JPH057579A - Irradiation area extractor for radiation image - Google Patents

Abstract

PURPOSE:To extract an irradiation area handily and accurately from a radiation image taken with the reduction of an irradiation field. CONSTITUTION:After the trimming of an original digital radiation image signal (S1) an image area is divided into small parts (S2). Here, a dispersion value of an image signal contained is calculated from each small area (S3) and columns and lines of the small areas in which the small areas with the dispersion values exceeding a specified value are contained in number larger than a specified number are recognized as candidates specifying the contour of the irradiation field. (S4) Then, the correctness or error in the recognition of the contour is judged depending on whether a representative value of an image signal in a column and line one column (line) outer side of those selected as candidate indicates a value corresponding to an area of stopping the irradiation field. (S5) When the recognition is correct, those inside the candidate column and line are extracted as area of irradiation field. (S6)

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a radiation image irradiation field region extracting apparatus, and more particularly to a device for extracting only a region portion of the irradiation field from a radiation image photographed by narrowing the irradiation field.

[0002]

2. Description of the Related Art Radiation images such as X-ray images are often used for diagnosing diseases, etc. In order to obtain this X-ray image, X-rays transmitted through a subject are applied to a phosphor layer (fluorescent screen). Conventionally, so-called radiography, in which visible light is generated by irradiation, and the visible light is applied to a film using a silver salt to develop the film, which is the same as in ordinary photography, has been conventionally used.

However, in recent years, a method for directly taking out an image from a phosphor layer has been devised without using a film coated with a silver salt. For this method,
Radiation that has passed through the subject is absorbed by the phosphor, and then the phosphor is excited by, for example, light or thermal energy to cause the phosphor to emit the radiation energy accumulated by the absorption as fluorescence. There is a method of obtaining an image signal by performing photoelectric conversion on the. The radiation image signal obtained in this way is visualized as it is or after being subjected to image processing and output to a printer, CRT or the like, but it may be digitized for image processing by a computer. Many.

[0004]

By the way, in radiography, there are humanitarian reasons to reduce the radiation-exposed portion as much as possible and diagnostic reasons such as preventing scattered radiation from a portion unrelated to the diagnosis. In some cases, a radiation non-transparent material such as a lead plate is installed on a part of the subject or a radiation source to perform an irradiation field stop that limits the irradiation field of the radiation to the subject.

On the other hand, in the image processing (gradation processing, spatial frequency processing, etc.) of the digital radiation image, the processing condition is determined from the statistical properties of the image data (maximum value / minimum value of data / histogram), Although the image processing is performed to finish the area so that it is easy to see, when the irradiation field is narrowed as described above, the statistical property of the image data indicates the portion of the irradiation field (the portion where radiation was not directly irradiated). The data of 1) influences and shows the property that the radiation transmission amount is generally small, and there is a problem that the image processing conditions cannot be set appropriately.

For example, as shown in FIG. 9, a lung field region, which is a region of interest in a chest radiographic image, is recognized by using the maximum and minimum values in the vertical and horizontal projections, and the image in the lung field region is recognized. It has been performed to determine the gradation processing condition based on the histogram of data or the cumulative histogram, but as shown in FIG. 10, when a part where the irradiation field is narrowed down enters the lower part of the image, the projection covers the narrowed part. Since it is created by including it, the lung field region cannot be correctly recognized, and the lung field region is recognized including the irradiation field stop portion. Therefore, the figure
In the case as shown in 10, if the gradation processing condition is set using the histogram or the cumulative histogram of the image data in the erroneously recognized lung field region, the gradation processing matched to the lower image data as a whole will be performed. As a result, the processed image becomes a whitish image with no delicate contrast.

Therefore, in determining the image processing condition, it is necessary to extract only the irradiation field portion in the image in advance, and to perform the region of interest detection and condition setting based on only the image data in the irradiation field. The following various methods have been proposed as methods for extracting fields. For example, in Japanese Unexamined Patent Publication No. 1-267634, a histogram of the entire image is created, and the image data is binarized with a value that maximizes the degree of separation between the peak in the irradiation field portion and the peak in the irradiation field stop portion. There is disclosed a method of extracting an irradiation field portion by using the above method.
Further, in Japanese Patent Application Laid-Open No. 2-81278, an image is divided into a plurality of small blocks in a grid pattern, various high frequency extraction processing is performed in each small block, and a portion containing a large amount of high frequency components is used as an irradiation field stop region. A method of recognition is disclosed. Further, Japanese Patent Laid-Open No. 63-183434 discloses a method of differentiating profile information of image data to detect the contour portion of the irradiation field area.

