JPH04242636A - Recognition device for radioactive ray irradiated field and radioactive ray image processing device - Google Patents

Abstract

PURPOSE:To precisely recognize a radioactive ray irradiated field by recognizing a rectangular field of a radioactive ray image which is formed in accordance with a transmittance degree of radioactive rays which transmit through each part of an object to be picked up, even though the objected to be picked up overlaps with the boundary of the irradiated filed. CONSTITUTION:Of the X- and Y-axes as the abscissa and the ordinate of an image area, the vertical Y-axis is selected as a specified axis since the front view of the chest of a human body is to be picked up, two band-like zones A1, A2 including the lungs field and one band-like zone including the back bones, that is, three zones in total, are set in accordance with thinned-out image data. With the use of such a fact that a difference between image data levels due to differences in transmittance of radioactive ray through objects to be picked up contained respectively in a plurality of band-like zones varies between the inside and outside of the irradiated area, the irradiated area is recognized. Accordingly, there is presented a difference in transmittance between parts of the object to be picked up included in each band-like zone, and accordingly, it is desire able to have a large deviation in the case of comparison of image data between the band-like zones.

Description

[Detailed description of the invention]

0001

[Field of Industrial Application] The present invention relates to a radiation irradiation field recognition device and a radiation image processing device. The present invention relates to a device that recognizes images based on image data, and a device that processes image data based on the recognized irradiation field.

0002

[Prior Art] Radiographic images such as X-ray images are often used for disease diagnosis. So-called radiography has been widely used in the past, in which a film using silver salt is developed by being irradiated with visible light, which is then irradiated onto a film using silver salt in the same manner as in ordinary photography.

However, in recent years, methods have been devised to directly extract images from the phosphor layer without using a film coated with silver salt. This method is
The radiation transmitted through the object is absorbed by a phosphor, and then this phosphor is excited, for example, with light or thermal energy, so that the phosphor emits the radiation energy accumulated through the absorption as fluorescence, and this There is a method of photoelectrically converting fluorescence to obtain an image signal. in particular,
For example, U.S. Patent No. 3,859,527 and JP-A-55
Publication No. 12144 and the like discloses a radiation image conversion method using a stimulable phosphor and using visible light or infrared rays as the stimulable excitation light.

[0004] The radiographic image signals obtained in this way are visualized as they are or after being subjected to image processing and output to a printer, CRT, etc., but they are not digitized for image processing by a computer. Often. In addition, the silver halide film on which the radiation image has been recorded is irradiated with light from a light source such as a laser or a fluorescent lamp to obtain transmitted light through the silver halide film, and the transmitted light is photoelectrically converted to obtain a radiation image signal. There is also a way to digitize it.

[0005] Generally, when taking a radiation image, an irradiation field diaphragm is sometimes used to limit the radiation irradiation area. In particular, when photographing the human body for medical diagnosis, it is important to minimize radiation exposure to parts of the body that are not necessary for diagnosis, and to prevent scattered radiation from areas that are not necessary for diagnosis from reaching the image areas necessary for diagnosis. An irradiation field diaphragm is used to prevent the image from entering and deteriorating the image quality. The shape of the irradiation field aperture is rectangular in most cases, and in particular, as shown in FIG.
When photographing the chest or abdomen, an aperture is often provided at the bottom of the image to prevent exposure of the genital organs.

[0006] When performing image processing and reproducing a digital radiation image taken with the irradiation field apertured in this way, it is convenient if it is possible to recognize which region is within the irradiation field. As a method for recognizing a radiation irradiation field, there is a method disclosed in, for example, Japanese Patent Laid-Open No. 133760/1983. This method adds and aggregates image data in the orthogonal X-axis and Y-axis directions, and determines the position on the Y-axis (X-axis) where the aggregated data in the X-axis (Y-axis) direction is a predetermined threshold. X at this position on the Y axis (X axis)
A straight line in the axis (Y-axis) direction is set as a boundary line that determines the irradiation field. That is, when a rectangular irradiation field is narrowed down, the irradiation field is recognized by utilizing the sudden change in the level of image data near the boundary of the irradiation field, as shown in FIG.

[0007]

However, with the method of recognizing the irradiation field based on changes in the image data level as described above, if the boundary of the irradiation field overlaps with the subject, in other words, the irradiation field may cause When a portion of the image is photographed in sections, there is a problem in that the recognition accuracy of the irradiation field deteriorates.

[0008] If the subject is smaller than the irradiation field and the radiation around the subject is directly irradiated onto the recording medium without passing through the subject, the dose level will be highest in the irradiation field near the boundary, and conversely, Since there is no direct radiation irradiation outside the irradiation field, the dose level is sufficiently small, so there is a large difference in image data between the boundaries of the irradiation field and the outside.

