JP2015080519A - Radiation imaging device and method for controlling the same, radiation image processing device and method, program and computer readable storage medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a device and method capable of determining whether or not a correction object pixel in which loss occurs in the radiation is appropriately corrected.SOLUTION: A radiation imaging device acquires a radiation image on the basis of the amount of radiation detected by a radiation detection unit, and corrects a first pixel that is a correction object present in the radiation image by using a second pixel other than the first pixel. Then, the radiation imaging device removes a periodic signal from each of a radiation image after the correction and a radiation image before the correction, and determines whether or not the correction is appropriate on the basis of the radiation image after the correction from which the periodic signal is removed and the radiation image before the correction from which the periodic signal is removed.

Description

  The present invention relates to a technique for correcting pixels present in a radiation image (for example, an X-ray image) in which radiation has been lost.

  In recent years, flat panel detectors (hereinafter referred to as FPD) that accumulate X-rays as charge signals and convert them into digital signals to provide diagnostic images have been put into practical use. Such an X-ray imaging apparatus using an FPD is configured to perform imaging while synchronizing the X-ray irradiation by the X-ray generator and the imaging operation by the FPD.

  On the other hand, there are many market demands to replace existing modality films and imaging plate portions with FPDs. When an imaging unit having such an existing modality is to be replaced with an FPD, it may be difficult to construct an interface for synchronizing the X-ray generator and the FPD. Patent Document 1 proposes an FPD that detects X-ray irradiation on the FPD side and automatically starts an accumulation operation without providing an interface between the X-ray generator and the FPD.

  However, when the X-ray irradiation is automatically detected and the accumulation operation is started on the FPD side, a certain amount of X-ray irradiation is necessary until the X-ray is detected and the accumulation operation is started. Will be done. For this reason, the charge due to the irradiated X-rays is discharged after the X-ray irradiation is actually started until the accumulation operation is started by detecting the X-rays, and the discharged charges contribute to the output value. Can not. Accordingly, a pixel in which charges are discharged and X-ray information is lost (hereinafter, this pixel is referred to as an X-ray defect pixel) and a pixel from which charges are not discharged (hereinafter, this pixel is referred to as a non-X-ray defect pixel). ) Has a problem that the output value varies even with the same incident X-ray dose.

  For example, in Patent Document 1, since the reset operation at the time of X-ray detection is performed for each line and the even-numbered line and the odd-numbered line are switched for each frame, the charge is discharged from the X-ray irradiation to the start of the accumulation operation. Is every other line. As a result, a line from which charges are discharged (hereinafter referred to as an X-ray defect line) and a line from which charges are not discharged (hereinafter referred to as a non-X-ray defect line) are alternately generated. There is a problem that a difference in output value between lines appears in a stripe pattern on an image. Regarding this problem, Patent Document 1 discloses a method of discarding X-ray defect line data as a defect and correcting it by linear interpolation of surrounding pixels.

  In addition, as a method other than the above-described method, a method of obtaining an X-ray defect rate in an X-ray defect line from a non-X-ray defect line adjacent to the X-ray defect line and digitally amplifying an output value according to the defect rate has recently been developed. Proposed. In this method, since the information of the X-ray defect line can be effectively used as compared with the method of discarding the X-ray defect line as a defect, a more preferable correction result can be obtained.

  By the way, in addition to the loss of X-rays due to the reset operation as described above, a signal unrelated to the subject may be superimposed on the X-ray image. For example, there may be a case where an instrument called a grid that removes scattered rays generated when X-rays pass through the inside of the subject is placed between the subject and the radiation receiving surface to perform imaging. This grid removes scattered rays by arranging radiation shielding materials such as lead and radiation transmitting materials such as aluminum and carbon alternately with a predetermined width. However, since a part of the direct line passing through the radiation shielding material is also removed when the grid is arranged, a periodic signal (also referred to as a grid stripe) is generated on the image.

JP2011-249891A

  Not only the reset operation but the X-ray deficient line and the X-ray deficient pixel as described above deteriorate the image quality, and correcting this is necessary to obtain an appropriate diagnostic image. However, when the X-ray information changes due to another factor due to a grid or the like, such a change in the X-ray information adversely affects the correction of the X-ray defect pixel described above, and the image is further deteriorated by the correction. There is. For example, if the grid is arranged under the condition that the direction of the grid stripes and the direction of the scanning lines are parallel, the direction of the stripes due to the X-ray defect line and the grid stripes generated when X-rays are automatically detected Will match. In the method of Patent Document 1, in such a case, the fringes of both interfere with each other and correction cannot be performed properly, and the image quality may be deteriorated by performing the correction. Note that the above-described problem can be solved by arranging the grid stripe direction and the scanning line direction to be perpendicular to each other. However, there is a possibility that the grid is erroneously arranged, and in this case, there is a possibility that correction cannot be performed properly.

  The present invention has been made in view of the above problems, and an exemplary object thereof is to provide an apparatus and a method capable of determining whether or not a correction target pixel in which a loss has occurred in radiation is appropriately corrected. It is in.