However, in the case of the above-described binarization method, the density value in the irradiation field diaphragm often overlaps with the low-concentration region of the human body in the irradiation field, so that even simple binarization is performed. It is difficult to accurately extract only the area of the irradiation field. Also, in the case of the method by high-frequency extraction for each small block, the fact that the area of the irradiation field diaphragm is noisy and there are many minute density changes is used, so at least 2000 × in order to capture ultrafine changes. It is necessary to perform highly accurate analysis using an image of about 2000 pixels, which is not feasible with automatic image processing that requires real-time processing.
Further, in the case of the method of detecting the contour based on the differential value of the profile information, after the contour candidate points of the irradiation field region are obtained for each line (column), the processing of connecting them to obtain the contour line is performed. However, in the work of observing on a pixel-by-pixel basis, inaccurate connection is often performed due to the influence of noise, and there is a problem that reliability is low even if linear approximation by an equation is performed.

The present invention has been made in view of the above problems, and an object of the present invention is to provide a radiation image irradiation field region extraction apparatus capable of accurately extracting the irradiation field region by a relatively simple process.

[0010]

Therefore, a radiation image irradiation field region extraction apparatus according to the present invention performs radiation field narrowing and forms radiation corresponding to the amount of transmission of each portion of the radiation irradiated to the object. An irradiation field area extraction device for an image, which is configured as shown in FIG. In FIG. 1, the small area dividing means divides the image area of the radiation image into a plurality of small areas in the vertical and horizontal directions, and the dispersion value calculating means divides the image data contained in the small area for each of the divided small areas. Calculate the variance.

The contour recognizing means recognizes the vertical and horizontal rows of the small areas including the predetermined number or more of the small areas having the variance value of the predetermined value or more as the contours of the irradiation field, and the irradiation field area extracting means recognizes the contours as the contours. The image area inside the small area sequence is extracted as the irradiation field area. Here, it is preferable that the image data in the small area for which the variance value is calculated by the variance value calculation unit is image data obtained by thinning out the original image data.

Further, the contour recognition of the irradiation field by the contour recognition means is determined as correct or incorrect based on the representative value of the image data in the small area corresponding to the contour or in the small area near the small area corresponding to the contour, and an error is detected. It is preferable to provide a correctness determination means for prohibiting the final contour recognition of the irradiation field based on the small area corresponding to the contour when it is determined that there is. Also,
When a plurality of contours are recognized in the inner and outer directions of the image by the contour recognizing means, an outer contour selecting means is provided for causing the irradiation field area extracting means to extract the irradiation field area based on the outermost contour. May be.

[0013]

When the image area of the radiation image is vertically and horizontally divided into a plurality of small areas, the radiation transmission amount is substantially uniform in the small areas corresponding to the areas where the irradiation of the radiation is blocked by the irradiation field diaphragm. Therefore, the image data is largely biased toward the low transmission amount side and the dispersion value is low. Further, in a small region corresponding to the irradiation field, various structures are present, and the radiation transmission amount varies, so that the dispersion value becomes relatively high. On the other hand, in the small area including the boundary line of the irradiation field stop, both the part of the lowest transmission amount where the irradiation field stop was performed and the part of relatively high transmission amount in the irradiation field are included, The image data will have the highest variance.

Therefore, the small area having the highest dispersion value is predicted to include both the area where the irradiation field is narrowed (the area where radiation is blocked) and the area of the irradiation field. A row of small areas in which a plurality of symbols are present can be recognized as they are, which indicate the contour of the irradiation field. As described above, since the contour that determines the irradiation field is determined by determining the variation in the image data for each small area, the accuracy of contour recognition does not greatly deteriorate even if the thinned image data is used. Therefore, it is preferable to simplify the processing by using the thinned data.

As described above, when a row including a plurality of small areas having a high dispersion value is recognized as the contour of the irradiation field, the dispersion value is relatively high in the irradiation field due to the variation of the transmission amount (for example, the object structure If there is a portion where the variance value is high due to the influence of noise or the like on the contour or the contour of the subject) or the irradiation field stop portion, this portion may be erroneously recognized as the contour of the irradiation field area.

