JP2014236480A - Radiation imaging system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a radiation imaging system that is able to obtain image information about satisfactory image quality by preventing noise resulting from a reset operation.SOLUTION: A radiation imaging system comprises: a detecting section (206) including a plurality of pixels arranged in a matrix and configured to output a pixel output value by converting radiation into electric charges and output a pixel output value; detecting means (203) configured to detect the start of emission of radiation; control means (202) configured to output image information, corresponding to the emission of radiation, to the plurality of pixels, thereby outputting dark image information; first correcting means (210) configured to make a correction by subtracting dark image information from image information corresponding to the emission of radiation; and second correcting means (211) configured to correct, in relation to image information corrected by the first correcting means, a step between an odd line and an even line, resulting from the reset operation of an interlace, and a step resulting before and after a line in which the reset operation of the interlace is stopped.

Description

  The present invention relates to a radiation imaging system suitably used for still image shooting such as general shooting in medical diagnosis and moving image shooting such as fluoroscopic shooting.

  At present, radiation imaging apparatuses using a flat panel detector (hereinafter referred to as FPD) made of a semiconductor material are widely used as imaging apparatuses used for medical image diagnosis and nondestructive inspection using X-rays. Such a radiation imaging apparatus is used as a digital imaging apparatus for still image shooting such as general shooting or moving image shooting such as fluoroscopic shooting in medical image diagnosis, for example. Such an imaging apparatus generally has a configuration in which the X-ray generator and the FPD are synchronized. However, when the FPD is installed, connection with the X-ray generator is necessary, and there is a problem that the installation location is limited.

  On the other hand, Patent Document 1 discloses a technique for detecting the start of radiation irradiation by a current flowing through a bias line of a conversion element without constructing an interface with a radiation generator. Specifically, the reset operation is performed by sequentially scanning non-adjacent rows. When radiation irradiation is detected, the reset operation is stopped and the operation proceeds to the accumulation operation. When radiation irradiation is completed, scanning is sequentially performed and an image data reading operation is performed. Patent Document 1 discloses a technique for reading offset correction data at the same timing from the reset operation to the read operation in a state where no radiation is irradiated after the image data read operation.

  In Patent Document 2, in order to correct the dark current response of the conversion element, the time from bias application to radiation imaging, the storage time, and the storage time of offset correction data are calculated from the response characteristics of the dark current. Based on the accumulation time, the offset correction data is read and corrected. Further, the scanning time of each row when obtaining the offset signal after X-ray imaging is calculated by calculation, and each row is scanned based on the calculated scanning time. In addition, in order to correct the dark current response of the conversion element, the time from bias application to radiography, the accumulation time, the response characteristics of the dark current, and the offset correction data acquired in advance are superimposed on the image data at the time of radiography. The dark current component to be calculated is calculated and corrected.

JP2011-249891A JP 2008-259045 A

  In the case where the dark current component of the conversion element fluctuates with time, the technique disclosed in Japanese Patent Application Laid-Open No. 2004-228561 may not be able to correct the step-like artifacts generated before and after the line in which the radiation is detected and the image quality may be deteriorated. They found out. In particular, when detecting X-rays while repeating a reset operation (empty reading) in which non-adjacent rows called interlaces, for example, even rows and odd rows are scanned separately, stripes of even odd rows generated in the row direction In some cases, the image quality may be deteriorated due to failure to correct the artifacts.

  On the other hand, in Patent Document 2, during the reset operation (empty reading), the scanning is stopped when radiation is detected, and the concept itself entering the accumulation operation or the reset operation is performed by sequentially scanning non-adjacent rows. Since there is no concept to perform, sufficient correction cannot be made. Therefore, in Patent Document 2, similar to Patent Document 1, artifacts that occur before and after the detection row and striped artifacts that occur in the even-numbered rows in the row direction may deteriorate the image quality. In addition, controlling the driving unit based on the scanning time of each row calculated by calculation requires a complicated circuit configuration, and may be difficult to realize.

  An object of the present invention is to provide a radiation imaging system that can prevent noise caused by a reset operation and obtain image information with good image quality.

  The radiation imaging system of the present invention includes a detection unit that is arranged in a matrix and includes a plurality of pixels that convert radiation into charges and output a pixel output value, a detection unit that detects the start of radiation irradiation, and the detection unit Until the start of radiation irradiation is detected, the interlace reset operation is performed on the plurality of pixels, and when the start of radiation irradiation is detected by the detection unit, the plurality of pixels are detected. The interlace reset operation is stopped and the charge accumulation operation is performed. After that, when the radiation irradiation is completed, the pixel output value reading operation of the plurality of pixels is performed, thereby responding to the radiation irradiation. The image information is output, and then the interlace reset operation, the charge accumulation operation, and the pixel output value read operation are performed again on the plurality of pixels. Control means for outputting dark image information, first correction means for performing correction by subtracting the dark image information from image information corresponding to the radiation irradiation, and the first correction means. The step of correcting the step between the odd-numbered row and the even-numbered row due to the interlace reset operation and the step generated before and after the row where the interlace reset operation is stopped is corrected for the image information corrected by 2 correction means.