However, if the object is larger than the irradiation field and there is a side that overlaps with the object most of the four sides of the rectangle of the irradiation field, the radiation in the part of the object where the boundaries of the irradiation field overlap The lower the level of transmission, the smaller the difference from the dose level outside the irradiation field, making it difficult to identify the boundary using the method described above.

That is, when recognizing an irradiation field based on the level difference of image data inside and outside the irradiation field boundary as described above, it is assumed that the image data level outside the irradiation field is sufficiently small. , when the boundary of the irradiation field overlaps the subject as described above, due to the influence of scattered radiation within the subject (hereinafter referred to as scattered rays), there may actually be light on the recording medium located outside the irradiation field. Figure 1
As shown in Fig. 2, the dose level (image data level) outside the irradiation field increases.

[0011] For this reason, the identification of boundaries based on level changes in image data becomes ambiguous, and since the influence of the scattered radiation has large variations depending on the subject, it is difficult to set an appropriate threshold value, and it is difficult to set a threshold value. Even a slight error could result in a large difference in the recognized irradiation field (see Figure 12). In particular, in a configuration where various image processing conditions such as gradation processing and frequency processing are determined based on image data, and when irradiation field aperture is performed, image data other than the irradiation field is used to determine the processing conditions. If it is included in the image data used, it will be affected by the image data outside the irradiation field, making it impossible to determine the optimal processing conditions. Therefore, when determining image processing conditions, it is necessary to exclude image data outside the irradiation field. However, due to the reasons mentioned above, errors occur in the recognition of the irradiation field, and the image data used to determine the processing conditions may not be included in the irradiation field. Field data was included, which could lead to inappropriate setting of processing conditions.

Conversely, due to errors caused by the above reasons, image data necessary for diagnosis may be recognized as image data outside the irradiation field, and processing conditions may be determined based on insufficient data, resulting in inappropriate image data. In some cases, treatment was applied. The present invention has been made in view of the above problems,
Even if the subject overlaps the boundary of the irradiation field,
It is an object of the present invention to provide an irradiation field recognition device that can accurately recognize an irradiation field, and to provide an image processing device that can accurately determine processing conditions based on image data based on the recognition of the irradiation field.

[0013]

[Means for Solving the Problems] Therefore, the radiation field recognition device according to the present invention recognizes a rectangular radiation field of a radiation image formed in accordance with the amount of radiation transmitted through each part of a subject. This recognition device is configured as shown in FIG. In FIG. 1, a band-shaped area setting means sets a plurality of band-shaped areas parallel to a specific axis along one side of a rectangular irradiation field, and a representative profile detection unit sets a plurality of band-shaped areas parallel to a specific axis along one side of a rectangular irradiation field, and a representative profile detection unit is configured to set a plurality of band-shaped areas parallel to a specific axis along one side of a rectangular irradiation field, and a representative profile detection unit sets a plurality of band-shaped areas parallel to a specific axis along one side of a rectangular irradiation field. A representative profile corresponding to a change in image data level is detected for each of the plurality of band-shaped areas. And the irradiation field recognition means is
A difference profile between the representative profiles of each of the plurality of band-shaped regions is determined, and a boundary line surrounding the irradiation field perpendicular to the specific axis is determined based on a change in the difference profile in the specific axis direction.

Further, as shown in FIG. 1, the radiation image processing device according to the present invention can process radiation field data for extracting only image data included in the radiation field recognized based on the radiation field recognition device according to the present invention. The image data processing means includes an extraction means and an image data processing means for processing the data using processing conditions determined based only on the extracted image data to obtain final radiographic image data.

[0015]

[Operation] According to this configuration, a plurality of band-like areas parallel to a specific axis along one side of the rectangular irradiation field are set, and each of the plurality of band-like areas corresponds to a change in image data level in the direction of the specific axis. Profile is detected. Then, a differential profile between the representative profiles is determined, and a boundary line surrounding the irradiation field perpendicular to the specific axis is recognized based on the change in this differential profile in the specific axis direction.

The difference profile between the representative profiles appears as a difference in radiation transmittance in each part of the object included in each band-shaped region, but the dose level leaking to the outside of the irradiation field due to the influence of scattered radiation within the object is It is not affected much by the radiation transmittance of the subject, and the range of variation in dose level is narrow. Therefore, if there is a clear difference in the radiation transmittance of the subject parts included in each of the plurality of band-shaped regions, if it is image data within the irradiation field, it will appear as a difference between the representative profiles, but if it is outside the irradiation field, it will appear as a difference between the representative profiles. In this case, the influence of transmittance is lost and the difference between the representative profiles becomes unclear. Therefore, the boundary line surrounding the irradiation field perpendicular to the specific axis can be recognized by observing the change in the difference profile in the specific axis direction.

