JP2005065992A - Image flaw classifying method, apparatus therefor and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To classify the image flaw on an X-ray image by the occurring origin. <P>SOLUTION: The image flaws on two X-ray images are detected on the basis of two image data showing two X-ray images photographed by a photographing system, which is composed of an X-ray generator 1, an accumulative phosphor sheet (IP) 2, a cassette 3 for housing IP 2 and a reader 4 for reading the IP 2 on the basis of the direction of the cassette to form image data, in a state that the housing direction of IP 2 is changed by 180°. The image flaw is classified so that the image flaws, which almost coincide with each other when two images are superposed one upon another as they are, are caused by the cassette, the image flaws which almost coincide with each other when one of the image is rotated by 180° to be superposed on the other image, are caused by IP and other image flaws are caused by the reader. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an image defect classification method and apparatus for classifying image defects on an X-ray image according to the generation source based on X-ray image data, and a program therefor.

  Conventionally, X-rays transmitted through a subject such as a human body are detected, image data representing an X-ray image of the subject is acquired, and based on the acquired image data, the X-ray image represented by the image data is converted to a CRT or the like. Displaying on a screen for observation, and analyzing an X-ray image using a computer are performed.

  The image data representing the X-ray image is usually an X-ray source that irradiates the subject with X-rays, and a detection medium that detects the X-rays transmitted through the subject and obtains X-ray image information of the subject (stores X-ray image information). Based on the X-ray image information obtained from the detection medium and the storage phosphor sheet to be recorded, the flat panel detector that outputs the X-ray image information as an electric signal, and the like, the image data representing the X-ray image of the subject is generated Image data generating means (reading device or flat panel that scans the stimulable phosphor sheet with excitation light to generate stimulated emission light and photoelectrically reads the obtained stimulated emission light to generate image data) An imaging member (a cassette for storing a stimulable phosphor sheet) interposed between an X-ray source and a detection medium and a conversion device that A / D converts an electrical signal output from the detector to generate image data , Accumulation Acquired by an imaging system including a phosphor sheet or a built-in type imaging apparatus with a built-in flat panel detector). If dust adheres to the X-ray detection surface of the medium or if noise enters the transmission path of the X-ray image information in the image data generation means, dust or dirt will appear on the X-ray image represented by the acquired image data. Shadows such as points and lines due to noise, that is, image defects occur.

  This image defect becomes an obstacle to observation or inspection of the subject in the image. Especially, in the image diagnosis in the medical field, not only the image information of the image defect part is lost, but also the image defect is mistaken as an abnormal shadow. Therefore, the occurrence of image defects is an important problem.

Therefore, various methods for detecting an image defect and a method for correcting the same have been proposed.For example, the stimulable phosphor sheet on which image information is accumulated and recorded is scanned with excitation light to generate stimulated emission light, In a reading device that photoelectrically reads the obtained photostimulated luminescence light and generates image data, the reflected light is monitored when scanning with excitation light, and dust is detected where the amount of reflected light falls below a threshold value. A method for detecting an adhering place (see Patent Document 1), a method for removing a high-frequency component from a solid image taken without a subject, and detecting an image defect from the image after the removal (see Patent Document 2) ), A method of correcting a defective pixel from an average pixel value of surrounding pixels (see Patent Document 3), and the like are known.
Patent registration 2589195 JP-A-10-282596 JP 2003-23572 A

  By the way, when an image defect is detected, correcting the defect is effective in suppressing adverse effects due to the image defect, but it is most important to eliminate the cause of the image defect itself, and therefore It is necessary to identify the source of the image defect.

  However, in the conventional method as described above, even if an image defect can be detected or corrected, the source of each detected image defect cannot be specified, and the user cannot Estimate the source based on intuition, experience, etc. and then remove the cause, or perform maintenance on all possible sources of image defects without specifying the source. There is a problem that efficient work is forced and the cause of the image defect cannot be removed efficiently.

  In view of the above circumstances, an object of the present invention is to provide an image defect classification method and apparatus, and a program therefor, which make it possible to efficiently remove the cause of occurrence of an image defect.

  According to the first image defect classification method of the present invention, a detection medium that detects X-rays emitted from an X-ray source as an X-ray image, and X-ray image data representing the X-ray image detected by the detection medium are generated. Two images shot by changing the relative positions of the detection medium and the imaging member by an imaging system including an image data generation unit and an imaging member interposed between the X-ray source and the detection medium Based on the two X-ray image data respectively representing the X-ray images, the image defect on the two X-ray images is detected, and the position of the detected image defect is based on the change due to the relative position change. The detected image defects are classified into types according to the generation sources.

  The second image defect classification method of the present invention includes an X-ray source, a detection medium for detecting X-rays emitted from the X-ray source as an X-ray image, and an X-ray representing the X-ray image detected by the detection medium. Two X-rays representing two X-ray images respectively taken by changing the relative positions of the X-ray source and the detection medium by an imaging system including image data generating means for generating image data Image defects on the two X-ray images are detected based on the image data, and the detected image defects are generated based on changes in the position of the detected image defects due to the relative position change. It is a method characterized by classifying into different types.

  Here, the “X-ray image” is an image obtained by X-ray imaging by the imaging system configured as described above, and the presence or absence of a subject at the time of X-ray imaging does not matter. Note that the X-ray image is preferably a solid image taken without a subject in order to facilitate detection of an image defect.

  The “image defect” is a shadow as noise appearing as a dot or a line on the X-ray image acquired by the above imaging system, and dust attached to the X-ray source, dust attached to the detection medium, detection This is caused by an electrical failure or scratch on the medium, dust adhering to the imaging member, dust mixed in a path for transmitting information representing an X-ray image in the image data acquisition means, or electrical noise.

