JP6035474B2 - X-ray equipment - Google Patents

Description

  The present invention relates to an X-ray imaging apparatus suitably used for dentistry and otolaryngology.

  In X-ray imaging technology, in recent years, X-ray imaging apparatuses using solid-state imaging devices have been widely used instead of X-ray photosensitive films.

  Also in X-ray imaging apparatuses used in dentistry and otolaryngology, X-ray imaging apparatuses using a solid-state imaging apparatus are used in various imaging modes such as CT imaging, panoramic imaging, and cephalo imaging.

  As a solid-state imaging device used for such an X-ray imaging apparatus, a CMOS image sensor is known, and among them, a passive pixel sensor (PPS: Passive Pixel Sensor) type is known. The PPS solid-state imaging device includes a pixel array in which PPS pixels including a photodiode that generates an amount of electric charge according to incident light intensity are two-dimensionally arranged in M rows and N columns. In response, the charge generated in the photodiode is changed to a voltage value in the integration circuit, and the voltage value is converted into a digital value and output.

  In general, the output ends of each of the M pixels in each column are connected to the input ends of the integration circuits provided corresponding to the columns through readout wirings provided corresponding to the columns. ing. Then, the charges generated in the photodiodes of each pixel are sequentially input from the first row to the M-th row for each row through the readout wiring corresponding to the column, and are input to the integration circuit. Are sequentially input to the analog / digital conversion circuit from the first column to the Nth column and output as digital values.

  In the X-ray imaging apparatus, the size required for the pixel array of the solid-state imaging device varies depending on the imaging application. For example, in X-ray imaging for dental diagnosis, the width in the longitudinal direction of the pixel array in cephalometric imaging is A long solid-state imaging device of 22 cm or more is required. When such a long solid-state imaging device is required, it may be difficult to manufacture the solid-state imaging device on a single substrate depending on the diameter of the semiconductor wafer used for the production of the solid-state imaging device. is there. In such a case, two substrates that are shorter than the dimensions required for the solid-state imaging device are arranged in the longitudinal direction, and the respective pixel arrangements are combined and used as one solid-state imaging device (so-called tiling). The dimensions can be satisfied.

  However, since it is difficult to eliminate the gap between the end of the substrate and the end of the pixel array produced on this substrate, when two substrates are used side by side, An area (dead area) where an X-ray image is not captured occurs at the boundary portion (joint). Depending on the imaging application, the position of such a dead area may be limited. For example, in dental cephalometric imaging, there is a temporomandibular joint near the center of the X-ray image, so if there is a dead area near the center of the entire tiled pixel array, the image data of the diagnostically important part is lost. There is a risk. In view of this, a solid-state imaging device has been proposed in which the positions of the dead areas are removed from the vicinity of the center by making the longitudinal widths of the pixel arrays on the two substrates different from each other (see Patent Document 1).

JP 2010-136 A

  In the above-described solid-state imaging device of Patent Document 1, it is possible to prevent a loss of image data in a portion important for diagnosis. However, since there is a dead area where an X-ray image is not captured at the boundary portion (joint), the problem that a line is generated in the image of the joint portion has not been solved.

  When two substrates are arranged adjacent to each other, as described above, there is a dead area at the boundary portion (joint). In order to eliminate this dead area as much as possible, there has been proposed a solid-state imaging device that is arranged and tiled so as to overlap an end portion of one substrate with an end portion of the other substrate (see Patent Document 1, FIG. 16C). . However, even when an X-ray image is picked up using such a solid-state image pickup device, there is a problem that a line is generated at a joint portion.

  The present invention has been made to solve the above-described conventional problems, and provides an X-ray imaging apparatus that suppresses the generation of joint lines and the like at the boundary and makes the joint lines inconspicuous. With the goal.

The X-ray imaging apparatus according to the present invention includes an X-ray irradiation unit that irradiates a subject with X-rays, an X-ray detection unit that outputs a digital electrical signal corresponding to incident X-rays, and the X-ray detection unit. A storage unit that stores an output image signal as projection image data; and an image processing unit that performs an image reconstruction operation based on the projection image data stored in the storage unit. A solid-state imaging device that uses a pixel array in which a plurality of pixels formed on a substrate are two-dimensionally arranged to be used as one imaging region, and the image processing unit includes a plurality of substrates arranged side by side from the storage unit. began read as a multiline correction range around including the boundary portion, an average luminance value obtained by averaging all pixel intensity values of each row in the column, the average to the longitudinal axis of each row luminance value, a horizontal axis line mean bright in the boundary portion in the space in which the number of Correcting the projection image data to be linear change in value, and reconstruction operation all projection image data including the corrected projection image data, and calculates the reconstructed image data.

Further, the image processing unit reads the i (i is an integer of 2 or more) rows of the correction range before and after including the boundary portion, the pixels adjacent to both ends of the correction range from the storage unit, the average Brightness value and the standard deviation, mean a correction value is calculated based on the Brightness value and the standard deviation, it may be configured to correct the pixel data of the correction range from the calculated correction value obtained.

Further, the image processing means reads from the storage means the pixels adjacent to the i (i is an integer of 2 or more) rows of the correction range including the boundary portion and both ends of the correction range, and corrects the correction range. mean brightness value (avg _i), the average luminance values of both pixels adjacent (avg _R, avg _L) and the standard deviation of both pixels adjacent (sd _R, sd _L) determined, adjacent both pixels of each correction line R1 = i / (n + 1) (1) The average value AVG _i and the standard deviation SD _i on the straight line connecting
r2 = 1-r1 (2)
SD _i = sd _R × r1 + sd _L × r2 (3)
AVG _i = avg _R × r1 + avg _L × r2 (4)
(Where n is an integer greater than or equal to 2, r1 and r2 are weighting coefficients corresponding to the row adjacent to the correction range and the position of each pixel in the correction range, sd _R is the standard deviation of one adjacent pixel, sd _L is the standard deviation of the adjacent pixels other hand, avg _i is the average luminance value of the range to be corrected, avg _R is the average luminance value of one of the adjacent pixels, avg _L is the other the average brightness values of adjacent pixels )
To calculate and
Using the calculated average luminance value , standard deviation, and average value, correction data IMG [x, y] of each pixel data in the correction range i is
IMG [x, y] = (img [x, y] −avg _i ) × (SD _i / sd _i ) + AVG _i
... (5)
(Here, img [x, y] is each pixel data in the correction range)
What is necessary is just to comprise so that it may calculate by.

In addition, the X-ray imaging apparatus of the present invention includes an X-ray irradiation unit that irradiates a subject with X-rays, an X-ray detection unit that outputs a digital electrical signal corresponding to incident X-rays, and the X-ray detection Storage means for storing the image signal output by the unit as projection image data, and image processing means for performing image reconstruction calculation based on the projection image data stored in the storage means,
The X-ray detection unit includes a solid-state imaging device that uses a pixel array obtained by two-dimensionally arranging a plurality of pixels formed on four or more substrates as one imaging region, and the image processing unit includes: multiple rows of the front and rear, including a plurality of boundary portions where the substrate are arranged side by side from said memory means correction range and to read out, the change in average luminance value at the boundary portion of the projection image data to be linear is corrected, the plurality of rows of the front and rear, including a boundary portion where a plurality of substrates are arranged side by side from said storage means out read as a correction area, an average luminance value obtained by averaging all pixel intensity values of each row in the column The projection image data is corrected so that the change of the average luminance value at the boundary portion is linear in the space where the vertical axis represents the average luminance value for each row and the horizontal axis represents the number of rows , and the corrected projection image data is All projection image data including Reconstruction calculation to to calculate the reconstructed image data.