[0017] Here, only the image data within the irradiation field recognized in this way is extracted, and the data is processed using processing conditions determined based only on the extracted image data to obtain final image data. If this is done, processing conditions will be determined including image data outside the irradiation field, that is, image data unrelated to the transmitted dose level of each part of the subject, and inappropriate image processing will be performed. You can prevent this from happening.

[0018]

[Examples] Examples of the present invention will be described below. FIG. 2, which shows one embodiment, is a block diagram showing the system configuration of a radiation image information recording and reading device including a radiation field recognition device and a radiation image processing device according to the present invention. An example of application to radiography is shown below.

Here, the radiation source 1 is controlled by a radiation control device 2 and irradiates radiation (generally X-rays) toward a subject M (such as a human chest). The recording/reading device 3 is equipped with a radiation image conversion panel 4 on a surface facing the radiation source 1 with the subject M in between.
stores energy in the stimulable phosphor layer according to the radiation transmittance distribution of the subject M with respect to the radiation dose irradiated from the radiation source 1, and forms a latent image of the subject M therein. In the above system, in order to prevent radiation from being irradiated to areas not necessary for diagnosis, a collimator (not shown) or the like is used to narrow down the radiation field to a rectangular shape during imaging.

The conversion panel 4 has a stimulable phosphor layer provided on a support by vapor phase deposition of the stimulable phosphor or coating of a stimulable phosphor paint, and the stimulable phosphor layer are shielded or coated with protective members to prevent adverse effects and damage from the environment. In addition, as the stimulable phosphor material, for example, Japanese Patent Application Laid-Open No. 61-72091,
Alternatively, a material as disclosed in Japanese Patent Application Laid-open No. 75200/1983 may be used.

The light beam generator (gas laser, solid laser, semiconductor laser, etc.) 5 generates a light beam whose emission intensity is controlled, and the light beam reaches the scanner 6 via various optical systems. There, the light is deflected, and the optical path is further deflected by the reflecting mirror 7, and the light is guided to the conversion panel 4 as stimulated excitation scanning light. The condenser 8 has a condensing end made of an optical fiber bundle located close to the conversion panel 4 scanned with the stimulated excitation light, and has a condensing end that is proportional to the latent image energy from the conversion panel 4 scanned by the light beam. receives stimulated luminescence with a luminescence intensity of 9 is a filter that allows only light in the stimulated emission wavelength range to pass from the light introduced from the condenser 8;
The light that has passed through is incident on the photomultiplier 10 and is photoelectrically converted into a current signal corresponding to the incident light.

The output current from the photomultiplier 10 is current/
The voltage signal is converted into a voltage signal by a voltage converter 11, amplified by an amplifier 12, and then converted into digital data (digital radiation image signal) by an A/D converter 13. Then, this digital data is sequentially subjected to image processing in the image processing device 14, and the image data after the image processing is transferred to the interface 1.
6 to the printer 17.

Reference numeral 15 denotes a CPU that controls image processing in the image processing device 14, and performs various image processing including gradation processing (for example, frequency processing) on the digital radiation image data output from the A/D converter 13. , enlargement, reduction, movement, rotation, statistical processing, etc.) in the image processing device 14 to form a form suitable for diagnosis, and then output to the printer 17 so that the printer 17 can obtain a hard copy.

Note that what is connected via the interface 16 may be a monitor such as a CRT, or may further be a storage device (filing system) such as a semiconductor storage device. Reference numeral 18 denotes a reading gain adjustment circuit, and this reading gain adjustment circuit 18 adjusts the light beam intensity of the light beam generating section 5, adjusts the gain of the photomultiple 10 by adjusting the power supply voltage of the high voltage power supply 19 for the photomultiple, and the current/voltage converter. 11 and amplifier 12 gain adjustment, and A/D converter 1
The input dynamic range of step 3 is adjusted, and the reading gain of the radiation image signal is comprehensively adjusted.

The image processing device 14 is configured to include the functions of a radiation field recognition device and a radiation image processing device, and is configured to perform radiation field aperture, perform imaging, and generate digital radiation obtained from the recording/reading device 3. Recognize the irradiation field based on the image signal. Then, only image data within the recognized irradiation field is extracted, gradation processing conditions are determined based on the extracted image data, and based on the determined processing conditions, the entire digital radiation image signal or the Only the extracted image data is subjected to gradation processing and output to the printer 17.