  Possible “detection media” include a stimulable phosphor sheet that stores and stores X-ray energy as an X-ray image, and a flat panel detector (including a portable type) that directly converts X-ray energy into an electrical signal. It is done.

  When the detection medium is a stimulable phosphor sheet, the image data generating means scans the stimulable phosphor sheet with excitation light to generate stimulated emission light, and photoelectrically reads the obtained stimulated emission light. Thus, the reader can generate X-ray image data.

  When the detection medium is a flat panel detector, the image data generation means can be a conversion device that A / D converts an electrical signal output from the flat panel detector to generate X-ray image data.

  As the “X-ray source”, an X-ray generator provided with an X-ray tube can be considered.

  The “change in relative position” does not include a change in the optical axis direction of the X-ray.

  In the first image defect classification method of the present invention, the imaging member is a surface on the X-ray source side of the cassette housing the detection medium, and the relative position change is performed on the detection medium stored in the cassette. The direction can be changed.

  In this case, the cassette is fixed in the imaging system, and the change of the direction of the detection medium is to rotate the detection medium by a predetermined amount. The classification is substantially performed when the two X-ray images are overlaid. Detecting a matching image defect as a type with cassette as the origin, detecting an image defect that substantially matches when at least one of the two X-ray images is rotated and superimposed so that the predetermined amount is compensated An image defect that does not correspond to any of the above-described image defects can be classified as a type whose origin is the image data generation unit as a type whose origin is the medium.

  Here, “the cassette is fixed in the imaging system” means that the spatial arrangement of the cassette during X-ray imaging and the insertion direction of the cassette into the reading device are the same. The orientation of the X-ray image follows the orientation of the cassette.

  The “predetermined amount” is preferably 90 ° or 180 ° when the X-ray detection medium is rectangular, but is not limited thereto.

  The phrase “so that the predetermined amount is compensated” means that the image shift amount in the X-ray image caused by rotating the detection medium is canceled out.

  In the first image defect classification method of the present invention, the detection medium is incorporated in an apparatus including at least a part of the imaging system so as to be rotatable or translatable, and the imaging member constitutes a part of the apparatus. The relative position change can be a position change by rotation or parallel movement of the detection medium with respect to the top plate.

  In the first image defect classification method of the present invention, before and after changing the positions of the two X-ray images so that the relative positions of the X-ray source, the detection medium, and the imaging member are different from each other. It may be taken. For example, an image taken before and after a change in which the X-ray source is rotated 90 ° to the right in the X-ray traveling direction and the detection medium is rotated 180 ° in the same direction can be considered. In this way, the detected image defect can be classified into a type having any one of the four types including the X-ray source in addition to the detection medium, the imaging member, and the image data generating means.

  As a method for detecting an image defect, a method for detecting a portion where an image signal value exceeds a predetermined threshold as disclosed in JP-A-3-277915 or an image defect, A method of detecting an image defect by removing a high frequency component of image data as disclosed in Japanese Patent No. 282596 can be used.

  A first image defect classification apparatus according to the present invention generates a detection medium that detects X-rays emitted from an X-ray source as an X-ray image, and X-ray image data representing an X-ray image detected by the detection medium. Two X images obtained by changing the relative positions of the detection medium and the imaging member by an imaging system including an image data generation unit and an imaging member interposed between the X-ray source and the detection medium. Image data input means for inputting two X-ray image data respectively representing line images, and image defects on the two X-ray images are detected based on the two X-ray image data input by the image data input means. The position information acquired by the image defect detection means for acquiring the position information of the image defect, the position change storage means for storing the change amount of the relative position due to the change, and the position information acquired by the image defect detection means, Two x-rays Based on the result of comparing the position of the image defect on the image with reference to the relative position change amount stored in the position change storage means, the detected image defect is classified into the type of generation source. And a classification means for classifying.

  The second image defect classification apparatus of the present invention includes an X-ray source, a detection medium that detects X-rays emitted from the X-ray source as an X-ray image, and an X-ray that represents the X-ray image detected by the detection medium. Two X-ray images respectively representing two X-ray images captured by changing the relative positions of the X-ray source and the detection medium by an imaging system including image data generating means for generating image data Based on the image data input means for inputting data and the two X-ray image data input by the image data input means, the image defect on the two X-ray images is detected and the position information of the image defect is acquired. Image defect detection means, position change storage means for storing the change amount of the relative position due to the change, and image defect on the two X-ray images represented by position information acquired by the image defect detection means Position change Classification means for classifying the detected image defects into types according to the generation source based on the result of comparison with reference to the amount of relative position change stored by the means. To do.

  Here, the “position change storage means” obtains relative position information before and after changing the position from a detection medium, a drive device that moves the X-ray source, and the like, and from these position information, “relative “Change in position” may be calculated and stored, or a predetermined change amount determined in advance may be stored as “relative position change amount”. The “relative position change amount” may be determined and stored based on information input by the user.

  In the first image defect classification apparatus of the present invention, the imaging member is a surface on the X-ray source side of the cassette that houses the detection medium, and the change in the relative position is performed by the detection medium housed in the cassette. The direction can be changed.

  In the first image defect classification apparatus of the present invention, the detection medium is incorporated in an apparatus including at least a part of the imaging system so as to be rotatable or translatable, and the imaging member constitutes a part of the apparatus. The relative position change can be a position change by rotation or parallel movement of the detection medium with respect to the top plate.

  The first and second image defect classification devices of the present invention may further include display means for displaying the detected image defect and the classified type of the image defect in association with each other.

