concave / convex portion has been calculated by the concave / convex portion calculation circuit 114 (step S205). When it is determined that “there is no concave area” in step S205, the third determination circuit 121 determines that the imaging position is the side surface (step S206).
This is because, as shown in FIG. 5 (f), when the photographing position is the side surface, the concave region does not exist. After this determination, the process ends.

On the other hand, if it is determined in step S 205 that “there is a concave area”, the density value calculation circuit 115 calculates the maximum value a ₁ of the left convex area based on the uneven area calculated by the uneven area calculation circuit 114. When, calculates the coordinates outermost high a _1, and the lowest value b of the recess region, and the coordinates of the minimum value b, the maximum value a ₂ of the right convex region, and the coordinates outermost high a ₂ (Step S207).

The first judgment circuit 123 calculates the maximum value a _{1 of} the left convex area calculated by the density value calculation circuit 115.
Is compared with the lowest value b of the recessed area, and the photographing position is determined (step S208). In particular,

[0050]

(Equation 4)

The determination is made from the value H obtained by the following equation (4). Here, for example, a table which is a correspondence table between the value H and the photographing position is provided in advance, and the photographing position corresponding to the actually obtained value H is obtained and determined from the table.

The second determination circuit 123 calculates a distance d between the coordinates of the maximum value a ₁ of the left convex area calculated by the density value calculation circuit 115 and the coordinates of the minimum value b of the concave area.
_1, and the coordinate maximum value a ₂ of the right convex region, compares the distance d ₂ between the coordinates of the minimum value b of the recess region, either the tube of the left and right sides of the chest (radiation emitting port ) Is determined (step S209). Specifically, “d ₁ <d ₂ ”
If it is a relationship, it is determined that the right side is a photographing position close to the tube in terms of distance, and if it is a relationship of “d ₁ > d ₂ ”,
It is determined that the left side surface is in the photographing position close to the tube in terms of distance. After this determination, the process ends.

As described above, in this embodiment, the through region and the in-vivo region having a fixed width in contact with the through region are deleted, so that the maximum density value of the in-vivo region can be reduced in a short time. Extraction can be performed stably with high accuracy.
In addition, since the imaging position is determined from the shape of the profile passing through the maximum value max of the density value of the in-vivo region, the imaging position is not affected by changes in imaging conditions and the like.
The photographing position can be determined accurately in a short time.
Further, the distance d ₁ between the coordinates of the highest value a ₁ of the left chest region and the coordinates of the lowest value b of the concave region, the coordinates of the highest value a ₂ of the right convex region of the chest, and the concave portion with the arrangements so as to compare the distance d ₂ between the lowest value b of the coordinate areas, accurately capturing posture can determine whether a left or a right.

(Second Embodiment)

In the present embodiment, the configuration of the image discriminating section 120 of the image discriminating apparatus 100 (FIG. 1) in the first embodiment is, for example, as shown in FIG. That is, in the present embodiment, the second determination circuit 1
The configuration for supplying the output of the second 22 to the first determination circuit 123 is different from that of the first embodiment.

Here, only the configuration different from that of the first embodiment will be specifically described.

First, as described above, the second determination circuit 122 calculates the coordinates of the maximum value a ₁ of the left convex area calculated by the density value calculation circuit 115 and the coordinates of the minimum value b of the concave area. the distance d ₁ between the maximum value a ₂ of the coordinates of the right convex region, the relationship between the distance d ₂ between the coordinates of the minimum value b of the recess region, the "d ₁ <d _2" Which _one of the right and left sides of the chest is a tube (radiation emission port), depending on whether or not “d ₁ > d ₂ ”
Is determined to be close to Therefore, the second determination circuit 122
Supplies this determination result to the first determination circuit 123.

When the relationship between the distance d ₁ and the distance d ₂ satisfies “d ₁ <d ₂ ” according to the result of the determination from the second determination circuit 122, the first determination circuit 123 Then, the photographing position is determined by using the above equation (4). On the other hand, if it is any other relationship,

[0059]

(Equation 5)

The photographing position is determined by using the following equation (5). That is, the photographing position is determined by comparing the highest value a _{2 of} the right convex region calculated by the density value calculating circuit 115 with the lowest value b of the concave region.