In order to achieve the above object, a radiation imaging apparatus according to an aspect of the present invention has the following arrangement. That is,
An acquisition means for acquiring a radiation image based on the radiation dose detected by the radiation detection means;
Correction means for correcting the first pixel to be corrected, which exists in the radiation image, using a second pixel other than the first pixel;
Removing means for removing a periodic signal from each of the radiation image after correction by the correction means and the radiation image before correction;
Determining means for determining the suitability of correction by the correcting means based on the corrected radiographic image from which the periodic signal is removed and the uncorrected radiographic image from which the periodic signal is removed.

  Determining whether or not the radiation deficient pixels that are generated when the radiation image is acquired by automatically detecting the radiation exposure is appropriately corrected, so that it is possible to suppress image quality deterioration due to the fact that the correction cannot be performed appropriately. .

1 is a configuration diagram of an entire X-ray imaging apparatus according to a first embodiment. The flowchart which shows the process sequence in 1st embodiment. FIG. 6 illustrates a circuit configuration of an FPD. The figure explaining the drive of FPD. The figure explaining the fall of the output value in a X-ray defect line. The figure explaining the calculation method of a correction coefficient. The figure explaining the difference in the correction coefficient by the presence or absence of a grid. The figure explaining the weighting function of local regression. The figure explaining the decreasing rate of a pixel value. The block diagram of the whole X-ray imaging apparatus by 2nd embodiment. The flowchart which shows the process sequence in 2nd embodiment.

<First embodiment>
Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings. The present invention is applied to a radiographic apparatus that acquires a radiographic image from a radiation dose detected by a radiation detection layer unit that detects radiation and performs radiographic image processing, for example, an X-ray imaging apparatus 100 as shown in FIG. . When the X-ray imaging apparatus 100 automatically detects X-ray irradiation and acquires an X-ray image, the X-ray imaging apparatus 100 corrects an X-ray deficient pixel in the X-ray image that is generated due to the loss of X-rays by a reset operation. A function of executing X-ray image processing for the purpose.

  In the X-ray imaging apparatus 100 as described above, an X-ray tube 101 that is an example of a radiation generation unit irradiates a subject 103 with X-rays. The X-ray generator 104 applies a high voltage pulse to the X-ray tube 101 when an exposure switch (not shown) is pressed to generate X-rays. The FPD 102 is controlled by the FPD control unit 105 to convert X-rays transmitted through the subject 103 into visible light using a fluorescent substance and detect it with a photodiode. The detected electrical signal is AD converted and transmitted to the FPD control unit 105. The FPD control unit 105 includes an image processing unit 109 and an image storage unit 108, and includes one or more computers (not shown).

  The computer included in the FPD control unit 105 includes a main control unit such as a CPU and a storage unit such as a ROM (Read Only Memory) and a RAM (Random Access Memory). Further, the computer may include a graphic control unit such as a GPU (Graphics Processing Unit), a communication unit such as a network card, an input / output unit such as a keyboard, a display, or a touch panel. Each of these components is connected by a bus or the like, and is controlled by the main control unit executing a program stored in the storage unit. The monitor 106 displays the received digital signal or the digital signal processed by the image processing unit 109 as an image. The operation unit 107 inputs an instruction from the user to the image processing unit 109 and the FPD 102. The image storage unit 108 stores the digital signal output from the FPD control unit 105 and the image data processed by the image processing unit 109. The image processing unit 109 corrects X-ray deficient pixels in the image captured by the FPD 102, and includes a first correction coefficient calculation unit 110, a second correction coefficient calculation unit 111, a correction unit 112, and periodic signal removal. Unit 113 and determination unit 114.

  In the X-ray imaging apparatus 100 having the above configuration, the operation of the characteristic part of the present embodiment will be specifically described with reference to the flowchart shown in FIG. First, a high voltage pulse is applied to the X-ray tube 101 by the X-ray generator 104 to irradiate the subject 103 with X-rays. After the X-ray irradiation is started, the FPD 102 starts an accumulation operation by automatically detecting the X-ray irradiation and acquires an X-ray image (S201).

  Here, the driving for automatically detecting and determining the start of X-ray irradiation will be specifically described with reference to FIGS. FIG. 3 shows a circuit configuration of the FPD 102. The FPD 102 has pixels formed of TFTs 302 and photodiodes 303, and several thousand pixels are formed on the glass substrate 301 both vertically and horizontally. Note that FIG. 3 shows an FPD composed of four pixels both vertically and horizontally for explanation. The scanning line control circuit 306 sequentially applies an ON signal to G1 to G4 to turn on the TFT 302. G1 to G4 are scanning lines. When an ON signal is applied to each scanning line, the TFT 302 is switched on, and the output charge of the photodiode 303 can be read line by line. S1 to S4 are signal lines. The charges read from the photodiode 303 are transmitted through the signal lines and read by the signal detection circuit 305. The signal detection circuit 305 holds the read signal, processes it such as amplification, AD conversion, and outputs it as a digital signal to the FPD control unit 105. A power supply 304 supplies an operating voltage to each photodiode 303.