Here, according to FIG.
The specific configuration of No. 4 will be explained. First, the original digital radiation image signal output from the recording/reading device 3 is input to the band-shaped area setting section 20 as a band-shaped area setting means, and here, a plurality of signals parallel to a specific axis along the contour of the rectangular irradiation field are A strip area is set. An averaged profile detection unit 21 serving as a representative profile detection unit indicates the average change in image data level along the specific axis direction for each of the plurality of band-shaped areas set by the band-shaped area setting unit 20. An averaged profile (corresponding to the representative profile in this embodiment) is detected (see FIGS. 6 to 8).

Next, the averaged profile deviation calculation section 22 calculates the averaged profile (representative profile) for each strip area detected by the averaged profile detection section 21.
Calculate mutual deviation (difference profile). and,
The irradiation field determining unit 23 determines the irradiation field as a straight line orthogonal to the specific axis based on the change in the image data deviation (difference profile) between the band-shaped regions obtained as described above in the specific axis direction. Find the enclosing boundary line. still,
In this embodiment, the function as an irradiation field recognition means is that of the averaged profile deviation calculation section 22 and the irradiation field determination section 2.
3.

When the irradiation field is determined by the irradiation field determination section 23, the image data extraction section 24 as an irradiation field data extraction means
Then, only the image data included in the determined irradiation field is extracted from the original digital radiation image signal output from the recording/reading device 3, and the processing condition determining unit 25 determines the image data based only on this extracted image data. Gradation processing conditions are determined. Then, the gradation processing unit 26 serving as an image data processing unit processes the image data (or the entire image signal) within the irradiation field extracted by the image data extraction unit 24 under the processing conditions determined by the processing condition determination unit 25. gradation processing is performed based on the gradation processing, and the image data after the gradation processing is output to the printer 17.

The printer 17 is configured as shown in FIG. 4, for example. In the printer 17 shown in FIG. 4, the digital radiation image signal read out via the interface 16 is first subjected to various signal correction processes in the signal correction circuit 31 via the buffer memory 30, and then to the D/A A converter 32 converts the digital signal into an analog signal. Then, in order to modulate the laser beam according to this analog signal, the output of the D/A converter 32 is sent to the modulator drive circuit 3.
3, and this modulator drive circuit 33 outputs a drive voltage corresponding to the output level of the D/A converter 32 to the optical modulator 34.

The optical modulator 34 modulates the laser light emitted from the laser light source 35 according to the image signal level based on the drive voltage, and the modulated laser light is transmitted to a deflection mirror (not shown) rotated by a motor (not shown). The light is reflected by the polygonal reflecting surface of the polygon mirror 36 and distributed in the main scanning direction. Note that a galvanometer mirror may be used as the deflection mirror.

The reflected light from the deflection mirror 36 has fθ
A recording medium (photosensitive material) 3 on which the scanning light is conveyed in the sub-scanning direction after passing through a lens 37 and being adjusted to a constant scanning speed.
8, a two-dimensional radiation image is recorded on the recording medium 38, and then the recording medium 38 is developed to obtain a hard copy of the radiation image.

Next, the irradiation field recognition and gradation processing (see FIG. 3) according to the present invention is performed in the image processing device 14 (see FIG.
Data processing) will be explained in detail. First, the band-shaped area setting unit 20 first thins out the original image data of, for example, about 2048 x 2464 pixels output from the recording/reading device 3 to about 128 x 154 pixels, and then performs the band-shaped area setting calculation described later. make it easier.

Next, the horizontal axis of the image area is the X axis, and the vertical axis is the Y axis. In this embodiment, as shown in FIG. Select the Y-axis, and add three regions parallel to the Y-axis, two belt-like regions A1 and A3 each containing the lung field, and a belt-like region A2 including the spine, to the thinned image data. Set based on.

[0034] In the irradiation field recognition according to the present invention, as will be described later, the difference in image data level due to the difference in radiation transmittance of the subject included in each of a plurality of strip areas changes inside and outside the irradiation field. This is used to recognize the irradiation field. Therefore, it is desirable that there is a difference in transmittance between the parts of the subject included in each band-shaped area, and that there is a large deviation when comparing image data between band-shaped areas. Since the transmittance differs greatly between the lungs and the spine, which are present in the lungs and the spine, multiple band-shaped areas are set in the direction parallel to the Y axis (vertical axis). This is particularly effective in recognizing the irradiation field when narrowing down the irradiation range in the vertical direction.

To specifically explain the setting of the band-shaped areas A1, A2, and A3, first, as shown in FIG.
The projection value PRJ(x), which is the cumulative value of the image data in the Y-axis direction, is calculated, and the projection level (passing radiation level) is found to be the minimum in the central part when the image area is divided into three equal parts in the X-axis direction. Let the point be the column XC of the midline. That is, in the case of chest imaging in this example,
Since the spine part with low transmittance is located between the lungs with high transmittance, the image data in the central part has the minimum value PC.
It is predicted that the point approximately indicates the center of the spine.