  By the correction of the second correction unit, noise caused by the reset operation can be prevented, and image information with good image quality can be obtained.

It is a block diagram which shows the structural example of a radiation imaging system. It is a circuit diagram which shows the structural example of a planar detector. It is a flowchart which shows operation | movement of a radiation imaging system. It is a drive timing diagram of a radiation imaging system. It is a drive timing diagram of a radiation imaging system. It is a figure for demonstrating a charge accumulation period. It is a figure for demonstrating 1st offset correction. It is a figure for demonstrating 2nd offset correction. It is a figure for demonstrating 2nd offset correction. It is explanatory drawing of a dummy pixel. It is a figure for demonstrating artifact correction.

  FIG. 1 is a block diagram illustrating a configuration example of a radiation imaging system according to an embodiment of the present invention. In addition to α rays, β rays, γ rays, etc., which are beams produced by particles (including photons) emitted by radiation decay, radiation has a similar level of energy, such as X-rays, particle rays, and space. Lines are also included. Hereinafter, an example in which the radiation is X-ray will be described.

  The radiation imaging system includes an X-ray generation apparatus 201, a control unit 202, an X-ray detection unit 203, a driving unit 204, a reading unit 205, a planar detector 206, a first memory 207, a second memory 208, and a correction unit 209. And display means (or a computer) 214. The correction unit 209 includes a first offset correction unit (first correction unit) 210, a second offset correction unit (second correction unit) 211, a gain correction unit 212, and an artifact correction unit 213. The control unit 202 controls the X-ray detection unit 203, the driving unit 204, and the reading unit 205. The driving means 204 drives the flat detector 206.

  The X-ray generator (radiation generator) 201 irradiates the flat detector 206 and the X-ray detector 203 with X-rays (radiation) through the subject. The flat detector 206 has a plurality of pixels arranged in a matrix of X columns × Y rows, and each pixel converts X-rays that have passed through the subject into charges and accumulates the accumulated charges. Output as output value. The X-ray detection unit 203 detects the start and end of X-ray irradiation and outputs an X-ray irradiation signal to the control unit 202. When the X-ray detector 203 detects the start of X-ray irradiation, the flat detector 206 starts accumulating charges according to the X-ray, and when the X-ray detector 203 detects the end of X-ray irradiation, A pixel output value corresponding to the accumulated charge is output to the reading unit 205. Then, the reading unit 205 outputs image information X-image (X, Y) corresponding to the X-ray irradiation of the pixels in X columns × Y rows. The first memory 207 stores image information X-image (X, Y).

  X-ray detection means 203 outputs periods TX and TE. The period TX is a period from the bias voltage application start time to the X-ray irradiation start time, as shown in FIG. The period TE is a period from the bias voltage application start time to the X-ray irradiation end time, as shown in FIG. As shown in FIG. 4, the driving unit 204 outputs the row RX in which the idle reading is stopped at the start of X-ray irradiation to the correcting unit 209.

  Thereafter, in a state where the X-ray generation apparatus 201 does not irradiate X-rays, the flat detector 206 performs the same operation (monomony drive) again based on the above-described periods TX and TE and the idle reading stop line RX. . Then, the reading unit 205 outputs dark image information Dark_A (X, Y) of pixels in X columns × Y rows. The second memory 208 stores dark image information Dark_A (X, Y). Here, the dark image information Dark_A (X, Y) is information regarding an offset component including fixed pattern noise for each pixel.

  The first offset correction unit 210 obtains dark image information Dark_A (X, Y) stored in the second memory 208 from image information X-image (X, Y) stored in the first memory 207. The first offset correction is performed by subtracting. The second offset correction unit 211 performs second offset correction on the image information corrected by the first offset correction unit 210 by correcting the offset difference between the odd and even rows. The gain correction unit 212 performs gain correction on the image information corrected by the second offset correction unit 211. The artifact correction unit 213 corrects the artifact caused by the detection delay of the X-ray detection unit 203 with respect to the image information corrected by the gain correction unit 212. The display unit (or computer) 214 displays (or processes) the image information corrected by the artifact correction unit 213.

  FIG. 2 is a circuit diagram showing a configuration example of the reading unit 205 and the flat detector 206 of FIG. The flat detector 206 includes a vertical drive circuit 114, a detection unit 112, and a bias power supply unit 103. The reading unit 205 includes a reading circuit 113, an output buffer amplifier 109, and an analog / digital (A / D) converter 110.