  FIG. 4 is a timing chart showing driving of the FPD 102 in this embodiment. When the X-ray is not irradiated, the FPD 102 stands by for the X-ray detection drive. At this time, the scanning line control circuit 306 sequentially drives the scanning lines by odd numbers such as G1, G3... In the first frame. Then, the dark charges of the pixels connected to the odd-numbered scanning lines are read and reset. In the next second frame, the scanning line control circuit 306 sequentially drives only the even lines such as G2, G4,..., And resets the pixels connected to the even-numbered scanning lines. Thus, the odd lines are reset in the odd frames, and the odd lines are reset alternately by resetting the even lines in the even frames.

  Here, the signal detection circuit 305 of the FPD control unit 105 monitors the reset charge to detect X-ray irradiation. When the X-rays are irradiated, charges are generated by the photodiode 303, so that the output of the signal line is increased. When this output exceeds a predetermined threshold value, the FPD control unit 105 considers that X-rays have been irradiated and starts the accumulation operation by turning off all the TFTs 302.

  Incidentally, X-ray irradiation can be detected more quickly as the threshold for detecting the above-mentioned X-ray irradiation is smaller, but erroneous detection due to noise or the like is likely to occur. Therefore, it is necessary to set the threshold value to a certain value in consideration of such a malfunction. As a result, there is a considerable time lag between the start of X-ray irradiation and the detection of X-ray irradiation. Due to this time lag, the charge due to the X-rays irradiated until the accumulation operation is started cannot contribute to the output value, and becomes smaller than the original output value. In other words, a correction target line in which correction target pixels (referred to as defective pixels in this specification) in which X-rays are lost is arranged is generated.

  FIG. 5 is a schematic diagram showing a decrease in the output value when uniform X-rays are irradiated. Here, FIG. 5A shows one rectangle as one pixel. FIG. 5B is a bar graph showing the sum of the output values of the pixels in each row of FIG. When uniform X-rays are irradiated, the sum of the output values of each row is substantially the same if there is no time lag until X-ray irradiation is detected. Here, it is assumed that X-ray irradiation is started on the even line G4 and X-ray irradiation is detected on the even line G8. In this case, X-rays are lost in the even-numbered lines G4, G6, and G8 in which the reset operation is performed as shown in FIG. 5B, and the value becomes smaller than the original output value. These lines G4, G6, and G8 are correction target lines. On the other hand, X-ray loss due to the reset operation does not occur in other lines, and these lines are pixels that do not need to be corrected (that is, pixels other than the correction target pixels, hereinafter referred to as non-correction target pixels). ) Are non-correction target lines.

  Note that the decrease in the output value depends on the time lag from when the X-ray irradiation is started until the reset operation is performed, and the decrease becomes larger as the time lag increases. Therefore, in the operation of this embodiment, since the reset operation is sequentially performed from G1, the output values decrease greatly from G4 to G6, G8 from which X-ray irradiation is started as shown in FIG. 5B.

  Also, the decrease in output value depends on the shooting conditions. For example, as the X-ray dose is increased, the proportion of the electric charge discharged to detect the X-ray irradiation contributes to the output value is small and can be visually ignored depending on the imaging conditions (correction of X-ray deficient pixels) In some cases.

  As described above, the X-ray image obtained by the operation as described above has portions in which correction target lines in which correction target pixels are arranged and non-correction target lines in which non-correction target pixels are arranged alternately exist. Then, such an X-ray image is transferred to the image processing unit 109 and a correction target line (hereinafter referred to as an X-ray defect line) whose output value is reduced is corrected. A specific correction method will be described below.

First, the correction coefficient calculation unit 110 calculates a correction coefficient for correcting the output value of the X-ray defect line (S202). Usually, the output value V of each pixel (hereinafter, the output value of each pixel may be referred to as a pixel value) is the sum of a gain component proportional to the X-ray dose X incident on the pixel and an offset component D due to dark current or the like. If the proportionality constant is A, the relationship between the pixel value V and the X-ray dose X is as follows.

In contrast, the pixel values V _d of the X-ray-deficient line, part of the X-ray dose X that is incident on the pixel is lost by the reset operation, the X-ray dosage contributes to the pixel value is reduced to 1 / G times. Therefore, the relationship between the pixel values V _d and X-ray dose X of the X-ray defective line is as follows.

なお、式（３）のＧ，Ｄの値は未知であり、これら２つを補正係数として算出することで、Ｘ線欠損ラインの画素値Ｖ_dを本来の画素値Ｖに補正することができる。 Therefore, the relationship between the pixel value V _d of the X-ray defect line and the original pixel value V is as follows from the equations (1) and (2).

Note that the values of G and D in Equation (3) are unknown, and the pixel value V _d of the X-ray defect line can be corrected to the original pixel value V by calculating these two as correction coefficients. .