  As the “display unit”, for example, a device that displays detected image defects and their types on a screen such as a CRT or a liquid crystal panel can be considered.

  In the first and second image defect classification apparatuses of the present invention, when there is a type in which the number of image defects classified by the classification unit reaches a predetermined threshold value or more, a warning that prompts maintenance of the generation source corresponding to the type is issued. You may make it further provide the warning output means to output.

  As the “warning output means”, for example, a message such as “There is dust on the cassette. Please clean or replace the cassette.” Is displayed on the screen of a CRT or liquid crystal panel, It is conceivable to sound a warning sound, turn on a warning lamp, or output a warning signal to a maintenance company that performs maintenance of the photographing system.

  The first program of the present invention generates a detection medium for detecting an X-ray emitted from an X-ray source as an X-ray image, and X-ray image data representing the X-ray image detected by the detection medium. Two X images obtained by changing the relative positions of the detection medium and the imaging member by an imaging system including an image data generation unit and an imaging member interposed between the X-ray source and the detection medium. Image data input means for inputting two X-ray image data respectively representing line images, and image defects on the two X-ray images are detected based on the two X-ray image data input by the image data input means. The position information acquired by the image defect detection means for acquiring the position information of the image defect, the position change storage means for storing the change amount of the relative position due to the change, and the position information acquired by the image defect detection means, The detected image defect is generated based on the result of comparing the position of the image defect on the two X-ray images with reference to the relative position change stored in the position change storage means. It is a program for functioning as a classification means for classifying by original type.

  The second program of the present invention includes a computer, an X-ray source, a detection medium that detects X-rays emitted from the X-ray source as an X-ray image, and an X-ray that represents the X-ray image detected by the detection medium. Two X-ray images respectively representing two X-ray images captured by changing the relative positions of the X-ray source and the detection medium by an imaging system including image data generating means for generating image data Based on the image data input means for inputting data and the two X-ray image data inputted by the image data input means, the image defect on the two X-ray images is detected and the position information of the image defect is obtained. An image defect on the two X-ray images represented by the position information acquired by the image defect detection means, the position change storage means for storing the relative change in position due to the change, and the image defect detection means Position of In order to function as a classification unit that classifies the detected image defects into types according to the generation source based on a result of comparison with reference to the relative change in position stored in the position change storage unit. It is a program.

  These programs may be supplied by being recorded on a computer-readable recording medium, or may be stored in a server connectable to the computer and supplied by downloading.

  According to the first image defect classification method and apparatus of the present invention, an image is taken by changing the relative position between the detection medium and the imaging member by the imaging system including the detection medium, the imaging member, and the image data generating means. By detecting image defects on the two X-ray images based on the two X-ray image data representing the two X-ray images, respectively, and by changing the relative positions of the detected image defects Based on the change, the detected image defects are classified into types according to the source, so the source of the detected image defect can be identified without depending on intuition and experience, and the cause of the image defect is determined. It can be removed efficiently.

  According to the second image defect classification method and apparatus of the present invention, an imaging system including an X-ray source, a detection medium, and an image data generation unit changes the relative position between the X-ray source and the detection medium. The image defect on the two X-ray images is detected based on the two X-ray image data respectively representing the two X-ray images, and the relative position of the detected image defect position is detected. Based on the change caused by the change, the detected image defects are classified into types according to the source, so the source of the detected image defect can be specified without depending on intuition and experience, and the occurrence of the image defect It becomes possible to remove the cause efficiently.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  First, a first embodiment which is an embodiment of a first image defect classification device according to the present invention will be described.

  FIG. 1 is a diagram showing an image defect classification apparatus 100 in the present embodiment, and FIG. 2 is a diagram showing an imaging system 10 and an image defect classification apparatus 100 for acquiring X-ray image data in the present embodiment. is there.

  An imaging system 10 in FIG. 2 includes an X-ray generator 1 as an X-ray source, a stimulable phosphor sheet 2 as a detection medium for detecting X-rays emitted from the X-ray source as an X-ray image, X-ray representing the X-ray image detected by the detection medium and the surface 3a on the X-ray source side of the cassette 3 containing the stimulable phosphor sheet 2 as an imaging member interposed between the radiation source and the detection medium As an image data generation means for generating image data, a stimulable phosphor sheet is scanned with excitation light to generate stimulated emission light, and the obtained stimulated emission light is photoelectrically read to obtain X-ray image data. And a reading device 4 to be generated.

  An image defect classification apparatus 100 shown in FIG. 1 is obtained by photographing two images taken by the photographing system 10 while changing the relative positions of the stimulable phosphor sheet 2 and the surface 3a of the cassette 3 on the X-ray source side. Based on the two X-ray image data P1 and P2 inputted by the image data input means 110 and the two pieces of X-ray image data P1 and P2 inputted by the image data input means 110, the two X-ray image data P1 and P2 respectively representing the X-ray images. Image defect detection means 120 for detecting image defects f1i and f2i on the X-ray images P1 and P2 and acquiring position information L1i and L2i of the image defects, and the relative position change amount E1 due to the change are stored. The position change storage means 130 and the position of the image defect on the two X-ray images represented by the position information L1i and L2i acquired by the image defect detection means 120 are stored in the position change storage means 130. Based on the comparison result with reference to the relative position change amount E1, the classification means 140 for classifying the detected image defect into the type according to the source, the detected image defect, If there is a display unit 150 that displays the classified types of image defects in association with each other, and there is a type in which the number of image defects classified by the classifying unit 140 exceeds a predetermined threshold, the generation source corresponding to the type Warning output means 160 for outputting a warning for urging maintenance is provided.