As described above, in the present embodiment, “d ₁
In the case of <d ₂ ”(when the right side is a photographing position close to the tube in terms of distance), the determination is made by using the equation (4) using the maximum value a ₁ of the left convex region. Yes, otherwise (when the left side is in a shooting position close to the tube in terms of distance)
, The determination is performed employing Equation (5) using the maximum value a ₂ of the right convex region, and so on, closer to the small tube fluctuation than the highest value farther to the tube By using the highest value of the convex region of the lung region in the judgment formula,
More stable determination of the photographing position can be performed.

(Third Embodiment)

In this embodiment, the maximum value calculation circuit 112 shown in FIG. 3 calculates the maximum density value from the image in which the through area and the area in contact with the through area at a constant width are deleted. In order to perform the process of calculating the coordinates,
For example, a processing program according to a flowchart as shown in FIG. 7 is employed. The processing program of FIG.
The maximum value calculation circuit 112, which is stored in the program memory 140 in advance and read and executed by the control unit 130, operates as follows.

First, a cumulative histogram of an image obtained by the pass-through deletion circuit 111 (an image obtained by deleting a pass-through area and an area in contact with the pass-through area with a certain width) is created (step S301).

Next, on the cumulative histogram created in step S301, a constant T determined by an experiment or the like is set.
The pixel of h3 or more is extracted (step S302).

Next, the center of gravity of the pixel extracted in step S302 is calculated (step S303), and a profile passing through the coordinates of the center of gravity is smoothed (step S304).
For the smoothing here, for example, a density average, an intermediate value in a certain area, a gray scale morphology and the like are used.

The highest density value and its coordinates on the profile smoothed in step S304 are set as the coordinates of the highest density value and the highest density value to be output to the subsequent profile creation circuit 113 (step S305).

As described above, in the present embodiment, the cumulative histogram is used, and the center of gravity having a certain density value (Th3) or more is set as the coordinates of the highest density value and the highest density value used in the profile creation circuit 113. By doing,
Noise can be removed, and a stable feature amount can be obtained. In addition, since the profile passing through the calculated center of gravity is smoothed and configured to be the coordinates of the highest density value and the highest density value again, the feature amount can be stably and accurately calculated without being affected by noise. Can be.

It should be noted that the object of the present invention is to
A storage medium storing software program codes for realizing the functions of the host and the terminal according to the embodiment is supplied to a system or apparatus, and a computer (or CPU or MPU) of the system or apparatus is stored in the storage medium. It goes without saying that this can also be achieved by reading and executing the program code. In this case, the program code itself read from the storage medium realizes the functions of the above-described first to third embodiments, and the storage medium storing the program code constitutes the present invention. Becomes

As storage media for supplying the program code, ROM, floppy disk, hard disk, optical disk, magneto-optical disk, CD-ROM, C
DR, a magnetic tape, a nonvolatile memory card, or the like can be used.

By executing the program code read by the computer, not only the functions of the above-described first to third embodiments are realized, but also the computer executes the program code based on the instructions of the program code. It is needless to say that the operating OS or the like performs a part or all of the actual processing, and the processing realizes the functions of the above-described first to third embodiments.

Further, after the program code read from the storage medium is written in a memory provided in an extended function board inserted into the computer or a function extension unit connected to the computer, based on the instructions of the program code, The CPU provided in the function expansion board or function expansion unit performs part or all of the actual processing,
It goes without saying that the processing may realize the functions of the above-described first to third embodiments.

[0073]

As described above, in each of the first and twelfth aspects of the present invention, the first density value of the first part (the lung part of the chest, etc.) and the second part (the chest part) And the second density value of the mediastinum and the like, and the photographing position of the subject is determined from the comparison result. This allows
The processing time can be shortened, and the photographing position can be determined efficiently and accurately. In addition, the first density value is the first density value.
If the maximum density value of the part is used and the minimum density value of the second part is used as the second density value, a more stable feature value of the density value can be obtained, so that the discrimination accuracy is improved. Can be done. Further, if the smoothed photographed image is configured to be used as the processing target image, the photographing posture can be stably determined without being affected by noise.