Therefore, in this embodiment, the correction coefficients G and D are calculated by regression analysis using pixel values of adjacent non-X-ray defect lines (that is, non-correction target lines) that have a high correlation with the X-ray defect lines. More specifically, the pixel value of each column i of the X-ray defect line is {x _i | i = 1, 2,..., N} as shown in FIG. Let the pixel values of _i be {y _{i, 1} | i = 1,2,..., n} and {y _{i, 2} | i = 1,2,. If it is assumed that the pixel values in each column have substantially the same value, the relationship between the pixel value x of the X-ray defect line and the pixel value y of the non-X-ray defect line is expressed by the following equation (4).

Here, as in Expression (4), the relationship between x and y is expressed by a linear expression of slope G and intercept D · (1-G), and G and D can be calculated by linear regression analysis. . For example, when least square regression is used as a method of linear regression analysis, a slope a and an intercept b that minimize the error E represented by the equation (5) are obtained, and corrected by the equation (6) from the obtained slope a and the intercept b. The coefficients G and D may be calculated.

  Here, in the present embodiment, the method using the least square regression has been described, but the present invention is not limited to this method, and the correction coefficient can be similarly set even if an already known method such as MA regression or RMA regression is used. Can be calculated. In this embodiment, the correction coefficient is obtained on the assumption that the pixel values of the respective columns have substantially the same value. However, when a steep edge or the like exists, there is a pixel for which this assumption does not hold. Therefore, as a countermeasure against such outliers, known robust regression such as M estimation, LMedS estimation, and RANSAC can be used.

  Although the calculation method of the correction coefficients G and D for one X-ray defect line has been described above, the same processing is performed for all X-ray defect lines, and the correction coefficients G and D for each X-ray defect line are obtained. calculate. Since the timing at which X-ray irradiation is started is unknown, it is unknown which line is the X-ray defect line. Therefore, in the present embodiment, the line where the reset operation is sequentially performed from the line where the reset operation was performed at the timing when the X-ray irradiation was detected is traced back, and the correction coefficient for the line of ceil (total number of lines / 2) −1. G and D are calculated. For example, in FIG. 5B, for all 10 lines, G6, G4, G2 and G2, which are not actually X-ray deficient lines, are sequentially from G8, which is reset when X-ray irradiation is detected. Including four lines of correction coefficients G and D are obtained. Note that a line to be corrected may be traced back a predetermined number of lines from the line where the reset operation was performed at the timing when X-ray irradiation was detected. In this case, the predetermined number of lines may be obtained in advance by calculation or experiment.

  Here, the correction coefficients G and D are obtained on the assumption that the pixel values of adjacent non-X-ray defect lines having a high correlation with the X-ray defect lines take substantially the same values after correction. However, when the grid is arranged under the condition that the grid stripe direction and the scanning line direction are parallel to each other, the correlation with the adjacent non-X-ray defect line may be low, and this assumption is not valid. Therefore, the periodic signal of the grid is superimposed on the correction coefficient. Therefore, the second correction coefficient calculation unit 111 calculates a correction coefficient obtained by removing the influence of the periodic signal caused by the grid from the correction coefficient obtained by the first correction coefficient calculation unit 110 (S203).

  By the way, FIG. 7 shows a graph of the correction coefficient G calculated from the line where the reset operation was performed at the timing when the X-ray irradiation was detected to the line where the reset operation was sequentially performed. In FIG. 7, lines that have been reset at the timing when X-ray irradiation is detected are designated as line number 0, and numbers are assigned sequentially. Here, when the grid is not mounted or when the grid is arranged under the condition that the direction of the grid stripes and the direction of the scanning line are orthogonal to each other, an accurate correction coefficient G as shown in FIG. Can be calculated. However, when the grid is arranged under the condition that the direction of the grid stripes and the direction of the scanning line are parallel to each other, the periodic signal due to the grid is superimposed as shown in FIG. The correct correction coefficient G cannot be calculated due to vibration. Although not shown, the same applies to the correction coefficient D. Therefore, a correction coefficient from which such a periodic signal is removed is calculated.

In the present embodiment, the periodic signal of the grid is removed using LOWESS, the line number in FIG. 7 is i, and the correction coefficient at the line number i is G _i. Gc _{i from} which the periodic signal is removed is calculated for each line.

Here, a _i and b _i in the equation (7) are unknown, and these values are calculated by local regression with the line number i as the center. More specifically, when calculating the regression coefficients a _i and b _i for the line number i, a value is calculated error E _i is the minimum of the following formula by a least square approximation.

Note that w _k is a weighting function as shown in FIG. 8, which is 1 when the line number is i, the value decreases as the distance from the sample k to be referred to increases from i, and becomes 0 when d exceeds. . Therefore, the correction coefficient Gc obtained above is a linear approximation of the correction coefficient G for each line in the range of ± d. Therefore, if d is set to a value somewhat larger than the grid period, the grid periodic signal cannot be applied to a straight line, and therefore the grid periodic signal can be removed as shown in FIG.