  Next, the operation of the image defect classification device 100 will be described. FIG. 3 is a diagram illustrating a processing flow in the first embodiment of the image defect classification apparatus 100.

  First, the image data input means 110 was photographed by changing the relative position between the stimulable phosphor sheet 2 as the detection medium and the cassette surface 3a as the photographing member acquired by the photographing system 10. Two pieces of X-ray image data P1 and P2 representing the two X-ray images P1 and P2 are input (step S11). Note that these two X-ray image data P1 and P2 are obtained by the method described below.

  In the imaging system 10, the cassette 3 in which the stimulable phosphor sheet 2 is stored is irradiated with X-rays by the X-ray generator 1, and the X-ray image P 1 is recorded on the stimulable phosphor sheet 2. The image is taken a second time, and the reading device 4 reads information representing the X-ray image P1 from the recorded stimulable phosphor sheet 2 stored in the cassette 3 with reference to the orientation of the cassette 3. X-ray image data P1 representing the image P1 is generated and acquired (the direction of the X-ray image follows the direction of the cassette). Thereafter, the storable phosphor sheet 2 that has been read by an erasing device (not shown) built in the reader 4 is irradiated with erasing light for erasing the stored and recorded information. The body sheet 2 is reset. The resettable stimulable phosphor sheet 2 is stored in the cassette 3 in the opposite direction (180 ° rotation) from the previous photographing. The arrangement of the cassette 3 is the same as the previous imaging, the second imaging is performed, and X-ray image data P2 representing the X-ray image P2 in a state where the stimulable phosphor sheet 2 is stored in the opposite direction is acquired. In these shootings, the presence or absence of a subject is not particularly defined, but in order to easily detect an image defect, shooting is performed with the subject removed, and the X-ray image may be a solid image without a subject image. desirable.

  FIG. 4 is a diagram illustrating an example of the X-ray images P1 and P2, and it is assumed that such an X-ray image is acquired in the present embodiment.

  When the X-ray image data P1 and P2 are input by the image data input means 110, the image defect detection means 120 is disclosed in JP-A-3-277915 based on the X-ray image data P1 and P2. Such a method for detecting a portion where the image signal value exceeds a predetermined threshold as an image defect, or removing a high frequency component of image data as disclosed in Japanese Patent Laid-Open No. 10-282596. Using a detection method or the like, the image defects f1i and f2i on the X-ray images P1 and P2 are detected, and position information L1i and L2i respectively representing the positions of the image defects are acquired (step S12). In this embodiment, it is assumed that image defects f11, f12, and f13 are detected on the X-ray image P1, and image defects f21, f22, and f23 are detected on the X-ray image P2.

  On the other hand, the position change storage means 130 determines the relative position change amount E1 between the stimulable phosphor sheet 2 and the cassette surface 3a based on information input from the user or predetermined information. Store it (step S13).

  Note that either acquisition of the position information L1i, L2i and storage of the change amount E1 may precede.

  Then, the classification unit 140 converts the detected image defect into the following based on the position information L1i, L2i acquired by the image defect detection unit 120 and the change amount E1 stored in the position change storage unit 130. As described above, the data is classified into types according to the generation source (step S14).

  The image defects f11 and f21 that substantially coincide when the two X-ray images P1 and P2 are overlapped as they are are classified as a type having the cassette 3 as the generation source (FIG. 4A), and the two X-ray images P1 and P2 Image defects f12 and f22 that substantially coincide when at least one of them is rotated 180 ° and overlapped in the opposite direction are classified as types having the stimulable phosphor sheet 2 as a generation source (FIG. 4B), and any of the above The image defects f13 and f23 that do not correspond to the image defect of FIG.

  The reason why it can be classified in this way is that image defects have the following characteristics according to their origins. However, in this embodiment, the spatial arrangement of the cassette 3 and the direction of insertion into the reading device 4 are the same. Therefore, it should be noted that the acquired two X-ray images are premised on the change of information fixed on the stimulable phosphor sheet 2 being reflected.

  First, image defects originating from the stimulable phosphor sheet 2 are considered to be substantially due to dust or scratches adhering to the stimulable phosphor sheet 2. The position of the stimulable phosphor sheet 2 is changed by the same amount as the change amount before and after the change of the direction of the stimulable phosphor sheet 2. Therefore, image defects that substantially coincide when at least one of the two X-ray images P1 and P2 is rotated by 180 ° (in the opposite direction) and overlapped are changed before and after the orientation of the stimulable phosphor sheet 2 is changed. It can be regarded as an image defect whose position has been moved by the same amount as the amount of change, and the generation source thereof can be the stimulable phosphor sheet 2.

  Next, since image defects originating from the cassette 3 are considered to be mostly due to dust adhering to the cassette, such image defects are caused by changing the orientation of the stimulable phosphor sheet 2. It has a feature that its position does not move before and after. Therefore, an image defect that substantially matches when the two X-ray images P1 and P2 are superimposed as they are can be regarded as an image defect whose position does not move before and after the change of the orientation of the stimulable phosphor sheet 2, The generation source can be the cassette 3.

  Further, an image defect originating from the reading device 4 causes the reading device 4 to scan the stimulable phosphor sheet 2 with excitation light to generate stimulated emission light. Since X-ray image data is generated by reading, most of this is considered to be caused by dust in the reading device or electrical noise in a photoelectric converter (for example, a photomultiplier tube). An image defect has a feature that its position appears at random regardless of before and after the change of the orientation of the stimulable phosphor sheet 2. Therefore, even if the two X-ray images P1 and P2 are overlapped as they are or when they are overlapped in the opposite direction, the position of the image defect is random regardless of the change in the orientation of the stimulable phosphor sheet 2. Can be regarded as an image defect appearing in FIG.