  The flat detector 206 is a sensor in which elements (pixels) for detecting radiation are arranged in a two-dimensional matrix, and detects radiation and outputs image information. FIG. 2 shows an example of the detection unit 112 having pixels of 3 rows × 3 columns for ease of explanation. However, the actual planar detector 206 has a larger number of pixels. For example, in the case of 17 inches, it has about 2800 rows × about 2800 columns of pixels.

  The detection unit 112 has a plurality of pixels arranged in a matrix. Each pixel has conversion elements S11 to S33 that convert radiation or light into electric charges, and switch elements T11 to T33 that output electric signals according to the electric charges of the conversion elements S11 to S33, and output pixel output values. . The conversion elements S11 to S33 are indirect type conversion elements or direct type conversion elements, and convert irradiated radiation into electric charges. The indirect conversion elements S11 to S33 include a wavelength converter that converts radiation into light and a photoelectric conversion element that converts the light into electric charges. The direct conversion elements S11 to S33 convert radiation directly into electric charges. As a photoelectric conversion element that converts irradiated light into electric charges, an MIS type photodiode that is disposed on an insulating substrate such as a glass substrate and mainly contains amorphous silicon can be used. The photoelectric conversion element may be a PIN photodiode.

  The switch elements T11 to T33 are transistors having a control terminal and two main terminals, and are preferably thin film transistors (TFTs). In each of the conversion elements S11 to S33, one electrode is electrically connected to one of the two main terminals of the switch elements T11 to T33, and the other electrode is electrically connected to the bias power supply unit 103 via a common bias line Bs. Connected. The plurality of switch elements T11 to T13 in the first row have their control terminals electrically connected in common to the drive line R1 in the first row. The plurality of switch elements T21 to T23 in the second row have their control terminals electrically connected in common to the drive line R2 in the second row. The control terminals of the plurality of switch elements T31 to T33 in the third row are electrically connected in common to the drive line R3 in the third row. The vertical drive circuit 114 is a shift register, for example, and supplies drive signals to the switch elements T11 to T33 via the drive lines R1 to R3, thereby controlling the conduction states of the switch elements T11 to T33 in units of rows.

  Each of the plurality of switching elements T11 to T31 in the first column has one main terminal connected to the conversion elements S11 to S31 and the other main terminal electrically connected to the signal line Sig1 in the first column. While the switch elements T11 to T31 in the first column are in the conductive state, an electrical signal corresponding to the charges of the conversion elements S11 to S31 in the first column is output to the readout circuit 113 via the signal line Sig1. Each of the plurality of switching elements T12 to T32 in the second column has one main terminal connected to the conversion elements S12 to S32, and the other main terminal electrically connected to the signal line Sig2 in the second column. While the switch elements T12 to T32 in the second column are in the conductive state, an electrical signal corresponding to the charges of the conversion elements S12 to S32 in the second column is output to the readout circuit 113 via the signal line Sig2. Each of the plurality of switching elements T13 to T33 in the third column has one main terminal connected to the conversion elements S13 to S33 and the other main terminal electrically connected to the signal line Sig3 in the third column. While the switch elements T13 to T33 in the third column are in the conductive state, an electrical signal corresponding to the charges of the conversion elements S13 to S33 in the third column is output to the readout circuit 113 via the signal line Sig3. A plurality of signal lines Sig <b> 1 to Sig <b> 3 arranged in the column direction output electric signals output from a plurality of pixels to the readout circuit 113 in parallel.

  The readout circuit 113 is provided with an amplifier circuit 106 for amplifying the electric signals of the signal lines Sig1 to Sig3 for each of the signal lines Sig1 to Sig3. Each amplification circuit 106 includes an integration amplifier 105, a variable gain amplifier 104, and a sample hold circuit 107. The integrating amplifier 105 amplifies the electric signals of the signal lines Sig1 to Sig3. The variable gain amplifier 104 amplifies the electric signal from the integration amplifier 105 with a variable gain. The sample hold circuit 107 samples and holds the electric signal amplified by the variable gain amplifier 104. The integrating amplifier 105 includes an operational amplifier 121 that amplifies and outputs electric signals of the signal lines Sig1 to Sig3, an integrating capacitor 122, and a reset switch 123. The integrating amplifier 105 can change the gain (amplification factor) by changing the value of the integrating capacitor 122. The operational amplifier 121 of each column has an inverting input terminal connected to the signal lines Sig1 to Sig3, a normal rotation input terminal connected to the reference power supply unit 111 of the reference voltage Vref, and an output terminal that outputs an amplified electric signal. To do. The reference power supply unit 111 supplies the reference voltage Vref to the normal rotation input terminal of each operational amplifier 121. The integration capacitor 122 is disposed between the inverting input terminal and the output terminal of the operational amplifier 121. The sample hold circuit 107 has a sampling switch 124 for the control signal SH and a sampling capacitor 125. Further, the reading circuit 113 includes a switch 126 and a multiplexer 108 in each column. The multiplexer 108 sequentially turns on the switches 126 in each column, and sequentially outputs the electrical signals output in parallel from the amplifier circuits 106 to the output buffer amplifier 109 as serial signals. The output buffer amplifier 109 impedance-converts the electrical signal and outputs it. The analog / digital (A / D) converter 110 converts the analog electric signal output from the output buffer amplifier 109 into a digital electric signal, and stores it in the first memory 207 or the second memory 208 of FIG. 1 as image information. Output.