  Here, although the value of d is not particularly limited, for example, in this embodiment, it is set to 5 times the period of the grid. The period of the periodic signal of the grid may be calculated in advance according to the density of the installed grid. Further, it may be calculated from the power spectrum of the correction coefficient G.

  The method for calculating Gc obtained by removing the periodic signal of the grid from the correction coefficient G has been described above. However, for the correction coefficient D, Dc obtained by removing the periodic signal of the grid is calculated by the same method. In the present embodiment, the linear expression is used as the regression expression, but the present invention is not limited to this, and a polynomial may be used as the regression expression. It is also possible to use Robust LOWESS or the like that is robust to abnormal values.

  In this embodiment, the periodic signal of the grid is removed by smoothing using local regression. Alternatively, the periodic signal of the grid may be removed by smoothing using a low-pass filter or the like. Further, the periodic signal of the grid may be removed by modeling the waveform of the correct correction coefficient and fitting the correction coefficient G to this model.

Next, the correction unit 112 corrects the defective pixel in the X-ray image using the obtained correction coefficients Gc and Dc (S204). Specifically, the pixel value of each column i of the X-ray defect line k is {V _k (i) | i = 1, 2,..., N}, and the correction coefficients for the line are Gc _k and Dc _k. For example, the corrected pixel value V ′ _k (i) is calculated for all X-ray defect lines by the following equation.

Next, the periodic signal removal unit 113 removes the periodic signal of the grid superimposed on the image by filtering (S205). Specifically, if the frequency of the grid on the image is fg (rad / sample), the grid stripes are removed using the Nth-order FIR filter h calculated by the following equation (0).

Note that the FIR filter calculated by the equation (10) is a low-pass filter that blocks frequencies higher than the grid frequency fg. Here, the density of the grid that is generally used is selected so as to have a high frequency on the image in consideration of the influence on the low-frequency component that is the main structure of the image. At 0.5π (rad / sample) or more. Therefore, in this embodiment, by setting f _g = 0.5π, an FIR filter corresponding to the density of the grid that is generally used is calculated.

  Next, the grid is removed using the FIR filter described above. Here, since the purpose is to remove a grid of stripes parallel to the direction of the scanning line, filtering may be performed with the FIR filter in the direction orthogonal to the scanning line on the image. In this embodiment, since the images before and after correction are used in the subsequent stage, filtering is performed on both the image before correction and the image corrected by the correction unit 112.

  Next, the determination unit 114 determines whether or not the correction performed in S204 is appropriate (s206). Specifically, except for the periodic signal of the grid, it is assumed that the corrected pixel values of non-X-ray deficient lines that are close to the X-ray deficient line have substantially the same value, and whether the corrected image is appropriate or not. judge.

  Here, in the driving according to the present embodiment, as described above, the rate of decrease in the pixel value is the largest in the line where X-ray irradiation is detected, and the rate of decrease decreases in order and becomes the minimum in the line where X-ray irradiation is started. . Also, X-ray defect lines occur every other line. A graph of this pixel value reduction rate is shown in FIG. 9A (in FIG. 9, the line number where X-ray irradiation is detected is represented as 0). Here, when the periodic signal removal described above is performed, the vibration components for every other line having a peak at the Nyquist frequency are removed together with the periodic signal of the grid, resulting in a wedge-shaped waveform as shown in FIG.

Therefore, in the present embodiment, it is evaluated whether or not the correction is appropriately performed by evaluating the step of the wedge-shaped waveform. Specifically, the pixel value of each column i of the line where the X-ray irradiation is detected is {x _i | i = 1, 2,..., N}, and is lower on the image (left in FIG. 9B). Let {y _{i, m} | i = 1, 2,..., N} be the pixel value of each column i of non-X-ray deficient lines that are separated by m lines. Then, assuming that the pixel values of both columns take substantially the same values when corrected appropriately, the relationship between the pixel value x of the X-ray defect line and the pixel value y of the non-X-ray defect line is as follows: It becomes.

ここで、ｍは任意の値を設定すれば良く、特に限定するものではないが本実施の形態では例えば１０とする。 Therefore, the slope a that minimizes the error E expressed by the following equation is obtained by least square approximation, and if the slope a is approximately 1, it can be determined that appropriate correction has been made.

Here, m may be set to an arbitrary value and is not particularly limited, but is set to 10 in the present embodiment, for example.

In the present embodiment, it is determined whether there is a possibility that the image quality may be deteriorated by performing correction. Therefore, the inclination a is obtained by the equation (12) for the line where the X-ray irradiation before and after the correction is detected, and when the inclination after the correction approaches 1 with respect to the inclination before the correction, that is, the step becomes small. In this case, it is determined that the correction is appropriate. Specifically, if the inclination obtained from the line before correction in S204 is a _o and the inclination obtained from the line after correction in S204 is a _c , it is determined that the correction is appropriate when the following condition is satisfied. To do.