Here, when the contour of the irradiation field is accurately recognized, the area outside the small area corresponding to the irradiation field contour is the irradiation field stop portion, and the small area corresponding to the irradiation field contour. Should include at least the irradiation field stop portion, and should not include the irradiation field stop portion inside the small area corresponding to the irradiation field contour. Therefore, before finally determining the small area corresponding to the contour of the irradiation field based on the variance value,
It is possible to determine whether the representative value of the image data in the small area corresponding to the contour or in the small area near the small area corresponding to the contour matches the characteristics when the true contour recognition is performed. For example, it is possible to avoid erroneously recognizing the contour of the irradiation field.

Further, it is predicted that the erroneous recognition of the contour will occur exclusively due to the variation of the radiation transmission amount in the irradiation field. Therefore, when a plurality of contours are recognized, the outermost contour is judged to be correct. .

[0018]

EXAMPLES Examples of the present invention will be described below. FIG. 2 showing an embodiment is a medical radiation image recording / reading device including a radiation image irradiation field region extracting device according to the present invention, which is applied to radiation imaging of the front of the chest of a human body for medical purposes. is there.

Here, the radiation source 1 is controlled by the radiation control device 2 to irradiate the human chest M (subject) with radiation (generally an X-ray), but in this embodiment, Between the radiation source 1 and the human chest M, a shielding plate 20 formed of a non-radiation transparent material such as a lead plate is provided, and irradiation is performed by using only the opening of the shielding plate 20 as a radiation irradiation field. A radiography is performed by narrowing the field.

The recording / reading device 3 is provided with a radiation image conversion panel 4 on the surface facing the radiation source 1 with the chest M sandwiched therebetween, and the conversion panel 4 is the radiation of the chest M with respect to the radiation dose from the radiation source 1. Energy according to the transmittance distribution is accumulated in the stimulable phosphor layer, and a latent image of the chest M is formed there. The conversion panel 4 is provided with a stimulable phosphor layer on a support by vapor deposition of the stimulable phosphor or coating of a stimulable phosphor coating, and the stimulable phosphor layer is an environment. It is shielded or covered by a protective member in order to block the adverse effects and damages caused by. Examples of the stimulable phosphor material include, for example, JP-A-61-72091,
Alternatively, a material as disclosed in JP-A-59-75200 is used.

The light beam generator (gas laser, solid-state 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. , Where it is deflected, and the reflector 7
The optical path is deflected by and is guided to the conversion panel 4 as stimulated excitation scanning light. The condensing body 8 has a condensing end made of an optical fiber located in the vicinity of the conversion panel 4 scanned by the stimulated excitation light, and is proportional to the latent image energy from the conversion panel 4 scanned by the light beam. The stimulated emission of the emission intensity is received. Reference numeral 9 is a filter that allows only light in the stimulated emission wavelength range from the light introduced from the light collector 8 to pass through. The light that has passed through the filter 9 is incident on the photomultiplier 10 and corresponds to the incident light. Is photoelectrically converted into a generated current signal.

The output current from the photomultiplier 10 is converted into a voltage signal by the current / voltage converter 11, amplified by the amplifier 12, and then digital data (digital proportional to the amount of radiation transmission) by the A / D converter 13. Radiation image signal). The digital image signal proportional to the radiation transmission amount of each part of the subject is sequentially subjected to image processing in the image processing device 14, and the image signal after the image processing is transmitted to the printer 17 via the interface 16. There is.

Reference numeral 15 denotes a CPU for controlling image processing in the image processing apparatus 14, which performs various image processing including gradation processing on digital radiation image data output from the A / D converter 13 (for example, spatial frequency). Processing, enlargement, reduction, movement, rotation, statistical processing, etc.) in the image processing device 14 to make it suitable for diagnosis and then output to the printer 17,
Make a hard copy of the chest radiograph available at the printer 17.

It should be noted that what is connected via the interface 16 may be a monitor such as a CRT, and further a storage device (filing system) such as a semiconductor storage device. Reference numeral 18 denotes a read gain adjusting circuit. The read gain adjusting circuit 18 adjusts the light beam intensity of the light beam generator 5, the gain of the Photomul 10 by adjusting the power supply voltage of the Photomulch high-voltage power supply 19, and the current / voltage converter. The gains of the amplifier 11 and the amplifier 12 are adjusted, and the input dynamic range of the A / D converter 13 is adjusted, so that the read gain of the radiation image signal is comprehensively adjusted.