Next, in the right and left 1/3 columns of the entire image, the image data is moved from the center to the outside, and the projection value at each moving point is compared with threshold values TR and TL, which will be described later. do. The positions where the projection value first becomes equal to or lower than the thresholds TR and TL are regarded as the boundaries between the lung field and the flesh, and are determined as the left end XR and right end XL of the region surrounding both lungs, respectively.

The threshold values TR and TL are determined by the following equations based on the maximum values PRX and PLX of the projection values in the right and left 1/3 columns, respectively, and the minimum value PC. TL={(K1-1)×PLX+PC}/K1TR={
(K2-1)×PRX+PC}/K2 In the above equation, for example, K1=K2=5.

Through the above processing, both ends of the lung field in the X-axis direction are defined, and the position of the spine is also estimated from the midline column XC. A predetermined width XC±W (predetermined value) is set as the central region A2 including the spine, and the remaining side portions are defined as regions A1 and A3 including the lungs, respectively. The predetermined value W can be, for example, a value obtained by dividing the maximum image width X in the X-axis direction by a predetermined value (for example, about 12).

Alternatively, the area sandwiched between the column where the projection value is equal to TR between XR and XC and the column where the projection value is equal to TL between XC and XL is set as A2, and the remaining both sides The portions may be respectively A1 and A3. Note that the setting of band-shaped areas is not limited to the method or the number of areas described above, but it should be noted that the subjects (human bodies) overlap and the radiation transmittance of each part of the subject included in each band-shaped area is clearly visible. There is a difference,
It is only necessary that a large difference in image level appears when the band-shaped areas are compared with each other. Therefore, instead of anatomically specifying a region as described above, multiple band-like regions may be specified in advance, and a straight line in the Y-axis direction perpendicular to one point in the X-axis direction may be It is also possible to have a band-shaped region, for example, the maximum value PRX (or P
Two straight lines in the Y-axis direction, passing through the point on the X-axis where the minimum value PC was obtained and a straight line in the Y-axis direction passing through the point on the X-axis where the minimum value PC was obtained, may be set as a band-shaped area. .

[0040] As another embodiment in which the band-shaped region is set based on image data, only the midline XC described above is detected;
For example, if V=X/4, W=X/12, (XC-V) ~
Let (XC-W) be a band-shaped region A1 including the lungs, and (X
CW) to (XC+W) may be defined as a band-shaped region A2 including the spine, and (XC+W) to (XC+V) may be defined as a band-shaped region A3 including the lungs. Further, in setting the band-shaped area as described above, XC can be simply set as the center of the image (X/2).

Furthermore, in the above embodiment, the original image data was thinned out to provide image data for recognizing the irradiation field, but it is possible to thin out the original image data to obtain image data for recognizing the irradiation field. The image data thus obtained may be used as data for irradiation field recognition. The above-mentioned "pre-reading" means that, prior to the "main reading" for obtaining a visible image for diagnosis, excitation light of a lower level than the excitation light used for this "main reading" is stored and recorded in the conversion panel 4 in advance. It reads the outline of image information.

Once the band-shaped regions are determined, the next averaging profile detection unit 21 calculates the average in the Y-axis (specific axis) direction for each region A1, A2, and A3, as shown in FIGS. 6 to 8. profile P1(y), P2(y), P3(y)(
A representative profile of each band-shaped region in this embodiment) is calculated. That is, the average value of the image data on the X-axis perpendicular to the specific axis Y-axis is determined for each strip-shaped region, and an averaging profile is set as a function of the average value, which is determined in a number equal to the number of pixels in the Y-axis direction. In this embodiment, this averaged profile corresponds to a representative profile corresponding to a change in image data level in a specific axis direction.

Next, the irradiation field determination unit 23 calculates the change in the deviation (difference value) between the image level of the band-shaped region including the lung field and the image level of the band-shaped region including the spine in the Y-axis direction according to the following equation. Ask along. In other words, the difference profile P(y) between the representative profile of the band-shaped region including the lung field and the representative profile of the band-shaped region including the spine is obtained. P(y)={P1(y)+P3(y)}/
2-P2(y), that is, P1(y) and P3(y) are both averaged profiles (representative profiles) of band-shaped areas A1 and A3 including the lung fields, and they are substantially similar (
(See Figures 6 and 7), and the average value is set as the averaged profile in the band-shaped area including the lung field, whereas P2(y) is the averaged profile in area A2 including the spine. Therefore, the above-mentioned P(y) shown in FIG. 9 is calculated based on the data correlating to the difference in radiotransmittance between the lung field and the spine (the difference profile between the representative profile of the lung field and the representative profile of the spine). Become.