  Further, the reason why the correction is not appropriately performed is when the influence of the periodic signal due to the grid is large. That is, when the correction coefficient obtained by the first correction coefficient calculation unit 110 is as shown in FIG. 7A, correction is appropriately performed. Even when the periodic signal due to the grid is superimposed as shown in FIG. 7B, if the grid amplitude or period is relatively small with respect to the wedge-shaped waveform to be obtained, the influence is small and correction is performed. Can be done appropriately. Therefore, the determination accuracy can be further improved if the determination by the equation (13) is performed only when the influence of the periodic signal due to the grid is large.

Therefore, in this embodiment, whether the influence of the periodic signal due to the grid is large using the correction coefficient G calculated by the first correction coefficient calculation unit 110 and the correction coefficient Gc calculated by the second correction coefficient calculation unit 111. Determine whether or not. Specifically, if the correction coefficients for the line number i are G _i and G c _i , respectively, the determination coefficient R ² is calculated by the following equation.

ここで、ＴＨはグリッドの影響度合いを判定する閾値であり、本実施形態では例えば０．９５とする。 Here, the coefficient of determination R ² is made of a smaller value the larger the difference between a value of 0 to 1, the correction coefficient G and Gc. Therefore, it superimposed periodic signal due to the grid as in FIG. 7 (b), and the larger the impact difference is increased, determined as a result the coefficient R ² takes a small value. Therefore, it is determined that the influence due to the grid is large under the following conditions.

Here, TH is a threshold value for determining the degree of influence of the grid, and is set to, for example, 0.95 in this embodiment.

  As described above, it is determined whether the correction is appropriate using the two determination criteria described above. Specifically, it is determined that the correction is not properly performed when the expression (15) is satisfied and the expression (13) is not satisfied, and it is determined that the correction is appropriately performed under other conditions. For example, the suitability of correction may be determined using only one of Equation (13) and Equation (15).

  Here, in this embodiment, when it is determined that the correction has not been properly performed, the correction result is rejected, and the image before correction is stored in the image storage unit 108 as processed data. Thereby, it is possible to reduce the possibility of deteriorating the image quality by the correction result. The case where the correction cannot be appropriately performed is a case where the grid amplitude and period are relatively large with respect to the wedge-shaped waveform to be obtained as described above. In other words, the wedge-shaped waveform to be corrected is relatively very small and is often at a level that can be visually ignored without correction.

If it is determined that the correction has not been properly performed, the image before correction is stored as processed data in the image storage unit 108, and the operator is notified via the monitor 106 that correction has not been performed. May be explicitly notified.
Second Embodiment

  The present invention is applied to, for example, an X-ray imaging apparatus 1000 as shown in FIG. The X-ray imaging apparatus 1000 shown in FIG. 10 is configured to include a third correction coefficient calculation unit 1001 with respect to the X-ray imaging apparatus 100 shown in FIG. In the second embodiment, the processing procedure of the image processing unit 109 is different from that of the first embodiment, and is an operation according to the flowchart shown in FIG.

  In the X-ray imaging apparatus 1000 of FIG. 10, the same reference numerals are given to portions that operate in the same manner as the X-ray imaging apparatus 100 of FIG. 1, and details thereof are omitted. In the flowchart shown in FIG. 11, the same steps as those in the flowchart shown in FIG. 2 are denoted by the same reference numerals, and only the configuration different from that of the first embodiment will be specifically described here. In S201 to S206, the same processing as that in the first embodiment is executed, and the suitability of the correction in S204 is determined.

  Here, in the first embodiment, when it is determined that the correction is not properly performed, the image before correction is stored in the image storage unit 108 as processed data. On the other hand, in this embodiment, when it is determined that the correction is not properly performed (NO in S1100), the third correction coefficient calculation unit 1001 calculates the correction coefficient again (S1101).

Here, the reason why the correction was not properly performed is that, as described above, the influence of the periodic signal due to the grid is large, and as a result, the second correction coefficient calculation unit 111 could not sufficiently remove the influence. Conceivable. Therefore, the third correction coefficient calculation unit 1001 calculates the correction coefficient by LOWESS in the same manner as the second correction coefficient calculation unit 111, but by changing the parameter used at that time, the periodic signal caused by the grid is calculated. A correction coefficient that removes the influence more strongly is calculated.

Specifically, the correction coefficient is calculated again by increasing d in the above equation (8). Here, d is a parameter for setting the width for performing linear regression. The larger this value is, the stronger the influence of the periodic signal can be removed. If this value is too large, the correct correction coefficient will not be applied properly, and the two are in a trade-off relationship.

  Therefore, in this embodiment, d is sequentially increased to obtain an optimal solution that makes correction appropriate. Specifically, the width is increased by, for example, 5% with respect to d set by the second correction coefficient calculation unit 111, and the correction coefficient is calculated (S1101). Next, the X-ray defect line is corrected using the correction coefficient newly obtained in S204. Next, a correction result is determined in S205 to S206. If the correction result is not appropriate (S1100), the width of d is further increased by 5% and the correction result is obtained again (S1101).

  The above operation is repeatedly executed until it is determined that the correction result is appropriate. Thereby, even when the influence of the periodic signal due to the grid is large, it is possible to perform appropriate correction.