The image processing device 14 performs image processing of the read radiation image signal. In such image processing, the area of the irradiation field in the image is extracted, and the statistical properties of the image data in the irradiation field are calculated. Alternatively, in order to determine the processing condition based on the statistical properties of the image data in the region of interest obtained from the profile information in the irradiation field, the image processing device 14 performs image processing as shown in the flowchart of FIG. The irradiation field area is extracted before the processing.

Here, according to the flow chart of FIG.
A brief description of the extraction of the irradiation field region is as follows: S1
In step S2, the number of data is reduced by thinning out the digital radiation image signal photoelectrically read from the conversion panel 4, and in step S2, the image area is divided into a plurality of small areas in the vertical and horizontal directions. Then, in S3, the variance value (image signal distribution width) of the image signal included in the region is obtained for each of the small regions divided in S2, and in S4, the irradiation field is calculated from the data of the variance value for each small region. Candidates for rows and columns (in the present embodiment, vertical divisions of an image are rows and horizontal divisions are columns) of a small area that defines the contour of the area (the boundary between the irradiation field area and the aperture area). Ask for.

In step S5, it is determined whether or not the candidate row / column of the small area defining the contour is correct or incorrect, and it is not the contour of the irradiation field but the contour of the human body structure or the human body. When it is determined that there is a high possibility, the candidate row / column determined in S4 is invalidated and S
Return to step 4 so that another line / column is searched for a contour candidate. On the other hand, when it is determined in S5 that the row / column of the small area selected as the candidate in S4 is correct, the contour of the irradiation field is recognized by the row / column of the small area, and the row / column of the small area that defines the contour is recognized. It is considered that the inner area is the irradiation field area and the outer area including the small area defined by the contour is the irradiation field stop area, and the irradiation field area is extracted.

In the present embodiment, the functions of the small area dividing means, the variance value calculating means, the contour recognizing means, the irradiation field area extracting means, the correctness determining means, and the outer contour selecting means are the CPU 1
5. It is assumed that the image processing device 14 is equipped. Next, the extraction of the irradiation field area outlined above will be described in detail. First, regarding the thinning-out processing in S1, the original image of the digital radiation image signal is generally about 2000 × 2000 pixels, but in order to reduce the processing time in the extraction processing of the irradiation field area, it is reduced to about 128 × 128 pixels. Signal is thinned out. As the thinning method, as shown in FIG. 4, a pixel value at one point in the matrix may be used, or an average value in the matrix may be used.

When the signals are thinned out to about 128 × 128 pixels, in the next step S2, the image area is divided vertically and horizontally in units of small areas of about 4 × 4 pixels, and the image is divided into 32 × 32 pixels.
Divide into about a grid (see FIG. 6). In S3, the variance value (signal distribution width) of the image signal of 16 pixels included in each small region of 4 × 4 pixels on the thinned-out data is calculated.

Here, the density distribution (signal distribution) in the small area has characteristics as shown in FIGS. 5A to 5D depending on which part of the image area the small area corresponds to. Become. That is, in the small area as shown in FIG. 5 (A), which corresponds to the unexposed portion in the irradiation field, the high density pixels (high signal values) are uniformly distributed in the small area, and therefore the dispersion value is high. Will be lower. Further, when the small area corresponds to the irradiation field stop portion, since the low-density pixels (low signal values) are uniformly distributed in the small area as shown in FIG. 5B,
In this case as well, the variance value is low. On the other hand, when the small area overlaps the human body, various structures exist,
As shown in FIG. 5C, the image density value (image signal) varies, so that the dispersion value is relatively high. However, in a small area of about 4 × 4 pixels, the density may vary depending on the structure of the structure. There is a small area in which a large variation does not appear in the value (signal value), and a high variance value cannot be obtained in all the small areas overlapping the human body part. Further, when the contour of the irradiation field is included in the small area, as shown in FIG. 5D, a low-density portion corresponding to the irradiation field diaphragm and a portion in the irradiation field exist in one small area. Therefore, the dispersion value is very high.