  In this manner, when the image defect is classified into the classification according to the generation source by the classification unit 140, the display unit 150 associates each image defect with the classified type and displays it on a screen such as a CRT. (Step S15). For example, the X-ray images P1 and P2 are displayed on the screen, and the detected image defect in the image is indicated by being circled based on the position information L1i and L2i, and corresponds to the type classified in the vicinity thereof. The type of generation source (cassette, stimulable phosphor sheet, reader) is displayed, or the circle line color surrounding the image defect is displayed for each type.

  When the number of classified image defects reaches a predetermined threshold or more in at least one of the classified types, the warning output unit 160 outputs a warning that prompts maintenance of the generation source corresponding to the type. (Step S16). For example, when the number of image defects classified into the category of the cassette generation source exceeds a predetermined threshold value, “Dust is attached to the cassette. Please clean or replace the cassette.” Displaying a message on the screen, sounding a warning sound, turning on a warning lamp, or outputting a warning signal via a network to a maintenance company that performs maintenance of the shooting system Good.

  Next, a second embodiment, which is another embodiment of the first image defect classification device according to the present invention, will be described. The image defect classification device in the present embodiment is the image defect classification device 100 shown in FIG. 1 as in the previous embodiment. FIG. 5 is a diagram showing an imaging system 20 for acquiring X-ray image data in the present embodiment.

  An imaging system 20 shown in FIG. 5 uses an X-ray generator 1 as an X-ray source and an X-ray image information as a detection medium for detecting X-rays emitted from the X-ray source as an X-ray image. Of the built-in type imaging apparatus 6 in which the FPD 5 is incorporated so as to be movable in the vertical direction as an imaging member interposed between the X-ray source and the detection medium. X / ray image data is generated by A / D converting the electrical signal output from the FPD 5 as image data generating means for generating X-ray image data representing the X-ray image detected by the plate 6a and the detection medium. It is comprised with the conversion apparatus 7. FIG.

  Next, the operation of the image defect classification device 100 will be described. FIG. 6 is a diagram illustrating a processing flow in the second embodiment of the image defect classification apparatus 100.

  First, the image data input means 110 acquires two X-ray images P3 acquired by the imaging system 20 by changing the relative positions of the FPD 5 as the detection medium and the top plate 6a as the imaging member. Two pieces of X-ray image data P3 and P4 representing P4 are input (step S21). Note that these two X-ray image data P3 and P4 are obtained by the method described below.

  In the imaging system 20, the top plate 6 a of the imaging device 6 in which the FPD 5 is built is irradiated with X-rays by the X-ray generation device 1, and information representing the X-ray image P 3 is output by the FPD 5 as an electrical signal. The first imaging is performed, and the conversion device 7 performs A / D conversion on the electric signal output from the FPD 5 with reference to the direction of the FPD 5 to generate and obtain X-ray image data P3 representing the X-ray image P3. (X-ray image orientation follows FPD orientation). Thereafter, the FPD 5 is translated upward by a predetermined amount with respect to the top plate 6a. The position of the top 6a is the same as the previous imaging, the second imaging is performed, and X-ray image data P4 representing the X-ray image P4 in a state where the FPD 5 is translated is acquired. In these shootings, the presence or absence of a subject is not particularly defined, but in order to easily detect an image defect, shooting is performed with the subject removed, and the X-ray image may be a solid image without a subject image. desirable.

  FIG. 7 is a diagram illustrating an example of the X-ray images P3 and P4. In this embodiment, it is assumed that such an X-ray image is acquired.

  When the X-ray image data P3 and P4 are input by the image data input means 110, the image defect detection means 120 is disclosed in JP-A-3-277915 based on the X-ray image data P3 and P4. Such a method for detecting a portion where the image signal value exceeds a predetermined threshold as an image defect, or removing a high frequency component of image data as disclosed in Japanese Patent Laid-Open No. 10-282596. Using a detection method or the like, the image defects f3i and f4i on the X-ray images P3 and P4 are detected, and position information L3i and L4i representing the positions of the image defects are acquired (step S22). In the present embodiment, it is assumed that image defects f31, f32, and f33 are detected on the X-ray image P3, and image defects f41, f42, and f43 are detected on the X-ray image P4.

  On the other hand, the position change storage means 130 has a drive amount for moving the FPD 5 relative to the change amount E2 of the relative position between the FPD 5 and the top plate 6a, information input from the user or predetermined information. In this case, the position is determined and stored based on the position information of the FPD 5 from the driving device (step S23).

  Note that either acquisition of the position information L3i, L4i or storage of the change amount E2 may precede.

  Then, the classification unit 140 converts the detected image defect into the following based on the position information L3i, L4i acquired by the image defect detection unit 120 and the change amount E2 stored in the position change storage unit 130. As described above, the data is classified into types according to the generation source (step S24).

  The image defects f31 and f41 that substantially coincide when the two X-ray images P3 and P4 are overlapped as they are are classified as a type having the FPD 5 as a generation source (FIG. 7A). When at least one is moved and overlapped so that the difference in image arrangement on the two X-ray images based on the relative position change is eliminated, for example, the X-ray image P4 is moved upward by the predetermined amount. The image defects f32 and f42 that substantially coincide with each other when they are moved to and overlaid are classified as the types having the top plate 6a as the generation source (FIG. 7B), and the other image defects f33 and f43 are converted to the conversion device 7. It is classified as a type that is the source.