  According to the present invention, the change in luminance value near the joint (boundary portion) is corrected so as to be linear, and there is no significant change in the luminance value before and after the joint, thereby suppressing the generation of image lines. it can.

It is a side view which shows the X-ray imaging apparatus used for the dentistry and otolaryngology concerning this invention. It is a front view which shows the X-ray imaging apparatus used for the dentistry and otolaryngology concerning this invention. It is explanatory drawing which shows the X-ray imaging apparatus at the time of a cephalometric imaging. It is a top view of the solid-state imaging device used for this invention. The solid-state imaging device used for this invention is shown, (a) is a sectional side view of the solid-state imaging device along the line aa in FIG. 4, and (b) is the solid state along the line bb in FIG. It is a sectional side view of an imaging device. It is a schematic diagram which shows a mode that it images using the solid-state imaging device by which two pixel arrangement | sequences were tiled up and down. FIG. 4 is a side sectional view showing a solid-state imaging device that is arranged and tiled so that the end of one substrate overlaps the end of the other substrate. It is a schematic diagram which shows the area | region containing the boundary part when the image data of the solid-state imaging device using two pixel arrangement | sequences is read, (a) is each n / 2 pixel area | region before and behind the joint of a boundary part. (B) shows the luminance value for each row (line) number. It is a flowchart which shows the image reconstruction process of this invention. It is a flowchart which shows the joint correction process of this invention. It is a flowchart which shows the image reconstruction process of this invention. It is a flowchart which shows the case where the noise process of the image reconstruction process of this invention carries out filter correction in a frequency domain. It is a flowchart which shows the case where the noise process of the image reconstruction process of this invention carries out the filter correction in a spatial domain. It is a schematic diagram showing a solid-state imaging device in which four pixel arrays are tiled vertically and horizontally. FIG. 5 is a schematic diagram in which n / 2 pixel regions are arranged before and after a joint between upper and lower boundary portions of a solid-state imaging device in which four pixel arrays are arranged vertically and horizontally. FIG. 6 is a schematic diagram in which n / 2 pixel regions are arranged before and after a joint between left and right boundary portions of a solid-state imaging device in which four pixel arrays are arranged vertically and horizontally.

  Hereinafter, embodiments of an X-ray imaging apparatus for imaging a subject (patient) will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals, and the description thereof will not be repeated in order to avoid duplication of description.

  FIG. 1 is a side view showing an X-ray imaging apparatus used in dentistry and otolaryngology according to the present invention, and FIG. 2 is a front view thereof. The X-ray imaging apparatus of this embodiment is configured to be capable of panoramic imaging, CT imaging, and cephalometric imaging.

  The X-ray imaging apparatus shown in FIGS. 1 and 2 is a standing type, and includes a support frame 7, a base 73 installed on the floor surface, and a support column 74 erected vertically from the base 73. The X-ray imaging apparatus includes a swing arm 3 for panoramic imaging / CT imaging, an X-ray irradiation unit 2, a panorama / CT X-ray detection unit 1, a head holding unit (not shown), and the like. The X-ray imaging apparatus also includes a cephalometric unit 4 for cephalometric imaging, a cephalometric X-ray detector 5 and the like.

  The support frame 7 is supported by the column 74. The support frame 7 is moved up and down along the column 74 by an elevating mechanism (not shown) according to the height of the patient. The support frame 7 supports the rotary drive unit 70 at the upper part and supports the head support unit 71 at the lower part. As shown in FIG. 2, the support frame 7, the rotation drive unit 70, and the head support unit 71 are configured in a U shape when viewed from the front.

  The rotation drive unit 70 supports the turning arm 3 so as to be rotatable in the horizontal direction. The head support unit 71 includes a head holding portion (not shown) that holds the head of the subject O, and a pair of handles 71b for the subject O to grip. The head support unit 71 includes an operation unit 71a on the distal end side, and an operator operates the lifting position of the support frame 7 with the operation unit 71a, opens / closes the head rod for fixing the head, and performs target imaging during CT imaging. Operations such as setting an area (image reconstruction range), adjusting a positioning laser beam during panoramic photography, and resetting the swivel arm 3 to the original position can be performed.

  The X-ray imaging apparatus includes a cephalo unit 4. The cephalometric unit 4 is attached to a unit arm 41 extending in the horizontal direction. The unit arm 41 is attached to the support frame 7. By raising and lowering the support frame 7, the cephalo unit 4 is raised and lowered according to the height of the subject O via the unit arm 41. The cephalometric unit 4 includes a unit holder 43 at the top thereof. The unit holder 43 includes a head fixing portion 42 that fixes the head of the subject O. The cephalometric unit 4 includes a cephalometric X-ray detection unit 5 adjacent to the head fixing unit 42. The Cephalo X-ray detection unit 5 is held by a sensor holder 8 fixed to the unit holder 43. In this embodiment, the Cephalo X-ray detection unit 5 uses two solid-state imaging devices 50 (see FIG. 3) in which two substrates are arranged in the longitudinal direction and the respective pixel arrays are combined. Yes. The solid-state imaging device 50 is constituted by, for example, a CMOS solid-state imaging device having a plurality of pixels arranged two-dimensionally. The solid-state imaging device 50 will be described later.

  In cephalometric imaging, it is necessary to separate the X-ray irradiation unit 2 and the cephalometric X-ray detection unit 5 by a predetermined distance. Therefore, the cephalo unit 4 is installed via a unit arm 41 having a predetermined length, and the distance between the X-ray irradiation unit 2 and the subject O, the subject O and the cephalo X-ray detection unit 5. Make the distance between the two constant.

  The swivel arm 3 includes a swivel shaft 32 (see FIG. 3) at the top. The pivot shaft 32 is supported by the support frame 7 via the rotation drive unit 70. The swivel arm 3 includes a panorama / CT X-ray detection unit 1 on one side and an X-ray irradiation unit 2 on the other side. The panorama / CT X-ray detection unit 1 and the X-ray irradiation unit 2 face each other. The panorama / CT X-ray detector 1 includes a panorama / CT two-dimensional X-ray detector 10 and exterior covers. The panorama / CT X-ray detection unit 1 is composed of a solid-state imaging device including a CCD sensor, an X-ray indirect conversion method (FPD: flat panel detector) sensor, and the like that are spread two-dimensionally.

