JP4574181B2 - Image processing method and apparatus - Google Patents

Description

  The present invention relates to an image processing method and apparatus suitable for processing an X-ray image. In particular, it is suitable for processing of X-ray image data in a digital X-ray imaging system using a solid-state imaging device.

  For X-ray imaging for medical diagnosis, a film screen system combining an intensifying screen and an X-ray photographic film is generally used. According to this system, X-rays that have passed through a subject are converted into visible light proportional to the intensity of the X-rays using an intensifying screen, and the X-ray photographic film is exposed to the visible light to form an X-ray image on the film. To do. Since the X-rays that have passed through the subject include internal information of the subject, the X-ray image formed on the film is an image representing the inside of the subject.

  Recently, X-rays are converted into visible light proportional to the intensity of X-rays by phosphors, converted into electrical signals by photoelectric conversion elements, and digitally converted by A / D converters. X-ray digital imaging devices that can be obtained with digital images are beginning to be used. As the photoelectric conversion element used here, one using amorphous silicon is known. According to such an X-ray digital imaging apparatus, not only the imaging method used in the conventional film screen system can be performed but also an X-ray using an image intensifier conventionally used. Compared to a digital imaging device, the X-ray sensor unit can be made larger, thinner and lighter.

  On the other hand, in a digital X-ray imaging apparatus using an image sensor, the gain of each pixel is not constant. Therefore, in order to generate a uniform output image with respect to an input image, gain correction for each pixel is required. A process for obtaining data for gain correction (gain correction data) is called calibration, and is generally performed by an operator's instruction at a certain interval. In the calibration, radiation is applied to the entire effective imaging area with the subject excluded. An image obtained by normalizing the output value of the captured image (hereinafter referred to as gain image) is stored as gain correction data. In actual subject photographing, the photographed image is corrected with the gain correction data.

  Also, it is redundant to use all of the captured images in display and image analysis. In general, when image processing parameters are determined based on a captured image, or when an image is displayed on a screen, a reduced image derived from the captured image is used.

  However, if such a reduced image is to be generated by calculation from all the images formed by all the pixels of the X-ray image sensor, the transfer of all the images, that is, the transfer of a large amount of data must be performed at high speed. Therefore, a high-speed I / F is required, and the product cost increases.

Therefore, by thinning out some pixels of the image obtained from the X-ray image sensor and transmitting it in advance, that is, by transmitting a reduced image derived from the entire image, it is possible to quickly present the image to the operator. Based on this, a method of determining success or failure of shooting and determining image processing parameters has been proposed (see, for example, Patent Document 1).
JP 2003-325494 A

  As described above, when high-speed image transfer is realized by using a reduced image, the main purpose is to determine success or failure of photographing, and therefore it is necessary to present the transferred reduced image to the user. However, no consideration is given to performing gain correction or the like on such a reduced image. Mounting a correction circuit for realizing gain correction on the X-ray imaging device side affects the product cost of the X-ray imaging apparatus. In addition, a defective pixel is an accessory of the solid-state imaging device, and it is necessary to register the defective pixel in advance and correct and display the pixel so that the defective pixel does not interfere with the diagnosis. However, when a defective pixel corresponds to an image generated by thinning out pixels from the entire image like the above-described reduced image, such a defective pixel remains as an abnormal pixel in the reduced image. For this reason, the image quality of preview display using reduced images and the determination of image processing parameters may be adversely affected.

  In X-ray imaging, a grid may be used to reduce the influence of scattered X-rays on an image. When such a grid is used, the pixel pitch of the solid-state image sensor and the grid density interfere with each other, and moire may occur on the display image. Moire does not cause much trouble for radiologists to determine the success or failure of X-ray imaging, whether the subject is not blurred or whether the concentration is normal, but it has a negative effect on the confirmation of radiographed images by radiologists The observation time tends to be longer. Further, although it is conceivable to perform filtering processing on the reduced image to suppress moiré, high-frequency components are cut by the filtering processing, which may affect the determination of image processing parameters by image analysis.

  The present invention has been made in view of the above problems. For example, in a system or apparatus that improves operability by acquiring a thinned image in advance as described above, reduction with appropriate correction is performed. It is possible to generate an image, and to improve the display quality and to determine a more accurate image processing parameter.