In each of the third and fourteenth aspects of the present invention, the distance between the first coordinate, which is the characteristic of the first part, and the first coordinate, which is the characteristic of the second part, in the captured image, The photographing position of the subject is determined. This allows
The direction of the photographing position can be determined. For example, in the chest image, the first region is defined as the left and right lung regions, and the second region is defined as the mediastinum, and the distance between the coordinates characteristic of the left lung region and the coordinates characteristic of the mediastinal region. By comparing the distance between the characteristic coordinates of the right lung and the characteristic coordinates of the mediastinum, and determining the photographing position based on the comparison result, the left and right photographing positions can be stably determined. Can be determined. Also,
Using the coordinates of the midpoint of the first part as the first coordinates,
If the middle point of the second part is used as the coordinates of, the distance information used for the determination can be obtained more accurately and stably, so that the direction of the photographing body position can be determined with higher accuracy. Further, if the smoothed photographed image is configured to be used as the processing target image, the photographing posture can be stably determined without being affected by noise.

In each of the fifth and sixteenth inventions, a profile that crosses the highest density value of the first part (the lung part of the chest, etc.) is created, and the profile corresponds to the concave part (the mediastinum part of the chest, etc.). And the coordinates of the density values, which are the characteristics of the convex portions (portions corresponding to the lungs of the chest, etc.), are obtained, and the photographing position is determined based on the results. This makes it possible to easily obtain the density values and the coordinates thereof, which are the characteristics of the concave portions and the convex portions, so that the processing time can be reduced and the photographing position can be determined efficiently and accurately. In addition, if the maximum density value of the convex portion is compared with the minimum density value of the concave portion, and the comparison result is used to determine the photographing position, the photographing position can be determined more accurately. If the configuration is such that the photographing position is determined based on the distance between the coordinates of the highest density value of the convex portion and the coordinates of the lowest density value of the concave portion, the direction of the photographing position can be determined. For example, in the chest image, the first part is the left and right lung parts, the second part is the mediastinum part, and the left convex part (left lung part)
, The distance between the coordinates of the highest density value of the concave portion (mediastinum portion), the coordinates of the highest density value of the right convex portion (right lung portion), and the coordinates of the highest density value of the concave portion (mediastinum portion). By comparing the distance with the coordinates of the density value and determining the photographing position based on the comparison result, the left and right photographing positions can be determined stably. Further, if the configuration is such that the photographing position is determined based on whether or not the concave portion is obtained, for example, in a chest image,
When the concave portion is not obtained, the photographing position is directed sideways (90
° orientation) and terminate the process, the processing time can be shortened and the photographing position can be determined efficiently. Even if a concave portion is not obtained, the stable photographing position can be determined. It can be performed. In addition, if a configuration is adopted in which a through region and a region in contact with the through region from a captured image are deleted within a predetermined width and the image is used as a processing target image, a density value as a feature amount of the concave portion and the convex portion is obtained. Can be stably extracted. For example, in a chest image, since the highest density value can be obtained only from the lung, the density value can be more stably extracted as a feature value. Further, if the smoothed photographed image is configured to be used as the processing target image, the photographing posture can be stably determined without being affected by noise.

According to the twenty-third aspect, it is possible to provide an apparatus and a system for accurately and efficiently determining the photographing position of a subject from a photographed image.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an image discriminating apparatus to which the present invention is applied in a first embodiment.

FIG. 2 is a flowchart for explaining a processing program executed by the image discriminating apparatus.

FIG. 3 is a block diagram illustrating a detailed configuration of a feature amount extraction unit and an image determination unit of the image determination device.

FIG. 4 is a diagram illustrating an example of image data input to the image discriminating apparatus.

FIG. 5 is a diagram illustrating an example of a profile of a lung region.

FIG. 6 is a block diagram illustrating a configuration of the image determination unit according to the second embodiment.

FIG. 7 is a flowchart illustrating a processing program executed by a maximum value calculation circuit of the feature amount extraction unit in the third embodiment.