  As shown in FIG. 3, the X-ray irradiation unit 2 includes an X-ray tube 21, a collimator 20, and exterior covers. The rotation drive unit 70 includes a turning shaft rotating means 30 for turning the turning arm 3. The revolving arm 3 is revolved horizontally by a revolving shaft rotating means 30 around a revolving shaft 32 extending in a substantially vertical direction. The rotary drive unit 70 includes a turning axis position setting means (not shown) for horizontally moving the turning axis 32 in the XY direction on a substantially horizontal plane. Then, the turning axis position setting means sets the turning axis 32 at the center position of the target imaging region (image reconstruction range).

  As shown in FIGS. 1 to 3, in panoramic imaging and CT imaging, the X-ray focal point of the X-ray tube 21 with respect to the subject (patient) O held by the head holding unit provided in the support frame 7. An X-ray cone beam is irradiated from 2a, and the X-ray transmitted through the subject O is detected by the panorama / CT X-ray detector 1.

  At the time of CT imaging, tomographic imaging of the target imaging region (image reconstruction range) is performed while rotating the swivel arm 3 about the target imaging region (image reconstruction range). Move to match the center position of the reconstruction range. During panoramic imaging, the pan / CT X-ray detector 1 and the X-ray irradiator 2 move the horizontal arm 3 horizontally and horizontally through the rotary shaft 32 so that a predetermined locus along the shape of the dental arch is drawn. Take tomography while turning. The trajectory of the panoramic X-ray beam forms an envelope-like trajectory because X-rays are incident on the dental arch (each tooth) so as to form a normal ray projection.

  As shown in FIG. 3, in cephalometric imaging, the subject O is inserted into the right and left outer ear holes of the subject (patient) O into the ear rods of the head fixing portion 42 provided in the cephalometric unit 4. The X-ray 21a is fixed and irradiated from the X-ray irradiation unit 2 provided on the turning arm 3, and the X-ray 21a transmitted through the subject O is detected by the solid-state imaging device 50 of the Cephalo X-ray detection unit 5. .

  The turning arm 3 includes an X-ray irradiation unit rotating unit 22 for rotating the X-ray irradiation unit 2. The X-ray irradiation unit 2 is rotated by the X-ray irradiation unit rotation unit 22 around a rotation shaft 220 extending in the vertical direction. At the time of cephalometric imaging, the panorama / CT X-ray detector 1 can be removed from the X-ray 21a irradiation field while the X-ray focal point 2a of the X-ray irradiator 2 is directed toward the solid-state imaging device 50 of the cephalo unit 4. .

  In the cephalometric imaging method, the distance (for example, 1500 mm) between the X-ray focal point 2a of the X-ray irradiation unit 2 and the center of the subject O, and the solid-state imaging of the center of the subject O and the Cephalo X-ray detection unit 5 At a distance from the device 50 (for example, 150 mm), the image is arranged and photographed so that the enlargement ratio is 1.1 times. Therefore, the cephalo unit 4 is provided at a predetermined distance from the X-ray irradiation unit 2 provided on the revolving arm 3.

  As shown in FIGS. 1 to 3, the X-ray imaging apparatus includes an operation / display unit 100 and an X-ray imaging apparatus main body 200. The X-ray imaging apparatus main body 200 includes an apparatus main body including a panorama / CT X-ray detection unit 1, a cephalo X-ray detection unit 5, an X-ray irradiation unit 2, a turning arm 3, and the like. (CPU) 35 is mounted. The control unit 35 controls the operation of each driving mechanism such as tube current and tube voltage (imaging conditions) of the X-ray irradiation unit 2. The operation / display unit 100 includes, for example, a personal computer (PC) connected to the control unit 35. The operation / display unit 100 includes an operation unit 136 including a keyboard, an operation unit (remote controller), a mouse, and the like, and an operator selects and inputs a shooting mode, shooting conditions, presence / absence of preliminary shooting, and the like. Further, the operation / display unit 100 includes a control device 110 including a CPU 111 and a ROM 112. The ROM 112 stores a control program such as an OS. The CPU 111 performs an image reconstruction operation, various control operations, and the like according to a program stored in a storage unit 130 or the like which will be described later. The CPU 111 is connected to the operation means 136 via the bus 120, and an operator inputs with an input means (not shown). Based on this input, the CPU 111 sets the turning axis position setting means via the control unit 35. Control is performed so as to perform a predetermined operation.

The operation / display unit 100 includes a storage unit 130. The storage unit 130 includes a work memory 131, an image memory 132, a frame memory 133, a program memory 134, and the like. Image signals output from the panorama / CT X-ray detection unit 1 and the cephalometric X-ray detection unit 5 are stored in the frame memory 133 of the storage unit 130 via the bus 120. The CPU 111 of the control device 110 functions as an image processing unit based on an image reconstruction program stored in the program memory 134 and performs an image reconstruction operation. In CT imaging is processing on the tomographic image along any tomographic plane is stored in the image memory 132 and is displayed on the display unit (monitor) 13 5. In addition, the image memory 132 can be provided with external memory means as necessary.

  The solid-state imaging device 50 converts the X-ray image that has passed through the subject O into electrical image data. Image data output from the solid-state imaging device 50 is once taken into the control device 110 including the CPU 111 from the bus 120 and then stored in the frame memory 133. From the image data stored in the frame memory 133, the CPU 111 performs a predetermined calculation process to reconstruct a tomographic image, a panoramic image, a cephalo image, and the like along an arbitrary tomographic plane.

  4 and 5 are diagrams showing an example of the configuration of the solid-state imaging device 50 used in this embodiment. 4 is a plan view of the solid-state imaging device 50, FIG. 5 (a) is a side sectional view of the solid-state imaging device 50 along the line aa in FIG. 4, and FIG. It is a sectional side view of the solid-state imaging device 50 along a line. 4 and 5 also show an XYZ orthogonal coordinate system for easy understanding.

  As shown in FIGS. 4 and 5A and 5B, the solid-state imaging device 50 includes a semiconductor substrate 53A (first substrate) and a semiconductor substrate 53B (second substrate). One imaging region is configured by the semiconductor substrates 53A and 53B. The size required for the imaging region of the solid-state imaging device 50 varies depending on the imaging application. However, in X-ray imaging for dental diagnosis, the length of the imaging region in the longitudinal direction is 22 cm or more in cephalometric imaging. Is required. Therefore, as in the present embodiment, two semiconductor substrates 53A and 53B shorter than the dimensions required for the solid-state imaging device 50 are arranged in the longitudinal direction, and the pixel arrays 50A and 50B are combined to form one imaging region. By using (so-called tiling), the required dimensions are satisfied. When two semiconductor substrates 53A and 53B are used side by side in this way, an area (dead area C) where an X-ray image is not captured occurs at the boundary portion (joint) of these pixel arrays. This is because it is difficult to manufacture gaps between the end portions of the semiconductor substrates 53A and 53B and the end portions of the pixel arrays 50A and 50B formed on the semiconductor substrates 53A and 53B. The

  The solid-state imaging device 50 includes a pixel array 50A (first pixel array) and a scan shift register 530A each formed on the main surface of the semiconductor substrate 53A, and a pixel array 50B formed on each main surface of the semiconductor substrate 53B. (Second pixel array) and a scan shift register 530B. The solid-state imaging device 50 further includes a signal output unit 520. The signal output unit 520 includes a plurality of signal reading units 521A to 521H formed on the main surface of the semiconductor substrate 53A and the main substrate 53B. A plurality of signal reading units 521I to 521L built in the surface, a plurality of analog / digital (A / D) converters 522A to 522L corresponding to the respective signal reading units 521A to 521L, and each A / D converter 522A And a plurality of FIFO (First-In-First-Out) data buffers 523A to 523L corresponding to .about.522L.