In order to achieve the above object, an image processing method according to the present invention comprises:
An image processing method in an image processing apparatus for processing an image captured using a grid in order to suppress scattered X-rays ,
An acquisition step in which an acquisition unit acquires a thinned image by extracting pixels arranged in a diagonal direction of the rectangular area for each rectangular area of a predetermined size from a captured image obtained by an imaging device including an imaging element;
A generating step for generating defective pixel correction data for a thinned image corresponding to a pixel configuration of the thinned image based on defective pixel data indicating a defective pixel position of the image sensor;
An integration step of integrating the pixels extracted from the same rectangular region in the thinned image to generate a reduced image by integrating the pixels into one pixel ;
Wherein the integrated process, with reference to the defective pixel correction data for the decimated image, to exclude the data of the defective pixels in the decimated image from the subject of the integration.

In order to achieve the above object, a photographing system according to the present invention comprises the following arrangement. That is,
An imaging system in which an imaging device and an image processing device for processing an image captured by the imaging device using a grid to suppress scattered X-rays are connected,
A thinned image is obtained by extracting pixels arranged in a diagonal direction of the rectangular area for each rectangular area of a predetermined size from the photographed image obtained by photographing, and the thinned image is acquired prior to transmission of the photographed image. Transfer means for transferring to the image processing device;
In the image processing device, generating means for generating defective pixel correction data for a thinned image corresponding to a pixel configuration of the thinned image based on defective pixel data indicating a defective pixel position of an imaging element in the photographing device ;
Integration means for generating a reduced image by integrating the pixels extracted from the same rectangular area in the thinned image image transferred by the transfer means into one pixel ;
Said integrating means refers to the defective pixel correction data for the decimated image, to exclude the data of the defective pixels in the decimated image from the subject of the integration.

  According to the present invention, it is possible to generate a reduced image that has been appropriately corrected from the thinned image, so that the display quality is improved and more accurate image processing parameters are determined.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings. In the following embodiments, the present invention is applied to an X-ray imaging system using a solid-state imaging device.

  FIG. 1 is a block diagram showing a configuration example of an X-ray imaging system according to the present embodiment. As shown in FIG. 1, the X-ray imaging system of this embodiment includes a flat panel detector (FPD) 1 having a phosphor and a large-screen photoelectric conversion device, an X-ray generation source, and an X-ray source having a control mechanism thereof. A device interface 3 for performing power supply control to the X-ray generator 2, FPD 1 and synchronous control of exposure with the X-ray generator 2, and a medical digital X-ray imaging device 4 (control / control the entire X-ray imaging system) In addition to managing the image, image processing such as correction of the acquired image is executed. The FPD 1 has a configuration in which thinned image data is derived from the obtained image data and transmitted, and then image data corresponding to all images is transmitted. For example, in this embodiment, a thinned-out image having a pattern as described later with reference to FIG. 6 is derived. Note that the thinning pattern of the FPD 1 may be variable (programmable) or fixed. In the present embodiment, the scintillator temporarily converts the light into light. However, a solid-state imaging device having sensitivity to the X-ray itself may be used.

  The X-ray imaging apparatus 4 includes a CPU 10, a RAM 6, a hard disk 12, a disk interface 13, an internal bus 14, LAN interfaces 11 and 15, and an input / output interface 16 for an input / output device for providing a user interface. .

  The CPU 10 executes various control programs in the X-ray imaging apparatus 4. In this embodiment, the CPU 10 uses Pentium (registered trademark) and Windows (registered trademark) as the OS. A control program 7 for realizing processing according to the present embodiment to be described later is loaded from the hard disk 12 to the RAM 6 by the CPU 10 via the OS and executed by the CPU 10. In addition, an image storage area 8 is secured on the RAM 6 and a captured image is temporarily stored therein. A correction data area 9 secured on the RAM 6 is an area for storing correction data for use in correcting image data.

  The hard disk 12 stores a control program for causing the CPU 10 to execute the processing described in the present embodiment, correction data necessary for photographing, and temporarily stores photographed image data. The internal bus 14 connects each component in the X-ray imaging apparatus 4. The LAN interface 15 is an interface for connecting to a local area network (LAN) to which external devices such as the PACS 21, the ordering device 22, and the printer 23 are connected. In the present embodiment, the X-ray imaging apparatus 4 receives an imaging order from the external ordering apparatus 22 via the LAN, and uses the image data as an external device, such as a PACS 21 or a printer 23, in order to use the captured image for diagnosis. Over the LAN.

  The LAN interface 11 connects the device interface 3 via the LAN cable 19. The input / output interface 16 connects a display device 17 and an input device 18 such as a keyboard and a mouse that serve as an interface with a radiographer who is an operator of the X-ray imaging apparatus 4. Of course, it is obvious that the display and input units 17 and 18 may be constituted by a touch panel device or the like.

  The FPD 1 and the device interface 3 are connected via a cable group 27 including a power cable, an X-ray control signal cable, a data cable for image transfer and control signals. In this embodiment, a general-purpose data communication technique such as Ethernet (registered trademark) can be used as the data cable.