[Explanation of symbols]

 Reference Signs List 110 feature amount extraction unit 111 slip-through deletion circuit 112 maximum value calculation circuit 113 profile creation circuit 114 unevenness calculation circuit 115 density value calculation circuit 120 image determination unit 121 third determination circuit 122 second determination circuit 123 first determination circuit

Claims (23)

[Claims] 1. An image discriminating method including a processing step of determining a photographing position of a subject from a photographed image of the subject having at least a first part and a second part, wherein the processing step includes the first step. A comparing step of comparing a first density value of a photographed image area corresponding to the part with a second density value of a photographed image area corresponding to the second part; A determination step of determining a body position. 2. The processing step comprising: obtaining a highest density value of a photographed image area corresponding to the first part as the first density value, and obtaining a lowest density value of a photographed image area corresponding to the second part. 2. A density value acquiring step of acquiring a value as the second density value, wherein the comparing step includes a step of comparing each density value obtained in the density value acquiring step. Image discrimination method. 3. An image discriminating method including a processing step of determining a photographing position of a subject from a photographed image of the subject having at least a first part and a second part, wherein the processing step includes the first step. A comparison step of comparing a distance between the first coordinate serving as a feature in the captured image area corresponding to the part and the second coordinate serving as a feature in the captured image area corresponding to the second part; A determining step of determining the photographing position from the comparison result of the comparing step. 4. The processing step comprises: obtaining coordinates of a middle point of the photographed image area corresponding to the first part as the first coordinates, and obtaining a middle point of the photographed image area corresponding to the second part. A coordinate acquisition step of acquiring the coordinates of the first coordinate as the second coordinate, and the comparing step includes a step of comparing distances obtained by the respective coordinates obtained in the coordinate acquisition step. 3. The image discriminating method according to 3. 5. An image discriminating method including a processing step of determining a photographing position of a subject from a photographed image of the subject having at least a first part and a second part, wherein the processing step includes the first step. A density value obtaining step of obtaining the highest density value of the photographed image area corresponding to the part, and a profile of an image density value passing through the coordinates of the highest density value obtained in the density value obtaining step in the captured image. A profile creation step, a concavo-convex portion acquiring step of obtaining concave portions and convex portions from the profile created in the profile creating step, and a density value and a characteristic value of each portion from the concave portion and convex portion obtained in the irregular portion acquiring step. A feature obtaining step of obtaining coordinates, and a determining step of determining a photographing position from the density values and the coordinates of the respective parts obtained in the feature obtaining step. Image discrimination method characterized by comprising the flop. 6. The processing step includes a pass-through deletion step of deleting, within a predetermined width, a pass-through area and an area adjacent to the pass-through area from the captured image, and the density value acquiring step includes: 6. The image discriminating method according to claim 5, further comprising a step of performing a process of obtaining the maximum density value on the photographed image processed in the step. 7. The feature obtaining step includes a step of obtaining a highest density value of the convex portion and obtaining a lowest density value of the concave portion, and the determining step includes the step of obtaining the convex portion obtained in the feature obtaining step. A first comparing step of comparing the highest density value of the concave portion with the lowest density value of the concave portion, and a first determining step of determining a photographing position based on a comparison result of the first comparing step. 6. The image discriminating method according to claim 5, wherein 8. The feature obtaining step includes a step of obtaining coordinates of a highest density value of the convex portion and obtaining a coordinate of a lowest density value of the concave portion, and the determining step is obtained in the feature obtaining step. A second comparing step of comparing the coordinates of the highest density value of the convex portion with the coordinates of the lowest density value of the concave portion, and a second discrimination for discriminating the photographing position based on the comparison result of the second comparing step. 6. The method according to claim 5, further comprising the steps of: 9. The image discriminating method according to claim 5, wherein said discriminating step includes a third discriminating step of discriminating a photographing position based on whether or not a concave portion is obtained in said uneven portion acquiring step. Method. 10. The method according to claim 1, wherein the photographed image to be processed is a smoothed image. 11. The imaging device according to claim 1, wherein the photographed image is a radiation chest image, the first part is a lung, and the second part is a mediastinum. , And 5. 