The solid-state imaging device 50 includes a flat base material 52, scintillators 54 </ b> A and 54 </ b> B, and an X-ray shielding member 55. The semiconductor substrates 53A and 53B described above are attached to the base material 52, and the scintillators 54A and 54B are disposed on the semiconductor substrate 53A and the semiconductor substrate 53B, respectively. The scintillators 54A and 54B generate scintillation light according to the incident X-rays, convert the X-ray image into an optical image, and output the optical image to the pixel arrays 50A and 50B, respectively. The scintillators 54A and 54B are respectively installed so as to cover the pixel arrays 50A and 50B, or are provided on the pixel arrays 50A and 50B by vapor deposition. The X-ray shielding member 55 is made of a material such as lead that has an extremely low X-ray transmittance. The X-ray shielding member 55 covers the peripheral portions of the semiconductor substrates 53A and 53B, particularly the regions where the scanning shift registers 530A and 530B and the signal reading units 521A to 521L are arranged. X-ray incidence to 521A to 521L is prevented.

  The pixel array 50A is configured by two-dimensionally arranging M × NA pixels (see FIGS. 5A and 5B) in M rows and NA columns. The pixel array 50B is configured by two-dimensionally arranging M × NB pixels in M rows and NB columns. In FIG. 4, the column direction coincides with the X-axis direction, and the row direction coincides with the Y-axis direction. Each of M, NA, and NB is an integer of 2 or more and satisfies NA> NB. The number of pixels in the row direction (NA + NB) in the pixel arrays 50A and 50B is preferably larger than the number M of pixels in the column direction. In this case, the imaging region including the pixel arrays 50A and 50B has a rectangular shape in which the row direction (Y-axis direction) is a long direction and the column direction (X-axis direction) is a short direction. The pixels are arranged at a pitch of 100 μm, for example, and are of the PPS system and have a common configuration.

  Here, in FIG. 4, among the NA columns included in the pixel array 50A, the column located at the left end (that is, the column with the smallest Y coordinate) is the first column, and the column located at the opposite right end is the NA column. And Also, in the figure, among the NB columns included in the pixel array 50B, the column located at the left end (the column with the smallest Y coordinate) is the first column, and the column located at the opposite right end is the NB column. . In this case, in the present embodiment, the pixel arrays 50A and 50B are arranged so that the first column of the pixel array 50B and the NA column of the pixel array 50A are along each other.

  In addition, one or a plurality of continuous columns including the first column of the pixel array 50A are covered with the X-ray shielding member 55, and are insensitive areas shielded from incident X-rays. That is, no light is incident on these columns and no charge is generated, which does not contribute to imaging. Similarly, one or a plurality of continuous columns including the NB column of the pixel array 50B are also covered with the X-ray shielding member 55, and are insensitive areas. Therefore, in the pixel arrays 50 </ b> A and 50 </ b> B, an effective region for imaging is configured by the other pixel columns except these pixel columns covered by the X-ray shielding member 55. In other words, the effective imaging area in the solid-state imaging device 50 is defined by the opening 55 a of the X-ray shielding member 55.

  The signal output unit 520 holds a voltage value corresponding to the amount of charge output from each pixel, converts the held voltage value into a digital value, and outputs the digital value to the data bus DB. The plurality of signal readout units 521A to 521H are provided corresponding to two or more pixel columns in the pixel array 50A per signal readout unit, and according to the amount of charge output from each pixel of the corresponding pixel column. The voltage value is held, and this voltage value is output to the corresponding A / D converters 522A to 522H. Similarly, the plurality of signal readout units 521I to 521L are provided corresponding to two or more pixel columns in the pixel array 50B for each signal readout unit, and charge output from each pixel of the corresponding pixel column is provided. A voltage value corresponding to the amount is held, and this voltage value is output to the corresponding A / D converters 522I to 522L. At this time, the scan shift registers 530A and 530B control each pixel so that the electric charge accumulated in each pixel is sequentially output to the signal reading units 521A to 521L for each row.

  The plurality of A / D converters 522A to 522L receive the voltage values output from the corresponding signal reading units 521A to 521L, perform A / D conversion processing on the input voltage values (analog values), A digital value corresponding to the input voltage value is generated. The plurality of A / D converters 522A to 522L output the generated digital values to the FIFO data buffers 523A to 523L corresponding to the A / D converters 522A to 522L.

  The plurality of FIFO data buffers 523A to 523L obtain all the digital values corresponding to the NA column included in the pixel array 50A and the NB column included in the pixel array 50B, and then transfer the digital values to the data bus DB. Output. At this time, the FIFO data buffers 523A to 523F are arranged on the left side of the digital values corresponding to the respective columns from the first column to the n-th column (2 ≦ n <NA) of the pixel array 50A (the boundary line E in FIG. 4). 6 digital data stored in the FIFO data buffers 523A to 523F) are sequentially output to the data bus DB. In parallel with this output operation, the FIFO data buffers 523G to 523L are arranged in the columns from the (n + 1) th column of the pixel array 50A to the NB column through the NA column and the first column of the pixel array 50B. Corresponding digital values (digital values stored in the six FIFO data buffers 523G to 523L arranged on the right side of the boundary line E in FIG. 4) are sequentially output to the data bus DB. That is, when viewed from the control device 110 such as the CPU 111 that controls the data bus DB, the six FIFO data buffers 523A to 523F arranged on the left side of the boundary line E constitute one output port. Six FIFO data buffers 523G to 523L arranged on the right side constitute another output port.

  A voltage value corresponding to the amount of charge output from each pixel of the pixel array 50A and the pixel array 50B is held, the held voltage value is converted into a digital value, and output to the data bus DB. The signal is stored in the frame memory 133 of the storage unit 130 via the bus 120.

  As described above, the size required for the pixel array of the solid-state imaging device 50 varies depending on the imaging application. For example, in cephalometric imaging in dental diagnosis, the pixel array of the solid-state imaging device 50 has a length of 22 cm or more. It is required to be a scale. In cephalometric imaging, we know the positional relationship between the skull of the subject (patient) and the maxilla and mandible, and obtain information such as which part to extract or whether the subject (patient) is easy or difficult to correct. This is because in order to obtain such information, the vertical width of the pixel array needs to cover almost the entire head of an adult.

  However, when such a long pixel arrangement is required, it may be difficult to produce the pixel arrangement on a single substrate depending on the diameter of the semiconductor wafer used for the production of the solid-state imaging device. . In such a case, two semiconductor substrates shorter than the dimensions required for the pixel array are arranged in the longitudinal direction, and the respective pixel arrays are combined and used as a single solid-state imaging device (so-called tiling). The dimensions can be satisfied.