Therefore, a row (horizontal small area column) or a column (vertical small area column) that includes many small areas having a very high variance value in the row or column direction of an image divided into small areas is , It is likely that the contour of the irradiation field is included. Therefore, S4
Then, a row or a column of a small area having a predetermined number Th2 or more of a small area having a variance value of a predetermined threshold Th1 or more (hereinafter referred to as a contour area) is detected, and the contour of the irradiation field is defined by the row or the column. To be a candidate for

By the way, since the contour area is detected as a small area in which the variance value of the included image signal is equal to or larger than the predetermined threshold Th1, the structure of the human body is not limited to the small area including the contour. In a small area (for example, an area that partially overlaps the lung field as shown in FIG. 7) that overlaps a portion where the signal variation due to the configuration of an object is large, the variance value is the threshold value Th1.
May be detected as a contour region, and the contour of the irradiation field may be erroneously recognized based on the erroneously detected contour region. Therefore, the row or column of the small region selected as a candidate in S5 is Whether or not it really shows the contour of the irradiation field is determined.

That is, the row or column of the small area recognized as the contour of the irradiation field in S4 is only a candidate and has not been confirmed, so it is a candidate row or candidate column as an undetermined element. . Whether or not the candidate row / candidate column is correct / incorrect in S5, that is, the candidate is the contour of the irradiation field area, or the contour of the human body structure or the contour of the human body instead of the contour of the irradiation field area. Whether or not there is a high degree of accuracy is determined as follows.

If the candidate row / candidate column truly shows the contour of the irradiation field, the area outside the candidate row / candidate column is the area of the irradiation field stop, so that it is more preferable than the candidate row / candidate column. The image signals of the pixels included in the outer one row / column should show the level corresponding to the irradiation field stop. Therefore, in S5, a representative value (minimum value, maximum value, average value, etc.) of pixel values included in one row / column outside the candidate row / candidate column of the irradiation field contour is calculated, and the representative value and the predetermined value are determined. Threshold Th
3 is compared, and if the representative value is less than the threshold value Th3, the candidate row / candidate column is considered to be correct, and if the representative value is not less than the threshold value Th3, the candidate row / candidate column. Does not indicate the contour of the irradiation field region, but determines that the contour is likely to be the contour of the human body structure or the human body. Candidate line
When the candidate column is correct, the outermost row / column is the irradiation field narrowed region and has a uniformly low density (low signal value). Therefore, when the representative value is the threshold value Th3 or more, the true value is true. The contour is presumed to be further outside.

If it is determined in S5 that the candidate row / column of the irradiation field contour does not really correspond to the contour of the irradiation field area but is highly likely to be the contour of the human body structure or the human body, the The candidate row / candidate column is invalidated, the process returns to S4 again, and a row / column that further includes a predetermined number of contour regions Th2 or more in the outer row or column is detected, and that row / column defines the contour of the irradiation field.・ Update setting is performed as a candidate column, and until a candidate row / candidate column that is determined to be correct in S5 is detected, S4,
The process of S5 is repeated.

It should be noted that the candidate rows / columns were sequentially detected from the center row and center column of the image as described above, and the row / column of the small area which would eventually become the contour of the irradiation field could not be detected. In this case, it is considered that the irradiation field was not narrowed in that direction. When it is determined in S5 that the candidate row / candidate column of the irradiation field contour is correct, it is finally recognized that the candidate row / candidate column includes the irradiation field contour line, and in S6, the contour is defined. The area inside the rows and columns of the small area (excluding candidate rows and columns) is extracted as the irradiation field area, and the outside including the rows and columns of the small area that defines the contour is defined as the irradiation field narrowed area. exclude.

As described above, the reason why the rows and columns of the small area defining the contour are excluded from the irradiation field area is that the irradiation field area may be deleted, but at least the area of the irradiation field diaphragm is reduced. This is for avoiding even a slight inclusion in the irradiation field area. The irradiation field area extracted as described above is provided for the subsequent setting of image processing conditions,
It is prevented that the processing condition is affected by the signal level corresponding to the area of the irradiation field stop and desired image processing cannot be performed.

For example, when the irradiation field region and the irradiation field stop region are distinguished as shown in FIG. 7, the lung field, which is the region of interest of the chest radiographic image, is determined based on the image signal in the irradiation field region. A region is detected based on the profile information, and image processing conditions such as gradation processing are determined based on the histogram or cumulative histogram of image signals in the detected lung field. The image processing condition may be set from the histogram of all the image signals in the irradiation field area or the cumulative histogram.