  The reason why the image defect can be classified in this manner is that image defects have the following characteristics depending on the generation source. However, in this embodiment, the spatial arrangement of the top plate 6a of the photographing apparatus 6 is the same. It should be noted that it is assumed that the acquired two X-ray images reflect a change in information fixed on the top plate 6a.

  First, since image defects originating from the top plate 6a are practically considered to be due to dust adhering to the top plate 6a, such image defects are before and after the change of the position of the FPD 5, The position is moved by the same amount as the amount of change. Therefore, an image defect that substantially coincides when at least one of the two X-ray images P1 and P2 is moved in a direction in which the amount of change is canceled and overlapped is regarded as the amount of change before and after the parallel movement of the FPD 5. It can be regarded as an image defect whose position has been moved by the same amount, and its source can be the top plate 6a.

  Next, image defects originating from the FPD 5 are actually considered to be mostly due to dust adhering to the FPD or electrical failure of the pixels. Therefore, such image defects are located at the position of the FPD 5. The position does not move before and after the change. Therefore, an image defect that substantially matches when the two X-ray images P1 and P2 are superimposed as they are can be regarded as an image defect whose position has not moved before and after the parallel movement of the FPD 5. can do.

  Furthermore, most of the image defects originating from the conversion device 7 are those in which the conversion device 7 performs A / D conversion on the electrical signal output from the FPD 5 to generate X-ray image data. Since this is considered to be due to electrical noise in the signal transmission path, such image defects have the feature that their positions appear randomly regardless of before and after the parallel movement of the FPD 5. Therefore, even if the two X-ray images P3 and P4 are overlapped as they are or when they are overlapped in the opposite direction, the image defects that do not coincide with each other regardless of before and after the parallel movement of the FPD 5 And the generation source can be the conversion device 7.

  Since the operations (steps S25 and S26) of the display unit 150 and the warning output unit 160 are the same as those in the first embodiment, they are omitted here.

  According to such an image classification device 100 in the first and second embodiments, the relative position between the detection medium and the photographing member is changed by the photographing system including the detection medium, the photographing member, and the image data generating unit. The image defects on the two X-ray images are detected based on the two X-ray image data respectively representing the two X-ray images photographed in the above, and the relative positions of the detected image defects are detected. Based on the change due to the change of position, the detected image defects are classified into types according to the source, so the source of the detected image defects can be identified without depending on intuition and experience, and the image defects It is possible to efficiently remove the cause of the occurrence.

  Next, a third embodiment that is an embodiment of the second image defect classification device of the present invention will be described.

  FIG. 8 is a diagram illustrating an image defect classification device 200 that is an embodiment of the second image defect classification device. In the present embodiment, the imaging system for acquiring the X-ray image data is the imaging system 20 shown in FIG. 5, but the X-ray generator 1 is configured such that the X-ray source head 1 a emits X-rays. The optical axis can be rotated by 90 °.

  The image defect classification apparatus 200 shown in FIG. 8 has two X-ray image data representing two X-ray images respectively captured by the imaging system 20 by rotating the source head 1a of the X-ray generator 1 by 90 °. Based on the image data input means 210 for inputting P5 and P6, and the two X-ray image data P5 and P6 input by the image data input means 210, image defects f5i on the two X-ray images P5 and P6, Image defect detection means 220 for detecting f6i to obtain position information L5i, L6i of the image defect, position change storage means 230 for storing the change amount E3 of the position of the source head 1a due to the rotation, and image defect detection The position of the image defect on the two X-ray images represented by the position information L5i and L6i acquired by the means 220 is referred to the change amount E3 stored in the position change storage means 230. Based on the comparison result, the classifying means 240 for classifying the detected image defect into the type classified by the generation source, the detected image defect, and the classified type of the image defect are displayed in association with each other. When there is a type in which the number of image defects classified by the classification unit 240 reaches a predetermined threshold or more, there is a warning output unit 260 that outputs a warning that prompts maintenance of the generation source corresponding to the type. ing.

  Next, the operation of the image defect classification device 200 will be described. FIG. 9 is a diagram showing a processing flow in the third embodiment of the image defect classification device 200.

  First, the image data input means 210 acquires two X-ray images P5 and P6 acquired before and after rotating the source head 1a of the X-ray generation apparatus 1 as an X-ray source by 90 °, acquired by the imaging system 20. Are input two pieces of X-ray image data P5 and P6 (step S31). These two X-ray image data P5 and P6 are obtained by the method described below.

  In the imaging system 20, the top plate 6 a of the imaging device 6 in which the FPD 5 is built is irradiated with X-rays by the X-ray generator 1, and information representing the X-ray image P5 is output as an electrical signal by the FPD 5. The first imaging is performed, and the conversion device 7 performs A / D conversion on the electrical signal output from the FPD 5 with reference to the direction of the FPD 5 to generate and acquire X-ray image data P5 representing the X-ray image P5. (X-ray image orientation follows FPD orientation). Thereafter, the radiation source head 1a of the X-ray generator 1 is rotated 90 ° clockwise toward the FPD 5 around the optical axis. X-ray image data representing the X-ray image P6 in a state in which the positions of the top plate 6a and the FPD 5 are the same as the previous imaging, the second imaging is performed, and the radiation source head 1a of the X-ray generator 1 is rotated by 90 °. Get P6. In these shootings, the presence or absence of a subject is not particularly defined, but in order to easily detect an image defect, shooting is performed with the subject removed, and the X-ray image may be a solid image without a subject image. desirable.

  FIG. 10 is a diagram illustrating an example of the X-ray images P5 and P6. In this embodiment, it is assumed that such an X-ray image has been acquired.