  As described above, when two semiconductor substrates are used side by side, as shown in FIG. 4, a dead area C occurs at the boundary portion (joint) between the pixel arrays. And depending on the imaging application, there are cases where there is a restriction on the position of such a dead area C. In the case of X-ray imaging in dental diagnosis, as shown in FIG. 6, two pixel arrays 50A and 50B are tiled in the vertical direction to perform imaging. By the way, when the vertical widths of the two pixel arrays are equal to each other, the boundary portion b2 of the two pixel arrays passes around the ear of the subject O. Note that areas FA and FB shown in the figure indicate imaging ranges by the pixel arrays 50A and 50B, respectively. In cephalometric imaging, information on the region G from the jaw of the subject O to the vicinity including the ears is important, but it is this region G that the boundary portion b2 portion of the two pixel arrays passes through the inside of the region G. This is not preferable because it leads to a lack of information regarding. Therefore, in such a case, the movement path of the boundary portion b between the pixel arrays can be excluded from the region G by making the widths in the longitudinal direction of the two pixel arrays 50A and 50B different from each other.

  In the solid-state imaging device 50 shown in FIGS. 4 and 5, the semiconductor substrate 53A and the semiconductor substrate 53B are arranged on the same plane, and the pixel array 50A and the pixel array 50B are tiled. In such a structure, the distance from the pixel located at the extreme end of each of the pixel array 50A and the pixel array 50B to the respective edges of the semiconductor substrate 53A and the semiconductor substrate 53B and between the semiconductor substrate 53A and the semiconductor substrate 53B. The total distance of the clearances (clearances) to be secured is at least the width of the dead area C. In addition, when the scintillator 54A and the scintillator 54B go around the side surfaces of the semiconductor substrate 53A and the semiconductor substrate 53B, the width of the dead area C is further increased by the amount of the rounding. As shown in FIG. 4, the dead area C does not have a pixel arrangement, and the portion appears in the image as a line.

Therefore, as shown in FIG. 7, in order to eliminate the dead area as much as possible, the solid-state imaging device 50 ₂ and Thaining are arranged so as to overlap an end portion of the other substrate on the end of one of the substrates has been proposed. The solid-state imaging device 50 _2, so that the semiconductor substrate 53B overlaps the end portion of the semiconductor substrate 53A, which are arranged shifted in the Z direction and the semiconductor substrate 53A and the semiconductor substrate 53B. The end of the pixel array 50A of the semiconductor substrate 53A and the end of the end of the pixel array B of the semiconductor substrate 53B are arranged so as to coincide with each other in the horizontal direction. Therefore, the pixel P at one end of the pixel array 50A and the pixel P at one end of the pixel array 50B coincide with each other at the line t, and the dead area can be theoretically eliminated.

However, that line is generated in the joint even when an image of the X-ray image using a solid-state imaging device 50 ₂ shown in FIG. 7 has been confirmed. This is presumably because the luminance of the pixels before and after the joint decreases due to the difference in the thickness of the scintillator at the joint.

Figure 8 shows the data of an area including a joint of the boundary portion when reading the image data of the solid-state imaging device 50 ₂ using the two pixel array having the structure shown in FIG. FIG. 8 shows a region in the longitudinal direction, that is, the Y direction (row (line) direction) including a boundary portion that is a joint between two pixel arrays. FIG. 8A shows n / 2 pixel regions before and after the boundary joint. FIG. 8B shows the luminance value for each number of rows. The solid-state imaging device 50 is configured with a two-dimensional sensor. In this example, 60 pixels are arranged in a column in the depth direction (X direction) in the drawing. The luminance values shown in FIG. 8 (b) represents the mean Luminance value obtained by averaging all pixel intensity values of the depth direction columns. In FIG. 8, n of the correction range is 20 rows (lines).

  As shown in FIG. 8B, the average luminance value changes greatly in the vicinity of the joint (boundary portion). For this reason, a line is generated in the vicinity of the joint of the pixel arrangement in an image such as a cephalo image. As shown in FIG. 6, the image is reconstructed as an image having a line at the boundary portion b. Note that the solid-state imaging device 50 shown in FIGS. 4 to 5 is also reconstructed as an image in which there is a line at the boundary portion b including the dead area C portion.

  Therefore, in the present invention, the change in the average luminance value in the vicinity of the joint (boundary portion) is corrected to be linear as shown by the broken line in FIG. 8B to eliminate the change before and after the joint. This suppresses the generation of image lines.

  As shown in FIG. 8, each row of the captured image from i = 1 to n before and after the joint (boundary portion) is calculated as a correction range. In FIG. 8, the pixel arrangement in the Y direction, that is, the long direction is shown from the left to the right. In FIG. 8A, L is the correction range (i), for example, the row adjacent to the correction range row in the lower pixel array 50A, and R is the correction range in the upper pixel array 50B. A line adjacent to the line is shown. That is, L indicates a row adjacent to a row where i = 1, and R indicates a row adjacent to a row where i = n.

Correction range (i = 1 to n) and Luminance value column of the row L for Luminance value of each row in the column and adjacent to the i = 1 of the correction range, column of the row R adjacent to correct the range of i = n based of the Luminance value, it performs a correction process of the joint. Hereinafter, correction processing according to the present invention will be described.

Correction processing, first, the control CPU111 device 110 accesses the frame memory 133, from the image data, correction range (i = 1 to n) of each row lines adjacent to the Luminance value and the correction range of the column of L, adjacent reads the Brightness value of each column of the row R of the average Luminance value column of each row of the correction range (i = 1 to n), adjacent rows L, and the average Luminance values of adjacent rows R, A standard deviation is calculated, and each calculated data is stored in the work memory 131. Here, the average Luminance value of the range to be corrected avg _i, averaged Luminance value avg _L adjacent rows L, and the average Luminance values of adjacent rows R and avg _R. The solid-state imaging device 50 is configured by a two-dimensional sensor. In this example, 60 pixels are arranged in a column in the depth direction (X direction) in the figure, and therefore, all the pixel luminance values in the depth direction column are obtained. The average brightness value obtained by averaging is obtained.

Further, the control circuit 110 uses the obtained average luminance value and standard deviation as sd _i , the standard deviation of the adjacent row L as sd _L , and the standard deviation of the adjacent row R as sd _R. The CPU 111 calculates an average value AVG _i and a standard deviation SD _i on a straight line connecting the row L and the row R adjacent to each correction row based on the following expressions, and stores the values in the work memory 131.

r1 = i / (n + 1) (1)
r2 = 1-r1 (2)
SD _i = sd _R × r1 + sd _L × r2 (3)
AVG _i = avg _R × r1 + avg _L × r2 (4)

  Here, r1 and r2 are weighting coefficients corresponding to the row adjacent to the correction range and the position of each pixel in the correction range, r1 is a coefficient corresponding to the adjacent row R, and r2 is the adjacent row L. Is a coefficient corresponding to.

Calculated average Luminance value, standard deviation, using the average value is calculated based on the correction data of each pixel data of the correction range i in the following equation.