[0044] The averaged profile (representative profile) of each of the strip areas A1, A2, A3, . . . An is expressed as P
1(y), P2(y), P3(y),...Pn(y)
When (n≧2), P(y) corresponding to the difference profile between the representative profiles is expressed by a general formula:

[0045]

[Math 1]

[0046]

[Math 2]

[0047] Among the plurality of band-shaped areas, it is preferable that the signs of ai (a is a real number) be made equal for areas in which the level of image data contained therein is similar, such as the above-mentioned A1 and A3. Furthermore, in an image that is approximately line symmetrical, it is preferable that the signs of ai be the same in areas that are symmetrical. That is, regions in which image data levels are similar are regarded as one group, and changes in image data deviation between groups in which image data level differences exist are determined along a specific axis direction.

[0048] Next, the difference profile P(y
), find the maximum value Pmax of
r is a constant, for example, about 0.3)
A point on the axis is set as point Y1 that defines the upper end of the irradiation field. In addition, the values of the difference profile P(y) are examined in reverse order from the maximum value Y to the decreasing direction, and only for the first time P(
The point on the Y axis where y)≧r×Pmax is set as the point Y2 that defines the lower end of the irradiation field.

[0049] By drawing a straight line parallel to the X axis through each of Y1 and Y2, the irradiation field narrowed in the vertical direction within the entire image area is recognized as the range surrounded by the straight line. Ru. In other words, as in this example, when the subject overlaps the boundary of the irradiation field, the dose level (image data) outside the irradiation field will decrease due to the influence of scattered radiation within the subject.
However, unlike transmitted radiation, scattered radiation has low directivity and is emitted from various parts of the subject in various directions and reaches the conversion panel, so the dose level reached is based on the dose level of the subject that overlaps the boundary. It is not affected by variations in transmittance and remains at approximately the same level regardless of whether it is a lung field or muscle area other than bone, or a spine area. Therefore, within the irradiation field, there will be a large difference in the corresponding image level between the part including the lung field and the part of the spine, but outside the irradiation field, the level difference will attenuate for the reasons mentioned above. . Therefore, the difference profile P(y), which indicates the level difference of image data between the part including the lung field and the spine part in the Y-axis direction, is observed, and the point where the P(y) falls is the irradiation field. This indicates the boundaries of the

As described above, according to this embodiment, the difference in image data based on the difference in transmittance of each part of the object within the irradiation field, rather than the change in the absolute level of the image data, is determined when the image data is outside the irradiation field. The irradiation field is recognized by utilizing the attenuation (the level of the difference profile P(y) is small and its fluctuation is small outside the irradiation field).
The greater the variation in transmittance in each part of the subject, the more accurately the irradiation field can be recognized from image data affected by scattered radiation without being affected by variations in the body shape of the subject.

[0051] In the boundary recognition of the irradiation field based on the difference profile P(y), the point on the Y axis where the first derivative value of P(y) has a sharp peak is taken as the point indicating the boundary of the irradiation field. It may be detected. Alternatively, the absolute value of the first-order differential value of the difference profile P(y) may be sequentially checked from the outside of the image area toward the inside, and the point on the Y axis at the time when it exceeds a predetermined threshold value may be set as the boundary. can. Furthermore, it is also possible to recognize boundaries using a combination of these.

Furthermore, in this embodiment, the average value of image data on a straight line perpendicular to the specific axis in each band-shaped area is used as a representative profile corresponding to the change in the image data level in the specific axis direction in each band-shaped area. Although a profile with a value is obtained, in addition to the average value, a maximum value, a minimum value, a median value, etc. can also be used as the representative value. A number of representative values equal to the number of pixels in the specific axis direction is obtained, and a function corresponding to the position of the representative value in the specific axis direction is used to determine how image data changes along the specific axis direction in the band-shaped region. Since it is a representative profile shown in the figure, it is called a representative profile.

When using the maximum and minimum values as the representative values, for example, in areas where the image data included is relatively high, such as areas A1 and A3 in the above embodiment, multiple pieces of image data exist in the X-axis direction. On the other hand, in a region where the image data included is relatively low, such as the region A2, the minimum value among the plurality of image data in the X-axis direction is sampled. For example, the difference in image data between the two regions can be further emphasized, and the attenuation of the difference when exiting from inside the irradiation field to outside the irradiation field can be accurately captured.