According to the extraction of the irradiation field area as described above,
Since it is used after thinning out the image signal, the processing time can be shortened, and the image area is divided into small areas, and the one including the contour line is detected in the small areas. Even if it is performed diagonally, the radiation field region can be extracted by absorbing it to some extent. Furthermore, it is intended to extract the irradiation field region based on a parameter that is a characteristic of the signal dispersion value in a small region, and for example, to extract the irradiation field region by capturing minute features such as spatial frequency. Therefore, even if the thinned image is used for shortening the processing time as described above, the accuracy does not decrease so much, and the irradiation field region can be stably and accurately extracted.

By the way, the correctness / incorrectness determination of the candidate row / candidate column in S5 is performed based on the image signal in the candidate row / candidate column in addition to the signal level in the outer row / column as described above. It can also be done. That is, if the candidate row / column is correct,
The areas of the irradiation field diaphragm should overlap each part in the candidate row, and the signal value corresponding to the area of the irradiation field diaphragm corresponds to the portion (for example, bone) with the smallest radiation transmission amount in the human body. Lower than the signal value to Therefore, when the minimum value of the pixel values included in the candidate row / candidate column is smaller than the value corresponding to the portion of the human body with the least amount of transmission and corresponds to the area of the irradiation field stop, the candidate row / candidate This means that at least the irradiation field stop portion overlaps the arrangement of the small areas set as columns, and the candidate row / candidate column is correct. On the contrary, when the minimum signal value in the candidate row / candidate column is not smaller than the level corresponding to the area of the irradiation field aperture, the candidate row / candidate column is not the contour of the irradiation field area, but the human body structure or human body. It is possible to consider that there is a high possibility that it is a contour of the above, and the row / column judged as the error is invalidated, and a row / column including a predetermined number or more of small areas having a variance value of a predetermined value or more is further outside. look for.

In the above-described embodiment, candidate rows and candidate columns that define the contour of the irradiation field are sequentially detected from the center of the image to the outside, and the contour of the human body structure or the human body is used instead of the contour of the irradiation field region. Only when it is determined that there is a high possibility, we searched for the candidate row / candidate column further outside, but instead of letting the user judge whether the candidate row / candidate column is set correctly, It is also possible to detect all the arrangements of the small areas that are the candidate rows and the candidate columns, and finally select which candidate is correct.

For example, as shown in FIG. 8, when a plurality of candidate rows / columns are detected in the inside / outside direction of the image, the outermost candidate rows / columns are judged to be correct and selected, and the selected ones are selected. An area inside the candidate row / column is extracted as an irradiation field area. In this way, the outside is treated as correct when the variance value is equal to or greater than the predetermined value even though the contour line is not included when the structure of the human body structure is affected. Is always within the irradiation field inside the true contour.

Further, in the area of the irradiation field diaphragm, the variance value of the small area may exceed a predetermined value due to the influence of noise and the like, and the contour of the irradiation field may be erroneously recognized. A representative value (minimum value, maximum value, average value, etc.) of pixel values included in a row / column inside one of the candidate columns is obtained, and the representative value is compared with a predetermined threshold Th4. Threshold T
If h4 or more, the candidate row / candidate column is regarded as an irradiation field contour, and if the representative value is less than the threshold value Th4, the candidate row / candidate column is within the irradiation field narrowed region. It may be determined that there is a high possibility that the recognition error is high based on noise. As described above, the pixel value inside the candidate is discriminated when the irradiation field contour is truly recognized, and the inside is the irradiation field area, and the signal level becomes higher than the diaphragm area. Because it should be.

Further, in the above-mentioned embodiment, the correct line is determined after specifying the candidate row / candidate column of the irradiation field contour. However, in the small area in which the variance value is determined to be equal to or more than a predetermined value,
Alternatively, it is determined whether or not the small area is a part of the contour of the irradiation field by the representative value (minimum value, maximum value, average value, etc.) of the signal values of the small area outside or inside one of the small areas. After that, a row / column in which a predetermined number or more of small areas determined to be a part of the contour are arranged may be recognized as the contour of the irradiation field.
Here, the correctness determination in small area units can be performed in the same manner as the above-described correctness determination in candidate row / candidate column units.