  In this embodiment, the LAN interface 11 is separated from the LAN interface 15. This is preferable from the viewpoint of preventing the LAN 19 for image transmission from being affected by the traffic of the hospital LAN (LAN 20). However, if a LAN having a sufficient transmission capacity is used, the LAN for image transmission and the in-hospital LAN can be configured by one common LAN.

  FIG. 2 is a screen configuration example of a user interface displayed on the display device 17 in the X-ray imaging apparatus 4. In the area 34, the subject information input from the ordering device 22 or the keyboard 18 is displayed. The button registration tray 31 is provided with a button for instructing an imaging parameter set in which part information to be imaged is preset. Reference numeral 33 denotes an example in which the photographing order is displayed as a button. In the area 32, shooting conditions preset for the input button are displayed. In the area 35, the state of the X-ray imaging apparatus 4 is displayed. Reference numeral 36 denotes a button for inputting an instruction for completion of photographing.

  The usage method of the X-ray imaging system of this embodiment provided with the above structures is demonstrated below using FIGS.

  First, when imaging a subject, the operator selects a subject to be X-rayed from the subject information already input by the ordering device 22 on the display terminal 17 of the X-ray imaging device 4. The subject information includes, for example, the subject name, date of birth, sex, and the like, and these pieces of information are acquired by the X-ray imaging apparatus 4 by selecting the subject name.

  Subsequently, in order to determine the image processing conditions of the image obtained by imaging, a part to be imaged to be performed is selected from the button registration tray 31. For example, when photographing the front of the chest, the operator selects a button representing a photographing object “front of the chest” from the button registration tray 31. The buttons corresponding to each imaging region are associated with imaging conditions, image processing conditions, correction data, and the like suitable for the imaging region, and these are stored in the hard disk 12 as a database. Accordingly, when the button 33 having imaging conditions for the front of the chest is selected, imaging conditions for the chest, image processing conditions, correction data, and the like are taken from the database on the hard disk 12 into the correction data area 9 on the RAM 6.

  Furthermore, the hard disk 12 stores ID information of each FPD owned and correction data adapted to each FPD in association with each other. In this embodiment, gain correction data and defective pixel correction data are held as correction data corresponding to each FPD. The gain correction data records the gain correction value for each pixel of the corresponding FPD, and the defective pixel correction data records the defective pixel position of the corresponding FPD.

  The CPU 10 acquires ID information from the FPD to be used, searches the hard disk 12 for corresponding correction data using the ID information, reads correction data (gain correction data and defective pixel correction data) that hit the search, Is loaded into the correction data storage area 9 on the RAM 6. As FPD ID information holding means, a flash memory capable of holding contents regardless of power ON / OFF, a DIP switch set at the time of factory shipment, and the like can be considered.

  After the above preparation, the engineer who is the operator positions the subject and presses the exposure switch 25 of the X-ray generator 2. The X-ray generator control unit 26 and the X-ray imaging apparatus 4 synchronize the X-ray exposure timing via the apparatus interface 3. X-rays output from the X-ray tube 24 in response to a binding person signal from the X-ray generator control unit 26 pass through the subject and enter the X-ray FPD1. In this case, the incident X-ray includes internal information of the subject. In the FPD 1, after scattered X-rays are blocked through a grid (not shown), it is converted into visible light proportional to the intensity of the X-rays by a phosphor, and a charge proportional to the visible light is accumulated by a photoelectric conversion device. The accumulated charge amount is digitized by A / D conversion and becomes X-ray image data.

  As will be described in detail later, in the present embodiment, the thinned image data obtained by thinning out the X-ray image data obtained as described above is first transmitted to the X-ray imaging apparatus 4. Thereafter, the X-ray image data (the entire imaging data) is transmitted to the X-ray imaging apparatus 4 by background processing, for example.

  The thinned image data is transferred and stored from the apparatus interface 3 via the LAN cable 19 and the LAN interface 11 to the image storage area 8 on the RAM 6 in the X-ray imaging apparatus 4. The thinned image is subjected to image correction processing using correction data under the control of the CPU 10 to become a reduced image, and is displayed on the display device 17 as a confirmation image.

  FIG. 3 is a diagram illustrating an example of a state in which the CPU 10 of the X-ray imaging apparatus executes the control program 7 and displays a reduced image on the user interface on the display device 17. An area 40 is an area for displaying a reduced image obtained based on the thinned image data transferred from the device interface 3. Displayed in this area 40 is a display image for confirmation, which is a reduced image based on thinned-out image data derived from all images obtained by photographing with the FPD 1. In the present embodiment, it is assumed that the FPD 1 has 2688 × 2688 pixels, but generally, a reduced image for confirmation display (and for generating image processing parameters) is about 336 × 336 pixels.