[0054] For this reason, when setting the band-shaped area, the projection and profile are created as described above.
There is no need to set up individual regions; instead, you can set multiple band-like regions in advance that are predicted to overlap with the subject, and use the image data to determine if areas with high transmittance, such as the lungs, are included based on the actual image data. The predicted region and the region predicted to include a region with low transmittance such as the spine may be selected, and the change in image data deviation between the selected regions may be determined in the Y-axis direction. .

Furthermore, the image data in the area A1 and the area A3 are compared in the X-axis direction, the higher data is sampled, and the deviation (difference) between the sampled image data and the image data in the A2 area is calculated. profile) may be calculated. Although this embodiment is an irradiation field recognition method when the irradiation range is narrowed down in the vertical direction, it can be similarly applied to a case where the irradiation range is narrowed down in the horizontal direction. To briefly explain one example, as shown in FIG. 12, a projection PRJ that accumulates image data in the X-axis direction
(y), and find a line YT where the projection level is minimum in the upper 1/2 region. y=0~Y
Let the region of T be a belt-like region B1 including the shoulder bone, and y=YT~
Let YT+V (V is a predetermined value) be a band-shaped region B2 including the lung field. The predetermined value V is, for example, about V=Y/3.

Next, an averaged profile (representative profile) P1(x
) and P2(x), P(x)=P2(x)−P
1(x). Then, the P(x) corresponding to the difference profile may be examined in the X-axis direction in the same manner as in the above embodiment. When the irradiation field is recognized as described above, the image data extraction unit 24 extracts only the image data included within the recognized irradiation field from the original image signal. Then, the processing condition determining section 25 determines conditions for tone processing in the tone processing section 26 based on the extracted image data. The gradation processing makes the density of the region of interest (the region including the image part necessary for medical diagnosis) in the reproduced image of the radiological image constant,
It also processes signals to make it easier to see the structures of the human body and the shadows of lesions.

The processing condition determining section 25 can determine the gradation processing conditions using a cumulative histogram as disclosed in, for example, Japanese Patent Laid-Open No. 63-31641. That is, as shown in FIG. 10, a histogram and a cumulative histogram are created based on the image data in the irradiation field extracted by the image data extraction unit 24, and when the cumulative histogram reaches a predetermined value (for example, 50%), The signal value of is set as the characteristic value Pc, and both the magnitude of this Pc is set within a predetermined range +a.
1, -a2 and this range (Pc+a1 to Pc
-a2) is set as the reference image signal range. Then, this reference image signal range is made to correspond to the input signal range (Qmax to Qmin) to an image reproducing device such as the printer 17, or from a predetermined value (for example, 5%) of the cumulative histogram to a predetermined value (for example, 95%). It is also possible to set the range up to the reference image signal range.

In addition, a nonlinear basic gradation processing table having the most preferable gradation for each photographing condition is prepared in advance, and the reference image signal level Qdef on the table is
By moving the table in parallel in the level direction of the input signal value so as to correspond to the signal value Pc at which the cumulative histogram reaches a predetermined value (for example, 50%),
A new table for determining gradation processing conditions can also be obtained. Alternatively, as disclosed in Japanese Patent Application Laid-open No. 59-83149, one may be selected from a plurality of preset basic gradation conversion curves and used by subjecting it to rotation and translation.

In addition to the method of using the cumulative histogram described above, methods for determining the gradation processing conditions include the maximum value, minimum value,
The gradation processing conditions can also be determined using the average value and median value. Here, in this embodiment, only the image data within the irradiation field is extracted from the image data obtained by imaging with irradiation field aperture, and the gradation processing conditions are determined based on the extracted image data. It is possible to prevent the setting of processing conditions from being adversely affected by originally unnecessary image data outside the irradiation field, and by setting appropriate processing conditions, it is possible to perform gradation processing that can contribute to improving diagnostic performance.

It should be noted that as the image data used for determining the gradation processing conditions, the image data extracted by the image data extracting section 24 is thinned out and used from the viewpoint of shortening calculation time and saving memory capacity. is also good, and in this case also the maximum value,
The minimum value, average value, and cumulative frequency distribution have almost no difference from those of the original image data. The preferred density range or brightness range of a reproduced image that has been subjected to gradation processing varies depending on the part of the image and the shooting conditions, but for example, in the case of hard copies, the lowest density range is from film fog density to (fog density + 0.3). is preferred, with a maximum concentration of 2
．． The range of 0 to 3.5 is preferable. In addition, particularly in chest images, the lung field is the most important region for diagnosis, and the highest density in the lung field is preferably in the range of 1.3 to 2.4, and preferably in the range of 1.6 to 2.2. is particularly preferred. That is, as the gradation processing conditions for the chest image, it is preferable to use, for example, a smooth nonlinear gradation processing table created so as to satisfy the three conditions of maximum density, minimum density, and maximum lung field density.