  When the X-ray image data P5 and P6 are input by the image data input means 210, the image defect detection means 220 is disclosed in Japanese Patent Laid-Open No. 3-277915 based on the X-ray image data P5 and P6. Such a method of detecting a portion where the image signal value representing the pixel density exceeds a predetermined threshold as an image defect, or removing high-frequency components of image data as disclosed in JP-A-10-282596 Then, the image defects f5i and f6i on the X-ray images P5 and P6 are detected by using a method for detecting image defects, and position information L5i and L6i representing the positions of the image defects are obtained (step S32). . In the present embodiment, it is assumed that image defects f51 and f52 are detected on the X-ray image P5 and image defects f61 and f62 are detected on the X-ray image P6.

  On the other hand, the position change storage means 230 uses the amount of change E3 of the source head 1a of the X-ray generator 1 as the information input from the user or predetermined information, and the source head 1a of the X-ray generator 1. If there is a drive device for movement, the drive device is determined and stored based on the positional information of the radiation source head 1a from the drive device (step S33).

  Note that either acquisition of the position information L5i, L6i or storage of the change amount E3 may precede.

  Then, the classification unit 240 converts the detected image defect into the following based on the position information L5i, L6i acquired by the image defect detection unit 220 and the change amount E3 stored in the position change storage unit 230. As described above, the data is classified into types according to the generation source (step S34).

  When at least one of the two X-ray images P5 and P6 is moved and overlapped so as to eliminate the difference in image arrangement on the two X-ray images based on the rotation of the source head 1a, for example, X-ray generator 1 (radiation head 1a) generates image defects f51 and f61 that substantially coincide when X-ray image P6 is rotated counterclockwise by 90 °, which is the amount of rotation of radiation source head 1a, and overlapped. As the type to be a source (FIG. 10B), the other image defects f52 and f62 are classified as types having a source other than the X-ray generator.

  Since the operations (steps S35 and S36) of the display means 250 and the warning output means 260 are the same as those of the display means 150 and the warning output means 160 of the first embodiment, they are omitted here.

  According to such an image defect classification device 200 in the third embodiment, the relative position between the X-ray source and the detection medium is changed by the imaging system including the X-ray source, the detection medium, and the image data generation means. The X-ray source of the position of the detected image defect is detected based on the two X-ray image data respectively representing the two X-ray images taken in Since the detected image defects are classified into types according to the source based on the change due to the change in the position of the image, the source of the detected image defect can be identified without depending on intuition and experience, and the image It is possible to efficiently remove the cause of the defect.

  In the classification of image defects, at least one of the two X-ray images is displayed on two X-ray images based on a change in the relative position between the detection medium and the imaging member or a change in the position of the X-ray source. When moving and superimposing images so as to eliminate the image difference at the same time, when classifying substantially coincident image defects into a predetermined type, one or both X-ray images are moved by a predetermined amount and then superimposed. However, it is conceivable to use a method for detecting image defects that substantially match, but in addition, image defects that appear to be the same in two X-ray images are preliminarily searched for from features such as shape, size, and density. Then, one or both images are moved until the image defect substantially matches, and the relative movement amount is dependent on the change in the relative position between the detection medium and the imaging member and the change in the position of the X-ray source. When the image defect is almost equal It is considered to use a method of detecting.

  It should be noted that a program for causing a computer to function as each means in the image defect classification apparatus of each of the above embodiments may be created, and this program may be recorded on a computer-readable recording medium and supplied. Alternatively, it may be stored in a server to which a computer can be connected and supplied by downloading. In this way, by installing the program on the computer and executing it, the computer can be operated as each image defect classification device, and the same effect can be obtained.

  For example, in the case of the first embodiment, as shown in FIG. 2, a maintenance company that operates a computer that executes the above-described program as the image defect classification apparatus 100 and connects the computer to the reading apparatus 4 and also performs imaging system maintenance. And may be connected via a network. As the network, a wired or wireless telephone line network is conceivable, but of course, it is not limited to this.

The figure which shows the image defect classification device 100 which is one Embodiment of the image defect classification device of this invention. The figure which shows the imaging | photography system 10 for acquiring X-ray image data The figure which shows the processing flow in 1st Embodiment of the image defect classification device 100. FIG. The figure which shows an example of X-ray image P1, P2 The figure which shows the imaging | photography system 20 for acquiring X-ray image data The figure which shows the processing flow in 2nd Embodiment of the image defect classification device 100. FIG. The figure which shows an example of X-ray image P3, P4 The figure which shows the image defect classification device 200 which is one Embodiment of the 2nd image defect classification device. The figure which shows the processing flow of the image defect classification device 200 The figure which shows an example of X-ray image P5, P6

Explanation of symbols

1 X-ray generator
2 Storage phosphor sheet
3 cassette
4 Reader
5 Flat panel detector (FPD)
6 Shooting device
7 Conversion Device 10 Imaging System 20 Imaging System 100 Image Defect Classification Device 110 Image Data Input Unit 120 Image Defect Detection Unit 130 Position Change Storage Unit 140 Classification Unit 150 Display Unit 160 Warning Output Unit 200 Image Defect Classification Unit 210 Image Data Input Unit 220 Image defect detection means 230 Position change storage means 240 Classification means 250 Display means 260 Warning output means

Claims (13)