  When a dedicated high-speed interface is used, all images can be transferred in about 1 second. Conventionally, all images are transferred using such a high-speed interface, and then all images are corrected, and the corrected image is reduced and used as a display image. However, in the X-ray FPD 1 of the present embodiment, for the purpose of reducing the circuit scale and cost, the image is transferred from the FPD 1 (device interface 3) through a general-purpose interface, for example, Ethernet (registered trademark) I / F. The data is transferred to the X-ray imaging apparatus 4. For this reason, it takes about 8 seconds to transfer all the images, and when the reduced image is displayed after waiting for the completion of all the image transfer as in the conventional machine using the high-speed interface, it seems that the performance is greatly deteriorated for the operator. The operability is also poor. Therefore, in this embodiment, a thinned image for the captured image is generated, and this thinned image is transmitted from the FPD 1 to the photographing apparatus 4 first. In the photographing apparatus 4, the received thinned image is subjected to correction processing or the like to obtain a reduced image, which is displayed. While the user performs image confirmation using the reduced image, the entire image from the FPD 1 is transferred to the X-ray imaging apparatus 4 as background processing and stored as inspection data.

  The parameter operation unit 42 provides a user interface for changing and setting density adjustment parameters and contrast adjustment parameters. For example, when the automatic density adjustment of the image displayed in the area 40 is not suitable, the operator can change the density of the captured image by adjusting the parameter operation unit 42 (density adjustment interface). When the operator finishes confirming the photographed image on the screen shown in FIG. 3, the operator presses the “next photographing” button 43 on the screen to perform the next photographing. When the “next shooting” button 43 is pressed, the user interface returns to the state shown in FIG.

  The control program 7 executed by the CPU 10 causes the operator to select the next imaging part ordered by the user interface shown in FIG. 2, and causes the operator to repeatedly execute the imaging flow described above until all imaging orders are completed. . When all the photographing orders are completed, there is no next photographing, so the CPU 10 changes the next photographing button 43 shown in FIG. When the operator presses the button 43 in this state, the test for the subject ends.

  When the inspection is completed, the control program 7 transmits the photographing conditions and photographing execution information stored in the hard disk 12 to the ordering device 22. The control program 7 notifies the ordering apparatus 22 that the examination imaging has been completed in accordance with a predetermined communication protocol. At this time, the imaging conditions under which the imaging was performed and the imaging execution information such as the irradiation dose are transmitted. In addition, the photographed image is obtained by using, for example, a medical PACS 21 or a medical printer for use in diagnosis according to a medical standard communication protocol called DICOM using image data with the above-described photographing conditions and photographing execution information as supplementary information. Output to 23. A doctor interprets an image stored in the PACS 21 by using an image display device for diagnosis (not shown) and makes a diagnosis.

  FIG. 4 is a diagram for explaining functions realized by the X-ray imaging apparatus 4 of the present embodiment, and shows a process for processing image data captured by the FPD 1. A square in the figure represents hardware or a processing unit realized by the CPU 10 executing predetermined software, and a circle represents data.

  The CPU 10 searches the correction data 72 from the hard disk 12 of the X-ray imaging apparatus 4 by the correction data search unit 61 using the configuration information 70 corresponding to the FPD 1, such as a serial number. The correction data 72 described here is gain correction data for correcting the gain of each pixel of the X-ray image sensor, defective pixel correction data for correcting a defective pixel of the X-ray image sensor, or the like as described above. The CPU 10 reads the display correction data 80 together with the search for the correction data 72. Alternatively, the display correction data is generated from the correction data 72 by the calculation unit 66 realized by the CPU 10. The correction data 72 is data acquired in advance in the calibration process of the X-ray imaging apparatus before imaging of the subject.

  When exposure is performed in the X-ray imaging system, the correction unit 62 performs correction processing on the captured image 71 (corresponding to all the images described above) captured by the FPD 1 using the correction data 72 corresponding to the FPD. And a corrected image 73 is created. As described above, since it takes time to transfer the captured image 71 in the general-purpose I / F, the CPU 10 that executes this embodiment first acquires the thinned image 81 for the captured image 71 and uses this to perform display processing. Etc. That is, the display image processing unit 63 processes the thinned photographed image 81 to create a corrected display image (reduced image) 83 and displays it on the display device 17. The display image processing unit 63 performs correction processing with reference to the display correction data 80 (gain correction data, defective pixel correction data) and the thinned dark image 82 transmitted from the FPD 1, details of which will be described later. To do. The operator looks at the image on the display device 17 to check whether there has been a shooting failure and whether there is a problem with the image processing parameters.