12. An image discriminating apparatus for discriminating a photographing position of a subject from a photographed image of the subject having at least a first part and a second part, wherein the photographing position of the photographed image area corresponding to the first part is determined. Comparing means for comparing a first density value with a second density value of a photographed image area corresponding to the second part, and determining means for determining the photographing position from the comparison result of the comparing means An image discriminating apparatus characterized in that: 13. A maximum density value of a captured image area corresponding to the first part is obtained as the first density value, and a minimum density value of a captured image area corresponding to the second part is determined as the second density value. 13. The image discriminating apparatus according to claim 12, further comprising a density value obtaining unit that obtains the density values of the image data, wherein the comparing unit compares the respective density values obtained by the density value obtaining unit. 14. An image discriminating apparatus for discriminating a photographing position of a subject from a photographed image of the subject having at least a first part and a second part, wherein the photographing position is within a photographed image area corresponding to the first part. Comparing means for comparing the distance between the first coordinate, which is the characteristic of the above, and the second coordinate, which is the characteristic in the captured image area corresponding to the second part; An image discriminating apparatus, comprising: discriminating means for discriminating a body position. 15. The coordinates of the midpoint of the photographed image area corresponding to the first part are acquired as the first coordinates, and the coordinates of the midpoint of the photographed image area corresponding to the second part are obtained as the first coordinate. 2. The apparatus according to claim 1, further comprising coordinate acquisition means for acquiring the coordinates as a second coordinate, wherein the comparison means compares the distances obtained by the respective coordinates obtained by the coordinate acquisition means.
4. The image discriminating apparatus according to 4. 16. An image discriminating apparatus for discriminating a photographing position of a subject from a photographed image of the subject having at least a first part and a second part, wherein the photographing position of the photographed image area corresponding to the first part is determined. Density value acquiring means for obtaining a maximum density value; profile creating means for creating a profile of an image density value passing through the coordinates of the maximum density value obtained by the density value acquiring means in the captured image; A concave and convex portion obtaining means for obtaining a concave portion and a convex portion from the profile created in the above; a characteristic obtaining portion for obtaining a density value and a coordinate of a characteristic of each portion from the concave and convex portions obtained by the concave and convex portion obtaining device; An image discriminating apparatus comprising: a discriminating unit that discriminates a photographing position from the density value and coordinates of each unit obtained by the feature acquiring unit. 17. A pass-through deletion unit for deleting a pass-through area and an area adjacent to the pass-through area from the photographed image within a predetermined width, wherein the density value acquisition unit is processed by the pass-through deletion unit. 17. The image discriminating apparatus according to claim 16, wherein the processing for obtaining the maximum density value is performed on the taken image. 18. The feature acquiring unit acquires a maximum density value of the convex portion, acquires a minimum density value of the concave portion, and the determining unit acquires a maximum density value of the convex portion obtained by the feature acquiring unit. A first comparing means for comparing a density value with a minimum density value of the concave portion, and a first determining means for determining a photographing position based on a comparison result of the first comparing means. Item 17. The image discriminating apparatus according to Item 16. 19. The feature acquiring means obtains the coordinates of the highest density value of the convex portion and the coordinates of the lowest density value of the concave portion, and the determining means obtains the coordinates of the lowest density value obtained by the feature obtaining means. Second comparing means for comparing the coordinates of the highest density value of the convex portion with the coordinates of the lowest density value of the concave portion, and second determining means for determining the photographing position based on the comparison result of the second comparing means. 17. The image discriminating apparatus according to claim 16, comprising: 20. The image discriminating apparatus according to claim 16, wherein said discriminating means includes a third discriminating means for discriminating a photographing position based on whether or not a concave portion is obtained by said uneven portion acquiring means. apparatus. 21. The photographed image to be processed is a smoothed image.
The image discriminating apparatus according to any one of Items 4 and 16. 22. The radiographic apparatus according to claim 12, wherein the radiographic image is a radiological chest image, the first site is a lung, and the second site is a mediastinum. , And 16. 23. A storage medium storing a computer-readable processing step of determining a photographing position of a subject from a photographed image of the subject having at least a first part and a second part. A storage medium comprising a processing step of the image determination method according to claim 1.