  As mentioned above, although preferable embodiment of this invention was described, it cannot be overemphasized that this invention is not limited to these embodiment, A various deformation | transformation and change are possible within the range of the summary. For example, in the above-described embodiment, a case has been described in which correction target lines in which correction target pixels are arranged and non-correction target lines in which non-correction target pixels are arranged alternately. However, the present invention is not limited to this. For example, if the correction target pixel in which X-ray loss has occurred is an image arranged so as to be correctable by the non-correction target pixel in which X-ray loss has not occurred, the correction and noise level reduction according to the above embodiment are applied. can do. Therefore, for example, the present invention may be applied to a configuration in which reset is performed for each column instead of a configuration in which reset is performed in units of lines. In addition, for example, it is obvious that the above-described processing can be applied to an X-ray image in which correction target pixels and non-correction target pixels are arranged in a lattice pattern. Furthermore, in the above-described embodiment, an example in which reset is performed line by line as the reset operation in units of lines has been described. However, a plurality of even lines or odd lines (for example, “G2 and G4”, “G6 and G8” in FIG. May be reset at the same time.

  Moreover, although the said embodiment demonstrated the case where the periodic signal formed by insertion of a grid was removed, it is not restricted to this.

Further, the present invention supplies a software program (in the embodiment, a program corresponding to the flowchart shown in the figure) to the system or apparatus directly or remotely, and the computer of the system or apparatus implements the functions of the above-described embodiments. Is also achieved by reading and executing the supplied program code.

  Accordingly, since the functions of the present invention are implemented by computer, the program code installed in the computer also implements the present invention. In other words, the present invention includes a computer program itself for realizing the functional processing of the present invention.

  Computer-readable recording media for supplying the program include, for example, hard disks, optical disks, magneto-optical disks, MO, CD-ROM, CD-R, CD-RW, magnetic tape, nonvolatile memory cards, ROM, There are DVDs (DVD-ROM, DVD-R).

As another program supply method, a client computer browser is used to connect to an Internet homepage, and the computer program of the present invention itself or a compressed file including an automatic installation function is downloaded from the homepage to a recording medium such as a hard disk. Can also be supplied. It can also be realized by dividing the program code constituting the program of the present invention into a plurality of files and downloading each file from a different homepage. That is, a WWW server that allows a plurality of users to download a program file for realizing the functional processing of the present invention on a computer is also included in the present invention.

  In addition, the program of the present invention is encrypted, stored in a storage medium such as a CD-ROM, distributed to users, and key information for decryption is downloaded from a homepage via the Internet to users who have cleared predetermined conditions. It is also possible to execute the encrypted program by using the key information and install the program on a computer.

  In addition to the functions of the above-described embodiments being realized by the computer executing the read program, the OS running on the computer based on the instructions of the program is a part of the actual processing. Alternatively, the functions of the above-described embodiment can be realized by performing all of them and performing the processing.

Claims (24)