  The bias power supply unit 103 includes a current-voltage conversion circuit 115 and an A / D converter 127. The current-voltage conversion circuit 115 converts the current flowing through the bias line Bs into a voltage while supplying the bias voltage Vs to the bias line Bs, and outputs the voltage to the A / D converter 127. The A / D converter 127 converts an analog voltage value having current information into a digital voltage value having current information and outputs the digital voltage value. The X-ray detection unit 203 in FIG. 1 detects the start and end of X-ray irradiation using the current information output from the A / D converter 127.

  In accordance with control signals D-CLK, OE, and DIO input from the drive unit 204 of FIG. 1, the vertical drive circuit 114 is turned on and turned off by the switch elements T11 to T33. Is output to each of the drive lines R1 to R3. Thereby, the vertical drive circuit 114 controls the conduction state and the non-conduction state of the switch elements T11 to T33, and drives the detection unit 112. The control signal D-CLK is a shift clock for a shift register used as the vertical drive circuit 114. The control signal DIO is a transfer pulse of the shift register of the vertical drive circuit 114. The control signal OE is an output enable signal for the shift register of the vertical drive circuit 114. As described above, the vertical drive circuit 114 sets the drive time and the scanning direction. The driving unit 204 controls the operation of each component of the reading circuit 113 by outputting the control signal RC, the control signal SH, and the control signal CLK to the reading circuit 113. The control signal RC is a signal for controlling the operation of the reset switch 123 of the integrating amplifier 105. The control signal SH is a signal for controlling the sampling switch 124 of the sample and hold circuit 107. The control signal CLK is a clock signal for controlling the operation of the multiplexer 108.

  3 is a flowchart showing a control method of the radiation imaging system of FIG. 1, FIG. 5 is a timing chart of the control method, and FIG. 4 is a timing chart enlarging a part of FIG. In step S301, the current-voltage conversion circuit 115 starts applying the bias voltage Vs to the bias line Bs as shown in FIG.

  Next, in step S302, the control means 202 determines whether or not X-ray irradiation has been started. The X-ray detection unit 203 outputs an X-ray irradiation signal to the control unit 202 when the current information output from the A / D converter 127 (electrical signal corresponding to the irradiated X-ray) is equal to or greater than a threshold value. . The control unit 202 determines that X-ray irradiation has been started when an X-ray irradiation signal is input, and determines that X-ray irradiation has not been started when no X-ray irradiation signal is input. If X-ray irradiation has been started, the process proceeds to step S304. If X-ray irradiation has not been started, the process proceeds to step S303. In step S303, under the control of the control unit 202, the detection unit 112 sets the drive lines R1 to R14 to the conductive voltage and sets the switch elements T11 to T33 and the like to the conductive state as illustrated in FIG. As a result, a reset operation (hereinafter referred to as idle reading) for resetting the charges of the conversion elements S11 to S33 and the like caused by the dark current charge accumulation is performed. Thereafter, the process returns to step S302. The detection unit 112 repeatedly performs a reset operation for resetting charges of the conversion elements S11 to S33 and the like generated by dark current before X-ray irradiation.

  In idle reading in step S303, as shown in FIG. 4, an interlace reset operation is performed. First, the drive lines R2, R4, R6,..., R14 of the pixels in the even rows are sequentially turned on, the switch elements T21, T41, etc. of the pixels in the even rows are sequentially turned on, and the pixels of the even rows are turned on. The charges of the conversion elements S21, S41, etc. are reset. Next, the drive lines R1, R3, R5,..., R13 of the odd-numbered pixels are sequentially turned on, the switch elements T11, T31 and the like of the odd-numbered pixels are sequentially turned on, and the odd-numbered pixels The charges of the conversion elements S11, S31, etc. are reset. The operation of the combination of the even row reset and the odd row reset is repeated until the start of X-rays is detected.

  After detecting the start of X-ray irradiation, in step S304, the control unit 202 stores the X-ray detection row RX via the driving unit 204. The X-ray detection row RX indicates a row RX in which idle reading is stopped by the start of X-ray irradiation, and is the fourth row corresponding to the drive line R4 in the case of FIG.