Once the gradation processing conditions are determined as described above, the gradation processing section 26 performs gradation processing on the entire image data or only the image data extracted by the image data extraction section 24 to obtain the final result. The final image data is output to the printer 17 to obtain a hard copy. In this example, the gradation processing conditions were determined as processing conditions based on the image data extracted as being included in the irradiation field, but in addition to this, the processing conditions for other image processing such as frequency processing In this case as well, the frequency processing conditions are determined based on the image data included in the recognized irradiation field, and the image data within the irradiation field or the entire image data is subjected to frequency processing according to such determination. That's fine.

Further, the image data extracted by the image data extracting section 24 may be stored in the filing device as it is, and then subjected to various image processing at the time of playback, and then played back on a CRT or as a hard copy. . In addition, in this embodiment, since the image data extraction unit 24 extracts only the image data within the irradiation field, unnecessary image data obtained when imaging with irradiation field aperture is removed, thereby reducing image processing time and memory. This also has the effect of reducing capacity.

Furthermore, in this embodiment, an example of chest imaging of a human body has been described, but the subject is not limited to the subject. For example, when the width in the vertical direction is narrowed down by the irradiation field aperture, the width in the horizontal direction may be If there is a change in transmittance, it is possible to recognize the boundaries between the upper and lower irradiation fields. In this example,
Although a system for obtaining digital radiation images using a stimulable phosphor was used, a system for obtaining digital radiation image signals by photoelectrically converting the transmitted light of a silver halide film on which a radiation image is recorded may also be used. The configuration for obtaining image signals is not limited.

[0064]

[Effects of the Invention] As explained above, according to the present invention, even when objects overlap the boundaries of the irradiation field and scattered radiation within the object affects the irradiation field, variations in individual objects can be avoided. In addition to being able to accurately recognize the irradiation field without being affected, only the image data included within the recognized irradiation field is extracted, and the data is processed using processing conditions determined only based on this extracted image data. Since the final radiation image data is obtained through processing, it is possible to prevent inappropriate setting of processing conditions due to the influence of image data outside the irradiation field.For example, in medical radiography, diagnostic performance can be improved. The effect is that it can be improved.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of the present invention.

FIG. 2 is a system schematic diagram showing an embodiment of the present invention.

FIG. 3 is a block diagram showing a detailed configuration of the image processing device shown in FIG. 2;

FIG. 4 is a system diagram showing the detailed configuration of the printer shown in FIG. 2;

FIG. 5 is a diagram showing how a band-shaped area is set in the embodiment.

FIG. 6 is a diagram showing an averaged profile (representative profile) of the A1 region shown in FIG. 5;

FIG. 7 is a diagram showing an averaged profile (representative profile) of area A3 shown in FIG. 5;

FIG. 8 is a diagram showing an averaged profile (representative profile) of area A2 shown in FIG. 5;

FIG. 9 is a diagram showing deviations (difference profiles) between the averaged profiles shown in FIGS. 6 to 8;

FIG. 10 is a diagram showing an example of determining gradation processing conditions.

FIG. 11 is a diagram showing an example of conventional irradiation field recognition.

FIG. 12 is a diagram showing how erroneous recognition occurs due to scattered radiation in conventional irradiation field recognition.

[Explanation of symbols]

1 Radiation source 2 Radiation control device 3 Recording/reading device 4 Radiation image conversion panel 14 Image processing device 17 Printer 20 Strip area setting section 21 Averaged profile detection section 22 Averaged profile deviation calculation section 23 Irradiation field determination section 24 Image data extraction section 25 Processing condition determining section 26 Gradation processing section

Claims (2)

[Claims] Claim 1: A radiation field recognition device that recognizes a rectangular radiation field of a radiation image formed in accordance with the amount of radiation transmitted through each part of a subject, the radiation field recognition device comprising: a strip-shaped area setting means for setting a plurality of strip-shaped areas parallel to the strip-shaped areas; and a representative profile that detects, for each of the plurality of strip-shaped areas, a representative profile corresponding to a change in the image data level in the specific axis direction based on pixel data of the strip-shaped areas. and a profile detection means, which determines a difference profile between the representative profiles of each of the plurality of band-shaped regions, and determines a boundary line surrounding the irradiation field perpendicular to the specific axis based on a change in the difference profile in the specific axis direction. A radiation field recognition device comprising: radiation field recognition means for determining a radiation field. 2. Radiation field data extraction means for extracting only image data included in the radiation field recognized based on the radiation field recognition device according to claim 1, and image data extracted by the radiation field data extraction means. 1. A radiation image processing apparatus comprising: image data processing means for processing data using processing conditions determined based solely on the data and obtaining final radiation image data.