  On the other hand, the corrected image 73 is processed by the image processing means 64 executed by the CPU 10 to create a diagnostic image 74. Subsequently, the image output means 65 is used to output the external device and diagnostic image data 74 to the external device via the LAN.

  FIG. 5 shows an example of a method for creating a thinned image. Reference numeral 90 denotes a part of the pixel configuration representing the image data 71 output from the FPD 1. For example, a pixel block 91 composed of, for example, 8 × 8 pixels is defined as one block, and only 92 as a representative point is sent to the X-ray imaging apparatus 4 as a thinned captured image 81. Then, the pixel 92 is used as the pixel 93 of the thinned image 94 and can be transferred at a high speed as a 1/64 thinned image of all the images.

  When the thinned image shown in FIG. 5 is used, gain correction can be performed by thinning out the gain correction value corresponding to the pixel at the same position from the gain correction data corresponding to the FPD 1 and applying it to the thinned image.

  However, the thinned image in FIG. 5 cannot cope with moire generated on the image due to interference between the grid density and the pixel pitch of the X-ray sensor in imaging using a grid to suppress scattered X-rays. Further, in the thinned image of FIG. 5, it is possible to deal with the defective pixel by using the average of both adjacent pixels, but there are cases where the amount of information for correcting the defective pixel is small and preferable correction cannot be performed. Therefore, in the present embodiment, a more preferable reduced image generation method that can perform gain correction and defective pixel correction and can also deal with moire is proposed. The reduced image generation method of the present embodiment will be described below with reference to FIG.

  In FIG. 6, reference numeral 100 denotes a part of the pixel configuration based on the image data 71 output from the FPD 1. For example, the pixel block 101 of 8 × 8 pixels in the image is one block, and the first line is a pixel having an x coordinate of 8n (n is a block counter) as the representative point of the block, and the second line has an x coordinate of 8n + 1. Pixels are transferred, and so on, and the 8th line transmits a pixel whose x coordinate is 8n + 7. In other words, as shown in FIG. 6, the pixels 102, 103, and 104 shown diagonally black in each block are transmitted as thinned data to be transmitted prior to the data of all images.

  The thinned data thus transferred is stored in the image storage area 8 provided in the RAM 6 of the X-ray imaging apparatus 4 as indicated by 105 in FIG. The 8 pixels of the block 106 correspond to the 8 × 8 pixels of the pixel block 101, the corresponding pixel of the pixel 102 is the pixel 107, the corresponding pixel of the pixel 103 is the pixel 108, and the corresponding pixel of the pixel 103 is the corresponding pixel 108, 104. A pixel corresponds like a pixel 109. In this case, for example, when all the original pixels are 2688 × 2688 pixels, in this embodiment, transmission is performed while thinning out only 1/8 pixel in the main scanning direction of transfer. Therefore, the image size of the thinned image to be transmitted is 336 × 2688 pixels, which is 1/8 of the captured image. In this case, the thinned image can be transferred in about 1 second compared to about 8 seconds required to transfer all the pixels of the photographed image. (Speed of time to display) is not impaired.

  4 performs correction processing on the thinned image as shown in FIG. 6 (even in the case of the thinned image shown in FIG. 5, correction processing for the thinned image in FIG. 6 except for defective pixel correction). The same correction process can be performed. The display image processing unit 63 performs offset noise correction, gain correction, and defective pixel correction. Thereafter, the thinned image composed of 336 × 2688 pixels is converted into a reduced image composed of 336 × 336. In the conversion to the reduced image, the block of 1 × 8 pixels shown in FIG. 6 is converted into one pixel. In the present embodiment, the defective pixel is also corrected at the time of conversion to the reduced image. Hereinafter, it demonstrates in order.

(1) Offset noise correction Generally, the FPD 1 is a dark image for correcting dark noise with a photographed image (the output of the FPD 1 without X-ray exposure, including fixed pattern noise of the X-ray image sensor). Therefore, the dark image is subtracted from the photographed image to remove dark noise. In the present embodiment, this dark image is also thinned out in the same manner as the captured image, and is sent as a thinned dark image 82 prior to the dark image 75 corresponding to all pixels. Therefore, the display image processing unit 63 first performs offset noise correction by subtracting the thinned dark image 82 from the thinned photographed image 81.