  Next, in step S <b> 305, the control unit 202 stores the period TX via the X-ray detection unit 203. The period TX is a period from the bias voltage application start time to the X-ray irradiation start time, as shown in FIG.

  Next, in step S306, the X-ray detection unit 203 determines whether or not the X-ray irradiation has ended. The X-ray detection means 203 determines the end of X-ray irradiation when the current information output from the A / D converter 127 (electrical signal corresponding to the irradiated X-ray) is less than the threshold value, and X-ray irradiation Stop signal output. Further, the X-ray detection means 203 may stop outputting the X-ray irradiation signal after a predetermined time (X-ray irradiation period) has elapsed from the detection time of the start of X-ray irradiation. The control unit 202 determines that the X-ray irradiation has ended when the input of the X-ray irradiation signal is stopped, and the X-ray irradiation has not ended when the X-ray irradiation signal is input. Judge. If the X-ray irradiation has been completed, the process proceeds to step S308. If the X-ray irradiation has not been completed, the process proceeds to step S307. In step S <b> 307, the detection unit 112 performs a charge accumulation operation under the control of the control unit 202. Thereafter, the process returns to step S306. As shown in FIG. 4, in the charge accumulation operation, all the drive lines R1 to R14 are set to a non-conductive voltage, and the switch elements T11 to T33 and the like of all the pixels are set to a non-conductive state in accordance with the X-ray irradiation. This is an operation of accumulating charges in the conversion elements S11 to S33. Until the X-ray irradiation is completed, the charge accumulation operation is performed in the conversion elements S11 to S33 and the like.

  After detecting the end of X-ray irradiation, in step S308, the control unit 202 stores the period TE via the X-ray detection unit 203. As shown in FIG. 4, the time TE is the time from the bias voltage application start time to the X-ray irradiation end time.

  Next, in step S <b> 309, the detection unit 112 performs a main read operation 502 for reading out charges corresponding to the X-ray irradiation under the control of the control unit 202. In the read operation 502, the drive lines R1 to R14 sequentially become conduction voltage pulses, the switch elements S11 to S33 and the like are sequentially turned on in units of rows, and from the first row of pixels to the last row of pixels. Then, electrical signals are sequentially output to the signal lines Sig1 to Sig3. The A / D converter 110 outputs image information X-image (X, Y) of pixels from the first row to the last row. The first memory 207 stores image information X-image (X, Y).

  Next, in step S310, the driving unit 204 performs the period 501 in FIGS. 4 and 5 based on the periods TX and TE and the idle reading stop line RX in a state where the X-ray generator 201 does not emit X-rays. The same control as that of the drive lines R1 to R14 is performed in the period 503. In other words, the monetary drive is performed.

  Next, in step S <b> 311, under the control of the control unit 202, the detection unit 112 performs a main read operation 504 for reading out charges from the conversion elements S <b> 11 to S <b> 33 and the like, similar to the main read operation 502. In the read operation 504, the drive lines R1 to R14 sequentially become conduction voltage pulses, the switch elements S11 to S33 and the like are sequentially turned on in units of rows, and from the first row of pixels to the last row of pixels. Then, electrical signals are sequentially output to the signal lines Sig1 to Sig3. The A / D converter 110 outputs dark image information Dark_A (X, Y) of pixels from the first row to the last row. The second memory 208 stores dark image information Dark_A (X, Y).

  FIG. 6 is a diagram illustrating a charge accumulation period of each drive line R1 to R14 in FIG. The charge accumulation period is a period from the last pulse falling time of each driving line R1 to R14 in the period 501 to the pulse rising time of each driving line R1 to R14 in the read operation 502. Since the odd row idle reading is performed after the even row idle reading, the charge accumulation period of the odd row is different from the charge accumulation period of the even row. Further, the even-numbered rows R2 and R4 below the idle reading stop row RX (= 4) have a charge accumulation period compared to the even-numbered rows R6, R8,..., R14 larger than the idle reading stop row RX (= 4). short. Further, since idle reading is sequentially performed in units of rows, the charge accumulation period becomes longer as the row number increases.

  FIG. 7A is a diagram illustrating output values for each row of the image information X-image (X, Y) and the dark image information Dark_A (X, Y). An output 701 indicates an output value of an odd-numbered row in the image information X-image (X, Y). An output 702 indicates an output value of an even-numbered row of the image information X-image (X, Y). An output 703 indicates dark current components of odd-numbered rows in the image information X-image (X, Y). An output 704 indicates the dark current components of even-numbered rows in the image information X-image (X, Y). An output 705 indicates an output value of an odd-numbered row in the dark image information Dark_A (X, Y). An output 706 indicates an output value of an even-numbered row in the dark image information Dark_A (X, Y).