IMG [x, y] = (img [x, y] −avg _i ) × (SD _i / sd _i ) + AVG _i
... (5)

  Here, IMG [x, y] is pixel data after correction of each x, y coordinate, and img [x, y] is pixel data before correction of each x, y coordinate.

  Each pixel data (x, y) in the correction range is read, each pixel data is corrected based on the above equation (5), and each pixel data is stored in the work memory 131 as corresponding pixel data. Then, using the corrected pixel data as image reconstruction data, image reconstruction is performed by a shift addition method or a filtered back projection method (FBP method, FBP: Filtered Back Projection) to obtain an image such as a cephalometric image.

Next, the operation of the present invention will be described with reference to the flowcharts of FIGS.
There are a shift addition method and an FBP method as image data reconstruction methods. Since the shift addition method and the FBP method have different reconstruction principles, the resolution and contrast in the tomographic direction are different. For this reason, each reconstruction method has the following advantages.

  First, the advantages of the shift addition method will be described. When the shift addition method is used, metal artifacts do not occur, and it becomes easy to diagnose how the bone is attached to the implant body. In addition, caries near the inlay and crown can be easily diagnosed. Furthermore, reconstruction calculation is fast, and it is possible to quickly display an image after diagnosis and make a diagnosis.

  The advantages of the FBP method will be described. When the FBP method is used, a high-contrast image can be created, and it becomes easy to diagnose the apex, root canal, periodontal ligament and the like.

  Thus, the shift addition method and the FBP method have their respective advantages, and it cannot be said that any reconstruction method is absolutely superior. Therefore, a reconstruction method that satisfies each requirement may be used.

  In this embodiment, the shift addition method as the reconstruction method is collected at a constant rate by scanning, and each pixel data near the joint (boundary portion) is corrected by the above-described method. Then, the frame data (image data) including the corrected data is shifted from each other by an amount corresponding to the position, for example, and mutually added.

  The FBP method as a reconstruction method corrects each pixel data near the joint (boundary part) as described above. A group of frame data (pixel data) including each corrected pixel data is individually subjected to a predetermined first weighting process, and then each image data after the first weighting process is subjected to a predetermined Perform convolution processing. Next, a predetermined second weighting process is performed on each projection data after the convolution process. Next, the image data after the second weighting process is individually subjected to a predetermined back projection (back projection: BP) process to generate a BP image.

Next, the processing operation of the present invention will be described.
First, when preparation for imaging such as positioning of the subject O is completed, the CPU 111 of the control device 110 responds to an operator instruction given through the operation unit 136 and instructs a scan for data collection. Then, pulsed or continuous wave X-rays are emitted from the X-ray irradiation unit 2 at a predetermined cycle or continuously. As a result, the X-rays emitted from the X-ray irradiation unit 2 pass through the subject O and are detected by the solid-state imaging device 50 of the X-ray detection unit 5. Then, as described above, digital frame data (pixel data) reflecting the X-ray transmission amount is output from the solid-state imaging device 50 of the X-ray detection unit 5. This frame data is temporarily stored in the frame memory 133.

  When the scan command is completed, the CPU 111 of the control device 110 functions as an image processing unit and performs image reconstruction processing. When the image reconstruction process is started, the CPU 111 of the control device 110 reads the captured image stored in the frame memory 133 and temporarily stores it in the work memory 131 (step S1).

  Subsequently, the CPU 111 of the control device 110 performs defect correction (step S2) and density correction (step S3) on the read captured image. Subsequently, the defect corrected and density corrected image data is logarithmically converted (step S4), and the logarithmically converted data is stored in the work memory 131.

  Subsequently, the CPU 111 of the control device 110 corrects the change in the average luminance value in the vicinity of the joint (boundary portion) so as to be linear, and eliminates the change before and after the joint to suppress the generation of the line of the image. A joint correction processing operation is performed (step S5).

When the processing operation for joint correction is started, as shown in FIG. 10, in order to calculate each row of captured images from i = 1 to n before and after the joint (boundary portion) as a correction range, the control device 110. the CPU111 accesses the work memory 131, the image data obtained by logarithmic transformation, the correction range (i = 1 to n) of each row of the row respectively adjacent to the Luminance value and the correction range of columns L, adjacent rows R reads the Brightness value of a column, a correction range (i = 1 to n) average Luminance value avg _i column of each row of adjacent rows L, mean Luminance value avg _L adjacent rows R, and avg _R Calculate the standard deviation. Then, the standard deviation of the correction range is calculated as sd _i , the standard deviation sd _{L of} the adjacent row _L , and the standard deviation sd _R of the adjacent row _R are calculated. Then, each calculated data is stored in the work memory 131 (step S51). The solid-state imaging device 50 is configured by a two-dimensional sensor. In this example, 60 pixels are arranged in a column in the depth direction (X direction) in the figure, and therefore, all the pixel luminance values in the depth direction column are obtained. The average brightness value obtained by averaging is obtained.

Subsequently, the CPU 111 of the control device 110 calculates an average value AVG _i and a standard deviation SD _i on a straight line connecting the row L adjacent to the both ends of each correction range and the row R adjacent to each other, and the values are calculated. Store in the work memory 131 (step S52). For this calculation, the CPU 111 of the control device 110 first calculates r1 and r2 based on the above equations (1) and (2), and calculates the calculated r1 and r2 and the standard deviation sd of the adjacent row L. _L, based on the above equation (3) using the standard deviation sd _R adjacent rows R, the standard deviation SD _i. Furthermore, CPU 111 of the controller 110, calculates the line L adjacent the calculated r1 and r2, the average luminance value avg _L of R, the average Luminance value AVG _i based on the equation (4) using avg _R To do. This process is repeated up to 1 to n lines of the correction range to obtain the average luminance value and the standard deviation correction value.

Mean Brightness value calculated in step S52, the standard deviation, using the average value, the pixel data of the correction range i is calculated based on the equation (5) (step S53). The CPU 111 of the control device 110 reads out each pixel data (x, y) in the correction range, corrects each pixel data based on the above equation (5), and stores each pixel data in the work memory 131 as corresponding pixel data. Store (step S53). When the correction processing for all the pixels in the correction range is completed, the correction processing ends and the process returns to the image reconstruction processing.

  Then, the data in the correction range near the joint (boundary portion) is replaced with the data corrected by the correction processing described above, and the replaced frame data (image data) is shifted to each other by an amount corresponding to the position. Shift addition processing is performed to reconstruct the image, and the reconstructed reconstruction data is stored in the image memory 132 (step S6). Then, the reconstruction process ends. Thereafter, the reconstructed image is displayed on the display unit 135.

  As described above, when the shift addition method is used, metal artifacts do not occur, so that it is easy to diagnose how the bone is attached to the implant body. In addition, caries near the inlay and crown can be easily diagnosed. Furthermore, reconstruction calculation is fast, and it is possible to quickly display an image after diagnosis and make a diagnosis. In addition, there is little noise, and it becomes easy to make a diagnosis even in a portion where the dose is slightly short, such as a median portion or a temporomandibular joint portion.