Further, when making an error judgment based on the signal value of a candidate row / candidate column or a small area outside or inside a small area having a variance value of a predetermined value or more, one small area outside or inside is not always required. It is not necessary that the two or more outer or inner small areas be used as neighboring small areas, and the correctness determination may be performed based on the image data in the small areas. still,
In the present embodiment, the radiation image signal image-processed based on the extraction result of the irradiation field area is immediately printed by the printer as described above.
Although it was made to make a hard copy by 17, it can be played back on a CRT, or once stored in a filing system and read out when necessary to make a hard copy.
It may be displayed on the CRT.

When the radiation image signal is stored in the filing system, the image-processed radiation image signal may be stored, but the unprocessed radiation image signal and the extraction result of the irradiation field area (or extraction). The image processing conditions based on the result) may be stored as a pair and the image processing may be performed at the time of reading. Furthermore, in the present embodiment, the irradiation field region is extracted from the radiation image signal photoelectrically read from the stimulable phosphor layer, but it is not limited to image reading using the stimulable phosphor. Instead, it may be a configuration using another one-dimensional or two-dimensional radiation detector.

[0047]

As described above, according to the present invention, the radiation image captured by performing the irradiation field stop is divided into small areas, and the area in which the signal dispersion value is high is the irradiation field area and the irradiation area. Since the irradiation field area is extracted by considering it as including the boundary with the field stop area (outline of the irradiation field area), it is possible to easily extract the irradiation field area with stable accuracy. The irradiation field area can be extracted with a certain degree of tolerance even for the oblique irradiation field diaphragm with respect to the image, and the accuracy can be secured by using the thinned image data. The effect is that it can be shortened.

[Brief description of drawings]

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

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

FIG. 3 is a flowchart showing how the irradiation field area is extracted in the embodiment.

FIG. 4 is a state diagram showing how image signals are thinned out.

FIG. 5 is a diagram showing a difference in density distribution for each small area.

FIG. 6 is a diagram showing a division state in a chest image with an irradiation field diaphragm.

FIG. 7 is a diagram showing an example of correctness of a candidate row.

FIG. 8 is a diagram showing an example in which a plurality of contours are recognized in the inner and outer directions of an image.

FIG. 9 is a diagram showing how a lung field region is recognized by projection.

FIG. 10 is a diagram showing how the lung field area is erroneously recognized by projection.

[Explanation of symbols]

1 Radiation source 4 Radiation image conversion panel 5 Light beam generator 10 Photomaru 14 Image processing device 15 CPU

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁵ Identification code Internal reference number FI Technical indication location H04N 5/238 Z 9187-5C (72) Inventor Hisonori Dono 1 Sakura-cho, Hino-shi, Tokyo Konica Stocks In the company

Claims (4)

[Claims] 1. An irradiation field area extraction device for a radiation image, which is formed in accordance with the amount of transmission of radiation that is applied to the object by subjecting the irradiation field to a subject. A small area dividing means for dividing into a plurality of small areas; a dispersion value calculating means for calculating a dispersion value of image data included in the small area for each of the plurality of small areas divided by the small area dividing means; Contour recognition means for recognizing the vertical and horizontal rows of small areas containing a predetermined number or more of small areas with the variance value calculated by the variance value calculating means being a predetermined value or more, as contours of the irradiation field, and being recognized as contours by the contour recognizing means. A radiation image irradiation field region extraction device, comprising: an irradiation field region extraction unit that extracts an image region inside the small region row as an irradiation field region. 2. An irradiation field of a radiation image according to claim 1, wherein the image data in the small area for which the variance value calculation means calculates the variance value is image data obtained by thinning out the original image data. Area extraction device. 3. The contour recognition of the irradiation field in the contour recognition means is judged as correct or incorrect based on a representative value of image data in a small area corresponding to the contour or in a small area near the small area corresponding to the contour, and an error is made. 3. The radiation image irradiation according to claim 1, further comprising: a correctness determination unit that prohibits final contour recognition based on the small area corresponding to the contour when it is determined to be true. Field extraction device. 4. An outer contour selection for causing the irradiation field area extraction means to extract the irradiation field area based on the outermost contour when the contour recognition means recognizes a plurality of contours in the inner and outer directions of the image. A radiation image irradiation field region extraction apparatus according to claim 1 or 2, further comprising means.