(2) Gain Correction On the other hand, when the correction data search unit 61 searches for the gain correction data 72 for performing gain correction for each pixel unique to the X-ray image sensor and loads it from the hard disk 12 to the correction data storage area 9. The decimation gain data 80 is derived and held in the correction data storage area 9 separately. This thinning gain data 80 is obtained by calculating gain correction data 72 storing gain correction values for each pixel into a pixel configuration similar to the thinned image data (thinned and configured by the method shown in FIG. 6). is there. It should be noted that the thinning gain data 80 may be created in advance from the corresponding gain correction data 72 and stored in the hard disk 12. The image processing unit 63 performs gain correction on the thinned image 81 corrected by the thinned dark image using thinning gain data that is a part of the display correction data 80.

(3) Defective pixel correction and conversion to a reduced image of 336 × 336 pixels When a defective pixel is included in the thinned pixel, the defective pixel is displayed as an abnormal pixel on the image, and the display quality is deteriorated. Also, if image processing parameters are generated using such reduced images, there is a possibility that correct parameters cannot be obtained. Therefore, in the present embodiment, the thinned-out defective pixel data converted from the defective pixel data unique to the X-ray image sensor, which is a part of the correction data 72, into the same configuration as the thinned-out image shown in FIG. Generate as part of Then, the display image processing unit 63 performs defect correction on the thinned image data after gain correction using the thinned defective pixel data.

  The image processing unit 63 reduces the thinned image subjected to dark correction, gain correction, and defect correction as described above to 336 × 336 pixels, and the obtained reduced image is displayed on the display device 17 as a corrected preview image 83. . The corrected reduced image is used for image analysis executed by the CPU 10 and used for determining image processing parameters for the captured image.

  By the way, reducing a defective pixel after correcting it by interpolation is equivalent to weighting one pixel by eight vertical pixels. That is, when deriving the thinned defective pixel data from the defective pixel data 72 for the photographed image, the weight of the defective pixel is set to 0, the weight corresponding to the pixel is distributed to other pixels, and defect correction and reduction processing is performed. Executable defective pixel data is generated, and the defect is corrected and reduced simultaneously using this correction data.

  For example, it is assumed that, among the pixels included in the vertical 8 pixel block 110 shown in FIG. 7, the hatched pixels are defective pixels. 111 is an example of a weighting pattern for each pixel included in the pixel block. The weight value pattern 111 is a reference weight value pattern. The total of all 8 pixels is calculated by multiplying the weight value corresponding to the 12-bit data possessed by each pixel, and the total value is divided by the total 26 of the weight values to obtain the pixel value corresponding to the pixel block. In this way, the value of one pixel is acquired from the pixel block of 1 × 8 pixels, and this is performed for all the blocks to obtain a reduced image of 336 × 336 pixels.

  As described above, weighting reduction is performed after interpolation with peripheral pixels as invalid data (= 0) as defective data, and weighting data corresponding to the defect pattern is calculated as shown at 112, Multiplying the weighted data and dividing by 26, the total weight, is equivalent. Furthermore, since the calculation for the defective pixel is not performed, there is an effect of reducing the calculation amount. Therefore, in the present embodiment, for example, when defective pixels are mixed as shown by 110 in FIG. 7, the weight values obtained by distributing the weight values corresponding to the defective pixels from the reference weight value pattern 111 to the surrounding weight values. The pixel value is calculated using the pattern 112.

  As is clear from the above description, in this embodiment, it is necessary to efficiently obtain a weight value pattern to be applied to each block as described with reference to FIG. Below, an example of the structure for the efficient acquisition of such a weight value pattern is demonstrated.

In FIG. 7, it is assumed that the defective pattern pixel 113 is MSB and the pixel 114 is LSB. Then, the defect pattern shown in 110 can be developed into an 8-bit value, and this defect pattern can be expressed as 23H in hexadecimal. That is, a table of 2 ⁸ = 256 types of patterns that can be expressed in 8 bits is created in advance as shown in FIG. 8, and a weight value pattern as shown in the weight value pattern 112 is stored. On the other hand, when the thinned defective pixel data is generated by thinning out the defective pixel data for the photographed image as shown in FIG. 6, the defective pixel pattern of each pixel block is held as an 8-bit value according to the rules described above. When correcting defective pixels in the thinned image as shown in FIG. 6, the table of FIG. 8 can be searched from the numerical values representing the defective pattern of each block from the thinned defective pixel data to obtain the weight value pattern of each block. . For example, the value of the defect pattern corresponding to the pixel block 110 in FIG. 7 is 23H, and the weight value pattern 112 is obtained by searching the table in FIG. Thus, a reduced image can be created by searching the table with a numerical value representing the defect pattern and calculating a weighted average with the obtained weight value pattern.