An acquisition means for acquiring a radiation image based on the radiation dose detected by the radiation detection means;
Correction means for correcting the first pixel to be corrected, which exists in the radiation image, using a second pixel other than the first pixel;
Removing means for removing a periodic signal from each of the radiation image after correction by the correction means and the radiation image before correction;
And a determination unit that determines whether or not correction by the correction unit is appropriate based on the corrected radiographic image from which the periodic signal is removed and the uncorrected radiographic image from which the periodic signal is removed. Radiography equipment. The correction means includes
First calculation means for calculating a correction coefficient for correcting the first pixel from the second pixel adjacent to the first pixel;
Second calculation means for calculating a correction coefficient obtained by removing the influence of the periodic signal from the correction coefficient calculated by the first calculation means,
The radiation imaging apparatus according to claim 1, wherein the first pixel is corrected using the correction coefficient calculated by the second calculation unit.   3. The radiation according to claim 2, wherein the first calculation unit calculates the correction coefficient by performing regression analysis on a relationship between pixel values of the second pixel adjacent to the first pixel. 4. Shooting device.   4. The method according to claim 1, wherein when the determination unit determines that the correction is not properly performed, the correction result of the correction unit is rejected and the correction by the correction unit is not performed. The radiation imaging apparatus of Claim 1.   The radiographic apparatus according to claim 1, wherein when the determination unit determines that correction is not properly performed, the fact is notified. And further comprising third calculation means for calculating a correction coefficient obtained by removing the influence of the periodic signal from the correction coefficient by a method different from the second calculation means,
The correction unit corrects the first pixel using the correction coefficient calculated by the third calculation unit when the determination unit determines that the correction is not appropriately performed. The radiation imaging apparatus according to claim 2 or 3.   The second calculation means or the third calculation means removes the influence of the periodic signal by smoothing the correction coefficient calculated by the first calculation means. The radiation imaging apparatus described.   The second calculating means or the third calculating means removes the influence of the periodic signal by fitting the correction coefficient calculated by the first calculating means to a modeled correction coefficient. The radiation imaging apparatus according to claim 6.   The determination unit performs regression analysis on the radiation images of the first pixel before and after correction by the correction unit, and determines whether the correction by the correction unit is appropriate based on the results of both regression analysis. The radiation imaging apparatus according to any one of claims 1 to 8.   The determination unit determines whether the correction by the correction unit is appropriately performed based on a difference between the correction coefficient calculated by the first calculation unit and the correction coefficient calculated by the second calculation unit. The radiation imaging apparatus according to claim 2, wherein the radiation imaging apparatus is a radiography apparatus.   The radiation imaging apparatus according to claim 1, wherein the first pixel is a pixel in which loss of radiation information has occurred due to a reset operation of the radiation detection unit.   12. The radiation image is an image having a portion where correction target lines in which the first pixels are arranged and non-correction target lines in which the second pixels are arranged alternately exist. The radiation imaging apparatus according to any one of the above. Means for determining the start of radiation irradiation based on a signal read at the time of resetting while resetting the radiation detection means in units of lines;
The correction target line and the non-correction target line in the radiographic image are obtained as a result of the reset being repeated after the radiation detection unit is irradiated with radiation until the unit for determining the irradiation start determines that the irradiation is started. The radiographic apparatus according to claim 12, wherein a portion in which and are alternately present is generated.   14. The periodic signal superimposed on the radiographic image acquired by the acquisition unit is generated by a grid arranged between the radiation detection unit and a radiation generation unit. The radiation imaging apparatus according to any one of the above. An acquisition means for acquiring a radiation image based on the radiation dose detected by the radiation detection means;
In order to correct the correction coefficient for correcting the first pixel to be corrected, which exists in the radiographic image, using a second pixel adjacent to the first pixel other than the first pixel. First calculating means for calculating a correction coefficient of
Second calculation means for calculating a correction coefficient obtained by removing the influence of a periodic signal from the correction coefficient calculated by the first calculation means;
A radiation imaging apparatus comprising: correction means for correcting the first pixel using the correction coefficient calculated by the second calculation means. Correction means for correcting the first pixel to be corrected, which exists in the radiation image, using a second pixel other than the first pixel;
Removing means for removing a periodic signal from each of the radiation image after correction by the correction means and the radiation image before correction;
And a determination unit that determines whether or not correction by the correction unit is appropriate based on the corrected radiographic image from which the periodic signal is removed and the uncorrected radiographic image from which the periodic signal is removed. A radiographic image processing apparatus. A correction coefficient for correcting the first pixel to be corrected, which exists in the radiographic image, is corrected using a second pixel adjacent to the first pixel other than the first pixel. First calculating means for calculating a correction coefficient;
Second calculation means for calculating a correction coefficient obtained by removing the influence of a periodic signal from the correction coefficient calculated by the first calculation means;
A radiation image processing apparatus comprising: correction means for correcting the first pixel using the correction coefficient calculated by the second calculation means. An acquisition step of acquiring a radiographic image based on the radiation dose detected by the radiation detection means;
A correction step of correcting the first pixel to be corrected existing in the radiation image using a second pixel other than the first pixel;
A removal step of removing a periodic signal from each of the radiographic image after correction by the correction step and the radiographic image before correction,
A determination step of determining whether or not correction by the correction step is appropriate based on the corrected radiographic image from which the periodic signal is removed and the uncorrected radiographic image from which the periodic signal is removed. Control method for radiation imaging apparatus. An acquisition step of acquiring a radiographic image based on the radiation dose detected by the radiation detection means;
In order to correct the correction coefficient for correcting the first pixel to be corrected, which exists in the radiographic image, using a second pixel adjacent to the first pixel other than the first pixel. A first calculation step of calculating a correction coefficient of
A second calculation step of calculating a correction coefficient obtained by removing the influence of a periodic signal from the correction coefficient calculated in the first calculation step;
And a correction step of correcting the first pixel using the correction coefficient calculated in the second calculation step. A correction step of correcting the first pixel to be corrected existing in the radiation image using a second pixel other than the first pixel;
A removal step of removing a periodic signal from each of the radiographic image after correction by the correction step and the radiographic image before correction,
A determination step of determining whether or not correction by the correction step is appropriate based on the corrected radiographic image from which the periodic signal is removed and the uncorrected radiographic image from which the periodic signal is removed. A radiation image processing method. A correction coefficient for correcting the first pixel to be corrected, which exists in the radiographic image, is corrected using a second pixel adjacent to the first pixel other than the first pixel. A first calculation step of calculating a correction coefficient;
A second calculation step of calculating a correction coefficient obtained by removing the influence of a periodic signal from the correction coefficient calculated in the first calculation step;
And a correction step of correcting the first pixel using the correction coefficient calculated in the second calculation step.   The program for functioning a computer as each means of the radiography apparatus of any one of Claims 1 thru | or 15.   The program for functioning a computer as each means of the radiographic image processing apparatus of Claim 16 or 17.   A computer-readable storage medium storing the program according to claim 22 or 23.

Images