  In the detection unit 112, dark current is generated even during a period in which X-rays are not irradiated. Therefore, in FIG. 6, the longer the charge accumulation period, the larger the noise of the dark current component, and the dark current noise in the image information X-image (X, Y) and the dark image information Dark_A (X, Y) increases. . Therefore, the outputs 701 to 706 in FIG. 7A correspond to the length of the charge accumulation period in FIG. 6 and dark current noise occurs.

  The period 711 corresponds to the period 511 in FIG. 5, and, similar to FIG. 6, the charge accumulation period is in the order of the sixth line of the drive line R 6, the fifth line of the drive line R 5, and the fourth line of the drive line R 4. It gets shorter. The subsequent period 712 corresponds to the period 512 in FIG. 5 and, similarly to FIG. 6, charge accumulation is performed in the order of the sixth line of the drive line R6, the fifth line of the drive line R5, and the fourth line of the drive line R4. The period is getting shorter. The dark current 713 becomes smaller as the time from the start of application of the bias voltage Vs elapses. Therefore, the dark current in the period 711 is larger than the dark current in the period 712. As a result, the dark current components 703 and 704 of the image information X-image (X, Y) in the period 711 are larger than the dark image information Dark_A (X, Y) in the period 712.

  Next, in step S <b> 312 of FIG. 3, the first offset correction unit 210 is stored in the second memory 208 from the image information X-image (X, Y) stored in the first memory 207. The dark image information Dark_A (X, Y) is subtracted. Thereby, the first offset correction is performed.

  FIG. 7B is a diagram of output values of each row showing the result of the first offset correction. The output 721 indicates the result of odd-numbered rows obtained by subtracting the dark image information Dark_A (X, Y) from the image information X-image (X, Y). The output 722 shows the result of even-numbered rows obtained by subtracting the dark image information Dark_A (X, Y) from the image information X-image (X, Y). If the dark current 713 in FIG. 7A is constant regardless of the time from the start of application of the bias voltage Vs, the output value of each row is the same in the outputs 721 and 722 in FIG. 7B. However, since the dark current 713 decreases with the passage of time from the start of application of the bias voltage Vs, the first offset correction result in FIG. 7B still includes dark current noise. In order to remove the dark current noise, the following second offset correction is performed.

  Next, in step S313 in FIG. 3, the second offset correction unit 211 performs the second offset correction. The second offset correction unit 211 performs a step between the odd and even rows due to the interlace reset operation and the interlace reset for the image information corrected by the first offset correction unit 210. The level difference generated before and after the row RX where the operation is stopped is corrected. 8 to 10 are examples different from each other for realizing step S313 in FIG. Each will be described in detail below.

  8A and 8B are diagrams for explaining the first method of the second offset correction. The outputs 721 and 722 in FIG. 8A correspond to the outputs 721 and 722 in FIG. The even-numbered output 722 is smaller than the odd-numbered output 721 in the even-numbered lines below the empty reading stop line RX, and is larger than the odd-numbered output 721 in the even-numbered lines larger than the empty reading stopped line RX. As a result, striped artifact noise as shown in FIG. 8B occurs in the output image information of the first offset correction unit 210. The output difference between the odd and even lines may not be constant from the first line to the last line. In the even-numbered output 722, there is a discontinuous point in the idle reading stop line RX, and in the odd-numbered output 721, there is no discontinuous point.

  Here, first, the second offset correction means 211 outputs an output difference a between the output 721 of the first odd row and the output 722 of the first even row, and the output 721 of the final odd row and the output of the last even row. The output difference b from 722 is measured. Next, the output difference A between the output 721 of the odd-numbered row immediately before the idle reading stop row RX and the output 721 of the even-numbered row immediately before the idle reading stopped row RX, and the output 721 of the odd row immediately after the idle reading stopped row RX. And the output difference B from the output 722 of the even-numbered row immediately after the idle reading stop row RX is measured. It should be noted here that the odd-numbered output 721 does not have a discontinuity near the stop row RX. Accordingly, the second offset correction unit 211 performs arithmetic processing using the measured output differences a, A, b, and B so that the odd-numbered row output 722 matches the even-numbered row output 721. Here, since the amount of step changes continuously between the output difference a and the output difference A and between the output difference b and the output difference B, linear correction or the like may be performed. In order to obtain the output differences a, A, b, and B with high accuracy, it is desirable to use a row average value or an average value of a plurality of pixels. By performing the second offset correction by the method shown in FIG. 8, it is possible to reduce steps before and after the idle reading stop line and even / odd steps with a simple configuration.