  Note that the thinned-out image in FIG. 6 is preferable because a weighted average is taken in an oblique direction, so that moire can be effectively suppressed. However, the configuration of the thinned image is the size of each block and the method of thinning out the blocks (in FIG. 6, it is a diagonal pattern with a downward slope to the right, but the pixels are thinned with a diagonal pattern with a downward slope to the left or with a zigzag pattern. The present invention is not limited to FIG. 6 and various modifications are possible. For example, if the third column from the left of each block is thinned out (becomes a vertical linear pattern), the 8n + 2nd pixel of each line in the block may be extracted.

  In the above-described defective pixel correction and reduced image generation, a reference weight value pattern in which a weight value 1 is assigned to all pixels is used, and the sum of the weight values of defective pixels is equally distributed to the weight values of normal pixels. It may be. In this case, a simple average of pixel values in the block 106 (simple average with defective pixels removed) is calculated. For example, in the case of a block having a defective pixel as shown in FIG. 7, the weight value for the normal pixel is 1 + 3/5, the pixel value of each normal pixel is multiplied by the weight value, and the result is divided by 8. As a result, the pixel values of the normal pixels are totaled, and this is equal to dividing this by the number (5 in FIG. 7).

  A processing procedure including acquisition and storage of captured images (all images) in the X-ray imaging apparatus 4 of the present embodiment that acquires reduced images as described above will be described with reference to the flowchart of FIG. In the present embodiment, a thinned image is transmitted in advance at the time of photographing of the subject, and a reduced image is generated by performing correction and the like, and all images are acquired and corrected in the background.

  When the operator selects a radiography method button drawn on the screen by the control program 7 executed by the CPU 10 of the X-ray imaging apparatus 4 and presses the exposure switch 25, the radiographing conditions stored in association with the button are stored. X-ray imaging is performed using. When the CPU 10 detects the end of the X-ray imaging, it requests the FPD 1 to transfer a thinned image in step S201. In step S202, thinned image transfer is executed, and in step S203, reception of the thinned image is completed. At this time, the thinned dark image 82 is also received.

  Subsequently, in step S204, offset correction is performed on the thinned image using offset data (thinned dark image) having a pixel configuration corresponding to the thinned image. In step S205, gain correction is executed using the thinned gain correction data created from the gain correction data. In step S206, thinning defect correction and image reduction are executed as described above. In step S207, image analysis is performed on the reduced image generated in step S206, and image processing parameters are determined in step S208. After the above processing, in step S209, the reduced image is displayed on the display device (40 in FIG. 3) and presented to the operator of the X-ray imaging apparatus. In the present embodiment, it has been described that thinning transfer is performed in order to reduce the circuit scale of the FPD 1, but it is also possible to perform a reduction process on the FPD side.

  In addition, the control program executed by the CPU 10 of the X-ray imaging apparatus 4 acquires all image data in parallel with the above-described reduction correction processing (S204 to S209) after completing the reception of the thinned image in step S203. The processing after step S211 is executed (in the background) in order to save.

  First, in order to acquire all image data, the process proceeds to step S211, and a request for transmission of all image data is sent to the FPD 1. In step S212, all images are transferred, and in step S213, all images are received. At this time, the dark image 75 is also received. When the reception is completed, offset correction of all image data is performed using the dark image 75 in step S215. In step S215, using the gain correction data (included in the correction data 72) of the FPD 1, gain correction is executed on all received images. Next, in step S216, defect correction is executed using defective pixel data of FPD1 (included in correction data 72). In step S217, together with the image processing parameters determined in step S208, all image data that has been corrected in step S216 is stored. Note that the reduced image and the image processing parameters determined from the image may be stored as header information separately from all image data.

<Other embodiments>
In the present invention, a recording medium recording software program codes for realizing the functions of the above-described embodiments is supplied to a system or apparatus, and the CPU of the system or apparatus reads the program codes stored in the recording medium. Needless to say, it can also be achieved through implementation. In this case, the program code itself read from the recording medium realizes the functions of the above-described embodiment, and the recording medium on which the program code is recorded constitutes the present invention. In this case, the program code itself read from the storage medium realizes the novel function of the present invention, and the storage medium storing the program code constitutes the present invention.

  The storage medium for supplying the control program 7 includes, for example, a flexible disk, a hard disk, an optical disk, a magneto-optical disk, a CD-ROM, a CD-R, a magnetic tape, and a nonvolatile memory card in addition to the control program RAM. EPROM or the like can be used.

  Further, by executing the program code read by the computer, not only the functions of the above-described embodiments are realized, but also an OS running on the computer performs actual processing based on an instruction of the program code. Needless to say, a case where the function of the above-described embodiment is realized by performing part or all of the processing is also included.