A detection medium for detecting X-rays emitted from the X-ray source as an X-ray image, image data generation means for generating X-ray image data representing the X-ray image detected by the detection medium, and the X-ray source; Two X-ray images respectively representing two X-ray images captured by an imaging system including an imaging member interposed between the detection medium and a relative position between the detection medium and the imaging member. Detecting image defects on the two X-ray images based on the line image data;
An image defect classification method, wherein the detected image defect is classified into a type according to a generation source based on a change in the position of the detected image defect due to the change of the relative position. The imaging member is a surface on the X-ray source side of a cassette that contains the detection medium and transmits X-rays;
The image defect classification method according to claim 1, wherein the change of the relative position is a change of an orientation of the detection medium stored in the cassette. The detection medium is incorporated in an apparatus including at least a part of the imaging system so as to be rotatable or translatable,
The imaging member is a top plate that forms part of the apparatus and transmits X-rays;
The image defect classification method according to claim 1, wherein the change in the relative position is a change in position of the detection medium by rotation or translation with respect to the top plate. The cassette is fixed in the imaging system;
The change of the direction of the detection medium is to rotate the detection medium by a predetermined amount,
The classification is
An image defect that substantially matches when the two X-ray images are overlaid is classified as a type with the cassette as a generation source.
An image defect that substantially coincides when at least one of the two X-ray images is rotated and overlapped so as to compensate for the predetermined amount is defined as a type having the detection medium as a generation source.
3. The image defect classification method according to claim 2, wherein an image defect that does not correspond to any of the image defects is classified as a type having the image data generation means as a generation source. An X-ray source, a detection medium for detecting X-rays emitted from the X-ray source as an X-ray image, and image data generation means for generating X-ray image data representing the X-ray image detected by the detection medium; Based on two X-ray image data representing two X-ray images respectively taken by changing the relative positions of the X-ray source and the detection medium. Detect image defects on line images,
An image defect classification method, wherein the detected image defect is classified into a type according to a generation source based on a change in the position of the detected image defect due to the change of the relative position. A detection medium for detecting X-rays emitted from the X-ray source as an X-ray image, image data generation means for generating X-ray image data representing the X-ray image detected by the detection medium, and the X-ray source; Two X-rays respectively representing two X-ray images captured by changing the relative positions of the detection medium and the imaging member by an imaging system including an imaging member interposed between the detection medium and the imaging medium Image data input means for inputting image data;
Image defect detection means for detecting image defects on the two X-ray images and obtaining positional information of the image defects based on the two X-ray image data input by the image data input means;
Position change storage means for storing a change amount of the relative position due to the change;
The position of the image defect on the two X-ray images represented by the position information acquired by the image defect detection means is referred to the relative position change amount stored by the position change storage means. An image defect classification apparatus, comprising: a classifying unit that classifies the detected image defect into a classification according to a generation source based on a result of comparison. The imaging member is a surface of the cassette containing the detection medium on the X-ray source side;
The image defect classification apparatus according to claim 6, wherein the change in the relative position is a change in a direction of the detection medium stored in the cassette. The detection medium is incorporated in an apparatus including at least a part of the imaging system so as to be rotatable or translatable,
The imaging member is a top plate that forms part of the apparatus and transmits X-rays;
The image defect classification apparatus according to claim 6, wherein the change in the relative position is a change in position of the detection medium by rotation or translation with respect to the top plate. An X-ray source, a detection medium for detecting X-rays emitted from the X-ray source as an X-ray image, and image data generation means for generating X-ray image data representing the X-ray image detected by the detection medium; Image data input means for inputting two X-ray image data respectively representing two X-ray images captured by changing the relative positions of the X-ray source and the detection medium by an imaging system including:
Image defect detection means for detecting image defects on the two X-ray images and obtaining positional information of the image defects based on the two X-ray image data input by the image data input means;
Position change storage means for storing a change amount of the relative position due to the change;
The position of the image defect on the two X-ray images represented by the position information acquired by the image defect detection means is referred to the relative position change amount stored by the position change storage means. An image defect classification apparatus, comprising: a classifying unit that classifies the detected image defect into a classification according to a generation source based on a result of comparison. 10. The image defect classification apparatus according to claim 6, further comprising display means for displaying the detected image defect and the classified type of the image defect in association with each other. When there is a type in which the number of image defects classified by the classification unit reaches a predetermined threshold value or more, a warning output unit is further provided that outputs a warning that prompts maintenance of the generation source corresponding to the type. The image defect classification device according to claim 6. Computer
A detection medium for detecting X-rays emitted from the X-ray source as an X-ray image, image data generation means for generating X-ray image data representing the X-ray image detected by the detection medium, and the X-ray source; Two X-rays respectively representing two X-ray images captured by changing the relative positions of the detection medium and the imaging member by an imaging system including an imaging member interposed between the detection medium and the imaging medium Image data input means for inputting image data;
Image defect detection means for detecting image defects on the two X-ray images and obtaining positional information of the image defects based on the two X-ray image data input by the image data input means;
Position change storage means for storing a change amount of the relative position due to the change;
The position of the image defect on the two X-ray images represented by the position information acquired by the image defect detection means is referred to the relative position change amount stored by the position change storage means. A program for causing the detected image defect to function as a classifying unit that classifies the detected image defect into a type according to a generation source based on the comparison result. Computer
An X-ray source, a detection medium for detecting X-rays emitted from the X-ray source as an X-ray image, and image data generation means for generating X-ray image data representing the X-ray image detected by the detection medium; Image data input means for inputting two X-ray image data respectively representing two X-ray images captured by changing the relative positions of the X-ray source and the detection medium by an imaging system including:
Image defect detection means for detecting image defects on the two X-ray images and obtaining positional information of the image defects based on the two X-ray image data input by the image data input means;
Position change storage means for storing a change amount of the relative position due to the change;
The position of the image defect on the two X-ray images represented by the position information acquired by the image defect detection means is referred to the relative position change amount stored by the position change storage means. A program for causing the detected image defect to function as a classifying unit that classifies the detected image defect into a type according to a generation source based on the comparison result.

Images