Next, a correction process different from that in FIGS. 8A and 8B performed by the second offset correction unit 211 will be described with reference to FIGS.
FIG. 9A is a diagram illustrating a configuration example of the detection unit 112 in FIG. 8A and 8B is that the detection unit 112 has a normal pixel region 901, a dummy pixel region 902 at the upper end, and a dummy pixel region 903 at the lower end. The dummy pixel regions 902 and 903 do not convert X-rays into charges even when they are irradiated with X-rays. The dummy pixel regions 902 and 903 are, for example, light-shielding pixel regions, and light corresponding to the X-ray is shielded, so that no charge corresponding to the X-ray is generated. In contrast, the normal pixel region 901 is not shielded from light, and charges corresponding to X-rays are generated. A dummy pixel region 902 at the upper end includes pixels in the first odd-numbered row and even-numbered row. The dummy pixel region 903 at the lower end includes pixels in the last odd-numbered row and even-numbered row.

  FIG. 9B is a diagram for explaining a correction method of the second offset correction unit 211. Similar to FIGS. 7B and 8A, the output 721 is an odd number obtained by subtracting the dark image information Dark_A (X, Y) from the image information X-image (X, Y) when there is no dummy pixel region. Indicates the line result. The output 722 shows the result of even-numbered rows obtained by subtracting the dark image information Dark_A (X, Y) from the image information X-image (X, Y) when there is no dummy pixel region. On the other hand, the output 911 is an odd-numbered output when it is assumed that the entire pixel area is a light-shielded dummy pixel area, and the output 912 is an even-number when the entire pixel area is assumed to be a light-shielded dummy pixel area. This is the output of the line. It should be noted here that the step component due to the dark current is irrelevant to the presence or absence of light shielding. That is, the difference between the output 722 and the output 721 and the difference between the output 912 and the output 911 are equal. Therefore, by shielding a part of the pixel area as shown in FIG. 9A, it is possible to accurately obtain the step amount at the upper end and the step amount at the lower end. First, the second offset correction unit 211 determines the difference a between the output 911 of the odd-numbered row and the output 912 of the even-numbered row 912 in the upper dummy pixel region 902 from the output 911 and the output 912 that are the results of the first offset correction. Is calculated. Next, the difference b between the odd-numbered row output 911 and the even-numbered row output 912 of the dummy pixel region 903 at the lower end is calculated. Next, the second offset correction unit 211 calculates a difference A between the output 722 of the even-numbered row and the output 721 of the odd-numbered row for each even-numbered row based on the empty reading stop row RX. Then, the second offset correction unit 211 calculates the difference B between the output 722 of the even-numbered row and the output 721 of the odd-numbered row based on the idle reading stop row RX. Next, the second offset correction unit 211 performs correction in the same manner as described above using the differences a, A, b, and B. Thereby, the second offset correction is performed. By this second offset correction, the noise of the striped artifact can be prevented. There are cases where the upper and lower edges of the image are in a so-called unclear state in which X-rays that do not transmit through the subject directly enter the pixel region. In this case, the output is saturated and there is a possibility that an even / odd step component cannot be obtained correctly. Subsequently, as shown in FIGS. 9A and 9B, by providing the dummy pixel region, the step component can be accurately corrected without being adversely affected by the missing elements.

  FIG. 10A corresponds to FIG. 9A and shows examples of other dummy pixel regions 1002 to 1005. The detection unit 112 includes a normal pixel region 1001, a dummy pixel region 1002 at the upper left corner, a dummy pixel region 1003 at the upper right corner, a dummy pixel region 1004 at the lower left corner, and a dummy pixel region 1005 at the lower right corner. Have The dummy pixel regions 1002 to 1005 do not convert X-rays into electric charges even when they are irradiated with X-rays, similarly to the dummy pixel regions 902 and 903 in FIG. In the normal pixel region 1001, charges corresponding to X-rays are generated in the same manner as the normal pixel region 901 in FIG. The dummy pixel areas 1002 and 1003 at the upper end include pixels in the first odd-numbered row and even-numbered row. The dummy pixel regions 1004 and 1005 at the lower end include the pixels of the last odd and even rows. Also in this case, the same processing as described above can be performed.

  FIG. 10B corresponds to FIG. 9A and shows an example of another dummy pixel region 1012. The detection unit 112 includes a normal pixel region 1011 and a dummy pixel region 1012 at the right end. Similarly to the dummy pixel regions 902 and 903 in FIG. 9A, the dummy pixel region 1012 does not convert X-rays into electric charges even when irradiated with X-rays. In the normal pixel area 1011, charges corresponding to X-rays are generated in the same way as the normal pixel area 901 in FIG. The dummy pixel region 1012 has pixels in all rows. In this case, with respect to all the rows, the output difference between the odd rows and the even rows can be detected and the offset correction can be performed. In this case, the information on the idle reading stop line RX is not necessarily required.

  Next, in step S <b> 314 of FIG. 3, the gain correction unit 212 performs gain correction on the image information corrected by the second offset correction unit 211. In FIG. 8A, after performing the second offset correction in step S313, the odd-numbered row output 721 and the even-numbered row output 722 indicate straight lines having a positive slope gain error. Therefore, the gain correction unit 212 reduces the gain error by correcting the gain variation.