  Next, a case where image reconstruction is performed by the FBP method will be described with reference to FIGS. 11 to 13. Similar to the reconstruction by shift addition, digital amount frame data (pixel data) reflecting the X-ray transmission amount is output from the solid-state imaging device 50 of the X-ray detection unit 5, and this frame data is temporarily stored in the frame memory 133. Is done. Then, the CPU 111 of the control device 110 reads the captured image stored in the frame memory 133 and temporarily stores it in the work memory 131 (step S1).

  Subsequently, the CPU 111 of the control device 110 performs defect correction (step S2) and density correction (step S3) on the read captured image. Subsequently, the defect corrected and density corrected image data is logarithmically converted (step S4), and the logarithmically converted data is stored in the work memory 131.

  Subsequently, the CPU 111 of the control device 110 corrects the change in the average luminance value in the vicinity of the joint (boundary portion) so as to be linear, and eliminates the change before and after the joint to suppress the generation of the line of the image. A joint correction processing operation is performed (step S5). Then, the seam correction process shown in FIG. 10 is performed. Then, the data in the correction range near the joint (boundary portion) is replaced with the data corrected by the correction processing described above, and the replaced frame data (image data) is reconstructed by reconstructing the image by the FBP method. The reconstructed data is stored in the image memory 132 (step S6a). In the FBP method, the image data in the vicinity of the joint (boundary portion) is replaced with the above-described correction data, and a group of data of the frame data (pixel data) of all the pixels including the correction data is individually set to the predetermined first data. A weighting process is performed, and then a predetermined convolution process is performed on each image data after the first weighting process. Next, a predetermined second weighting process is performed on each projection data after the convolution process. Next, the image data after the second weighting process is individually subjected to a predetermined back projection (back projection: BP) process to generate a BP image. Thereafter, the reconstructed image is displayed on the display unit 135.

  Next, in the FBP method, processing when noise processing is subjected to filter correction in the frequency domain will be described with reference to FIG.

  The CPU 111 of the control device 110 obtains image data obtained by replacing the data in the correction range near the joint (boundary portion) with the data corrected by the above correction processing from the work memory 131 (step S61).

  The CPU 111 of the control device 110 performs Fourier transform on the corrected image data to create spectrum data of the corrected image data (step S62).

  Subsequently, the CPU 111 of the control device 110 performs a filter process by taking the product of the spectrum data of the corrected image data and the noise removal filter (step S63). Then, the CPU 111 of the control device 110 performs inverse Fourier transform on the spectrum data of the image data to which the filter is applied, and obtains filter-corrected image data (step S64).

  Thereafter, the image data subjected to the filter correction is back-projected to obtain reconstructed image data, which is stored in the image memory 132 (step S65). Then, the reconstructed image data is read from the image memory 132 and displayed on the display unit 135 as a diagnostic image (step S66).

  Next, according to FIG. 13, processing when noise processing is filter corrected in the spatial domain will be described.

  The CPU 111 of the control device 110 obtains image data obtained by replacing the data in the correction range near the joint (boundary portion) with the data corrected by the correction processing described above from the work memory 131 (step S61a).

  The CPU 111 of the control device 110 performs inverse Fourier transform on the noise removal filter to create a spatial domain filter (step S62a).

  Subsequently, the CPU 111 of the control device 110 performs convolution integration between the corrected image data and the spatial domain filter (step S63a).

  Thereafter, the image data subjected to the filter correction is backprojected to obtain reconstructed image data, which is stored in the image memory 132 (step S65a). Then, the reconstructed image data is read from the image memory 132 and displayed on the display means 135 as a diagnostic image (step S66a).

In the above-described embodiment, the solid-state imaging device 50 (50 ₂ ) using two semiconductor substrates side by side has been described. However, the present invention is not limited to two semiconductor substrates to be tiled, but three or more semiconductor substrates. Can be applied. That is, the area before and after the boundary portion may be set as a correction range, and the seam correction according to the present invention may be applied to the range. In the above-described embodiment, the widths in the longitudinal direction of the two pixel arrays 50A and 50B are different from each other. However, the present invention can also be applied when the vertical widths of the two pixel arrays are equal to each other. Needless to say.

In the above-described embodiment, the solid-state imaging device 50 (50 ₂ ) using two semiconductor substrates arranged in the vertical direction has been described. However, FIG. 14 illustrates solid-state imaging using four semiconductor substrates arranged in the vertical and horizontal directions. It shows the device 50 _3. This solid-state imaging device 50 ₃ has a pixel array 50A, a pixel array 50B, a pixel array 50C, and a pixel array 50D arranged above and below, a pixel array 50A and a pixel array 50C, and a pixel array 50B and a pixel array 50D arranged on the left and right. Each is tiling. In such a structure, the pixel array 50A and the pixel array 50B in the row direction, the pixel array 50C and the pixel array 50D, the pixel array 50A and the pixel array 50C in the column direction, and the pixel array 50B and the pixel array 50D are connected to each other. Part) will exist. That is, there are joints (boundary portions) on the top, bottom, left, and right.

  As described above, the luminance value greatly changes near the joint (boundary portion). Therefore, according to the present invention, the change in luminance value near the joints (boundary portions) in the upper, lower, left, and right directions is corrected to be linear, and the change before and after the joints is eliminated to suppress the generation of image lines. Is.

  In the upper and lower pixel arrays 50A and 50B, the pixel array 50C and the pixel array 50D, as shown in FIG. 15, each row of the captured images from before and after i = 1 to n including the upper and lower joints (boundary portions). Is calculated as a correction range. This vertical correction is performed in the same manner as the above correction.

  That is, the correction is performed as described above, with each row of the captured image from i = 1 to n including the upper and lower joints (boundary portions) between the pixel array 50A and the pixel array 50B as a correction range. Further, each row of the captured image from i = 1 to n before and after the upper and lower joints (boundary portions) of the pixel array 50C and the pixel array 50D is corrected as described above as a correction range. Based on the above equation (5), each pixel data is corrected, and the data in the correction range near the upper and lower joints (boundary portion) is replaced with the image data corrected by the correction processing described above, and the upper and lower joints (boundary portion). The correction process is performed.

  Subsequently, the right and left joints (boundary portions) are corrected. In the left and right pixel arrays 50A and 50C, and the pixel array 50B and pixel array 50D, as shown in FIG. 16, each of the captured images from j = 1 to m including the left and right joints (boundary portions) is included. The column is calculated as a correction range. In FIG. 16, the pixel arrangement in the X direction is shown from the left to the right. Then, XL is the correction range (j), for example, the column adjacent to the correction range column in the pixel array 50B (50A) located on the left side, and XR is the correction range column in the pixel array 50D (50C) located on the right side. Adjacent columns are shown. That is, XL indicates a column adjacent to the column where j = 1, and XR indicates a column adjacent to the column where j = m.

And Luminance values of columns XL adjacent to j = 1 of the bright degree value and the correction range of each column of the correction range (j = 1~m), the Luminance value of the column XR adjacent correction range of j = m Based on this, the left and right joints (boundary portions) are corrected.