  Further, after the program code read from the recording medium is written in a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, the function expansion is performed based on the instruction of the program code. It goes without saying that the CPU or the like provided in the board or the function expansion unit performs part or all of the actual processing and the functions of the above-described embodiments are realized by the processing.

  Note that the present invention is also applicable to a case where the program is distributed to a requester via a communication line such as the Internet from a storage medium in which the program code of the software realizing the functions of the above-described embodiments is recorded. is there.

  As described above, according to the X-ray imaging system of this embodiment, the thinned image data of the image obtained by the X-ray imaging is transmitted, and the thinned image data using the correction data corresponding to the X-ray imaging device used. Correction data is generated, and image processing such as offset correction, gain correction, and defect correction is performed on the transmitted thinned image data. Therefore, it is possible to provide a system for determining an X-ray image at a practical speed even with an inexpensive I / F, and to obtain a high-quality reduced image display. In the configuration in which the image processing parameter is determined from the reduced image, a more accurate parameter can be determined. Further, when grid photography is performed, it is possible to suppress a moire generated by weighted average reduction as shown in FIG. 7 using a thinned image as shown in FIG. 6 and to deal with defective pixels.

It is a block diagram which shows the structure of the medical X-ray imaging system of embodiment. It is a figure which shows the example of a screen display of the user interface by an X-ray imaging apparatus. It is a figure which shows the example of a screen display of the user interface by an X-ray imaging apparatus. It is a figure explaining the image processing process in the medical X-ray imaging system of embodiment. It is a schematic diagram which shows general thinning image transfer. It is a schematic diagram which shows the thinned image transfer by embodiment. It is a figure explaining a thinning defect correction | amendment and a reduction coefficient. It is a figure which shows an example of the holding method of the table which implement | achieves a thinning defect pattern and a reduction coefficient. It is a flowchart explaining the process sequence of the X-ray imaging apparatus at the time of imaging | photography.

Claims (7)

An image processing method in an image processing apparatus for processing an image captured using a grid in order to suppress scattered X-rays ,
An acquisition step in which an acquisition unit acquires a thinned image by extracting pixels arranged in a diagonal direction of the rectangular area for each rectangular area of a predetermined size from a captured image obtained by an imaging device including an imaging element;
A generating step for generating defective pixel correction data for a thinned image corresponding to a pixel configuration of the thinned image based on defective pixel data indicating a defective pixel position of the image sensor;
An integration step of integrating the pixels extracted from the same rectangular region in the thinned image to generate a reduced image by integrating the pixels into one pixel ;
Wherein the integrated process, with reference to the defective pixel correction data for the thinned-out image, an image processing method characterized that you exclude data of defective pixels in the decimated image from the subject of the integration.   The image processing method according to claim 1, wherein the display unit of the image processing apparatus further includes a display step of displaying the reduced image.   The image processing method according to claim 1, wherein the parameter generation unit of the image processing apparatus further includes a parameter generation step of generating an image processing parameter for the captured image from the reduced image. An image processing apparatus for processing an image captured using a grid to suppress scattered X-rays,
Acquisition means for acquiring a thinned image by extracting pixels arranged in a diagonal direction of the rectangular area for each rectangular area of a predetermined size from a captured image obtained by an imaging device including an imaging element;
Generating means for generating defective pixel correction data for a thinned image corresponding to a pixel configuration of the thinned image based on defective pixel data indicating a defective pixel position of the image sensor;
An integration unit that generates a reduced image by integrating pixels extracted from the same rectangular region in the thinned image into one pixel ;
Said integrating means refers to the defective pixel correction data for the thinned image, the image processing apparatus characterized that you exclude data of defective pixels in the decimated image from the subject of the integration. An imaging system in which an imaging device and an image processing device for processing an image captured by the imaging device using a grid to suppress scattered X-rays are connected,
A thinned image is obtained by extracting pixels arranged in a diagonal direction of the rectangular area for each rectangular area of a predetermined size from the photographed image obtained by photographing, and the thinned image is acquired prior to transmission of the photographed image. Transfer means for transferring to the image processing device;
In the image processing device, generating means for generating defective pixel correction data for a thinned image corresponding to a pixel configuration of the thinned image based on defective pixel data indicating a defective pixel position of an image sensor in the imaging device;
Integration means for generating a reduced image by integrating the pixels extracted from the same rectangular area in the thinned image image transferred by the transfer means into one pixel ;
Said integrating means includes an imaging system above with reference to the defective pixel correction data for the thinned image, characterized that you exclude data of defective pixels in the decimated image from the subject of the integration. A control program for causing a computer to execute the image processing method according to any one of claims 1 to 3 . Computer-readable storage medium storing a control program for executing the image processing method described by computer to any one of claims 1 to 3.

Images