  Here, in FIG. 9B, when the average value of the difference between the odd-numbered row output 721 and the even-numbered row output 722 is ½ or less of the random noise, the second offset in step S313 in FIG. It is not necessary to perform correction. Further, in FIG. 4, when the period from the bias voltage application start time to the X-ray irradiation start detection time is longer than the predetermined period, the time change of the dark current is small, so that the corrections in steps S313 and S314 in FIG. 3 are performed. It does not have to be.

  Next, in step S315, the artifact correction unit 213 corrects the artifact. Specifically, the artifact correction unit 213 detects the image information corrected by the gain correction unit 212 until the interlace reset operation is stopped after the X-ray detection unit 203 detects the start of X-ray irradiation. The artifacts caused by the delay time are corrected. As shown in FIG. 11, a case will be described in which the X-ray generator 201 starts X-ray irradiation after idle reading of the row RA. In that case, when the X-ray detection unit 203 detects the start of X-ray irradiation, due to the detection delay of the X-ray detection unit 203, the detection unit 112 performs idle reading of the even rows from the row RA to the row RX, and then The reading operation is stopped and the operation proceeds to the charge accumulation operation. Ideally, when the X-ray generator 201 starts X-ray irradiation after idle reading of the row RA, it is desirable to stop the empty reading operation and perform charge accumulation operation after empty reading of the row RA. . However, since a predetermined time is required for the detection and control of the X-ray detection means 203, the even-numbered rows from the row RA + 2 to the row RX are read out after the X-rays are irradiated, and thus are accumulated by the X-ray irradiation. The charges of the conversion element S41 and the like are reset. As a result, the even numbered rows from the row RA + 2 to the row RX have a smaller amount of charge of the conversion element S41 and the like accumulated by X-ray irradiation than the other rows. As a result, the even-numbered rows from the row RA + 2 to the row RX have a smaller output value than the other rows, and a wedge-shaped artifact occurs. Therefore, in step S315, the artifact correction unit 213 corrects the output values of the pixels in the even rows from the row RA + 2 to the row RX, thereby reducing the noise of the wedge-shaped artifact.

  In addition, since the wedge-shaped artifact exists in the even-numbered rows from the row RA + 2 to the row RX, the correction in steps S313 and S314 in FIG. 3 may not be performed. In that case, the artifacts in even rows from row RA + 2 to row RX can be reduced by the correction in step S315.

  The above-described embodiments are merely examples of implementation in carrying out the present invention, and the technical scope of the present invention should not be construed in a limited manner. That is, the present invention can be implemented in various forms without departing from the technical idea or the main features thereof.

202 control means, 203 X-ray detection means, 206 planar detector, 210 first offset correction means, 211 second offset correction means

Claims (6)

A detector that is arranged in a matrix and includes a plurality of pixels that convert radiation into electric charges and output pixel output values;
Detecting means for detecting the start of radiation irradiation;
Until the detection means detects the start of radiation irradiation, an interlace reset operation is performed on the plurality of pixels. When the detection means detects the start of radiation irradiation, the plurality of pixels Then, the interlace reset operation is stopped and the charge accumulation operation is performed.After the radiation irradiation is completed, the pixel output values of the plurality of pixels are read out, thereby performing the radiation output operation. Dark image information is output by outputting image information corresponding to irradiation, and then performing the interlace reset operation, the charge accumulation operation, and the pixel output value read operation again for the plurality of pixels. Control means for outputting
First correction means for performing correction by subtracting the dark image information from image information corresponding to the radiation irradiation;
For the image information corrected by the first correction unit, before and after the step between the odd-numbered row and the even-numbered row due to the interlace reset operation and the row where the interlace reset operation is stopped. A radiation imaging system, comprising: a second correction unit that corrects a generated level difference. The detection unit includes a plurality of dummy pixels that do not convert radiation into charges,
The radiation imaging system according to claim 1, wherein the second correction unit performs the correction using pixel output values of the plurality of dummy pixels.   The radiation imaging system according to claim 2, wherein the plurality of dummy pixels include at least a first odd-numbered row and even-numbered row and a last odd-numbered row and even-numbered row pixel.   The radiation imaging system according to claim 1, further comprising gain correction means for performing gain correction on the image information corrected by the second correction means.   Further, for the image information corrected by the second correction unit, an artifact caused by a delay time from when the detection unit detects the start of radiation irradiation to when the interlace reset operation is stopped is corrected. The radiation imaging system according to any one of claims 1 to 4, further comprising an artifact correcting unit.   Furthermore, it has a radiation generator which irradiates a radiation, The radiation imaging system of any one of Claims 1-5 characterized by the above-mentioned.

Images