Correction processing of the left and right joint (boundary) from the image data corrected in the row direction, the column adjacent to the correction range and Luminance values of rows in each column of the correction range (j = 1 to m) XL, adjacent reads the Brightness value of each row of the column XR for the average Luminance values of rows in each column of the correction range (j = 1 to m), adjacent columns XL, the average Luminance values of adjacent rows XR Calculate the standard deviation. The average Luminance value avg _j range correcting, avg _XL average Luminance values of adjacent columns XL, the average Luminance values of adjacent rows XR and avg _XR.

Further, the standard deviation sd _j ranges to correct, standard deviation sd _XL adjacent columns XL, the standard deviation of the adjacent rows XR as sd _XR, using the average luminance value and the standard deviation determined, the corrected column The average value AVG _j and the standard deviation SD _j on the straight line connecting the columns XL and XR adjacent to each other are calculated based on the following formulas, and the values are stored.

k1 = j / (m + 1) (11)
k2 = 1-k1 (12)
SD _j = sd _XR × k1 + sd _XL × k2 (13)
AVG _j = avg _XR × k1 + avg _XL × k2 (14)

  Here, k1 and k2 are weighting coefficients corresponding to the columns adjacent to the correction range and the positions of the respective pixels in the correction range, k1 is a coefficient corresponding to the adjacent column XR, and k2 is the adjacent column XL. Is a coefficient corresponding to.

Calculated average Luminance value, standard deviation, using the average value is calculated based on the correction data of each pixel data of the correction range j to the following equation.

IMG [x, y] = (img [x, y] −avg _j ) × (SD _j / sd _j ) + AVG _j
... (15)

  Here, IMG [x, y] is pixel data after correction of each x, y coordinate, and img [x, y] is pixel data before correction of each x, y coordinate.

  Each pixel data (x, y) in the correction range is read out, each pixel data is corrected based on the above equation (15), each pixel data is replaced with the corresponding pixel data, and a horizontal joint (boundary portion) The correction process is performed. Then, image data such as a cephalometric image is obtained by performing image reconstruction by a shift addition method or a filter back projection method using pixel data in which correction of upper, lower, left, and right joints (boundary portions) is performed as image reconstruction data. Get.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is shown not by the above description of the embodiments but by the scope of claims for patent, and is intended to include all modifications within the meaning and scope equivalent to the scope of claims for patent.

DESCRIPTION OF SYMBOLS 1 X-ray detection part for panorama / CT 2 X-ray irradiation part 3 Revolving arm 4 Cephalo unit 5 Cefaro X-ray detection part 50 Solid-state imaging device 7 Support frame 73 Base 74 Prop 100 Operation / display part 110 Control apparatus 111 CPU
DESCRIPTION OF SYMBOLS 130 Memory | storage part 131 Work memory 132 Image memory 133 Frame memory 134 Program memory 135 Display means 136 Operation means

Claims (5)

Projection image data includes an X-ray irradiation unit that irradiates the subject with X-rays, an X-ray detection unit that outputs a digital electrical signal corresponding to the incident X-rays, and an image signal output by the X-ray detection unit Storage means for storing, and image processing means for performing image reconstruction calculation based on the projection image data stored in the storage means,
The X-ray detection unit includes a solid-state imaging device that uses a pixel array in which a plurality of pixels formed on a plurality of substrates are two-dimensionally arranged as a single imaging region, and the image processing unit includes the storage unit multiple rows of the front and rear, including a plurality of boundary portions where the substrate are arranged side by side from the unit as a correction area out read, an average luminance value obtained by averaging all pixel intensity values of each row in the column, a vertical axis line by line The projection image data is corrected so that the change of the average luminance value at the boundary portion is linear in the space with the horizontal axis representing the average luminance value and the horizontal axis, and all the projection image data including the corrected projection image data An X-ray imaging apparatus characterized by calculating reconstruction image data by performing reconstruction calculation.   The X-ray imaging apparatus according to claim 1, wherein the X-ray detection unit is a Cephalo X-ray detection unit that performs Cephalo imaging. Said image processing means reads the i (i is an integer of 2 or more) rows of the correction range before and after including the boundary portion, the pixels adjacent to both ends of the correction range from the storage unit, the average Luminance value and the standard a deviation, calculated to calculate the correction value from the average Brightness value and the standard deviation, according from the calculated correction value to claim 1 or claim 2, wherein the correcting the pixel data of the correction range X X-ray equipment. The image processing means reads out from the storage means i (i is an integer of 2 or more) rows of the correction range before and after the boundary portion, and both ends of the correction range, and reads out the average brightness of the correction range. The degree value (avg _i ) , the average luminance value (avg _R , avg _L ) of both adjacent pixels, and the standard deviation ( sd _R , sd _L ) of both adjacent pixels are obtained, and the adjacent pixels in each correction row are connected. The average value AVG _i on the straight line and the standard deviation SD _i are r1 = i / (n + 1) (1)
r2 = 1-r1 (2)
SD _i = sd _R × r1 + sd _L × r2 (3)
AVG _i = avg _R × r1 + avg _L × r2 (4)
(Where n is an integer greater than or equal to 2, r1 and r2 are weighting coefficients corresponding to the row adjacent to the correction range and the position of each pixel in the correction range, sd _R is the standard deviation of one adjacent pixel, sd _L is the standard deviation of the adjacent pixels other hand, avg _i is the average luminance value of the range to be corrected, avg _R is the average luminance value of one of the adjacent pixels, avg _L is the other the average brightness values of adjacent pixels )
To calculate and
Using the calculated average luminance value , standard deviation, and average value, correction data IMG [x, y] of each pixel data in the correction range i is
IMG [x, y] = (img [x, y] −avg _i ) × (SD _i / sd _i ) + AVG _i
... (5)
(Here, img [x, y] is each pixel data in the correction range)
The X-ray imaging apparatus according to claim 3, which is calculated by: Projection image data includes an X-ray irradiation unit that irradiates the subject with X-rays, an X-ray detection unit that outputs a digital electrical signal corresponding to the incident X-rays, and an image signal output by the X-ray detection unit Storage means for storing, and image processing means for performing image reconstruction calculation based on the projection image data stored in the storage means,
The X-ray detection unit includes a solid-state imaging device that uses a pixel array obtained by two-dimensionally arranging a plurality of pixels formed on four or more substrates as one imaging region, and the image processing unit includes: multiple rows of the front and rear, including a plurality of boundary portions where the substrate are arranged side by side from said memory means correction range and to read out, the change in average luminance value at the boundary portion of the projection image data to be linear is corrected, the plurality of rows of the front and rear, including a boundary portion where a plurality of substrates are arranged side by side from said storage means out read as a correction area, an average luminance value obtained by averaging all pixel intensity values of each row in the column The projection image data is corrected so that the change of the average luminance value at the boundary portion is linear in the space where the vertical axis represents the average luminance value for each row and the horizontal axis represents the number of rows , and the corrected projection image data is All projection image data including Reconstruction calculation to, X-rays imaging apparatus characterized by calculating the reconstructed image data.

Images