JP2017224991A - Radiation imaging apparatus, its driving method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique advantageous to improve accuracy in offset correction and improve radiation image quality accompanying thereof.SOLUTION: A radiation imaging apparatus comprises: an imaging section formed from a plurality of pixels arranged in rows and columns; a plurality of column signal lines corresponding to the plurality of columns and respectively connected to the pixels of the corresponding columns; a plurality of current sources corresponding to the plurality of columns, and each of which supplies bias current to the corresponding column signal line; and a processor that acquires image data on the basis of a pixel signal output via the plurality of column signal lines from the plurality of pixels of the imaging section after emission of radiation, and that corrects the image data on the basis of correction data. Using, as a plurality of dark image data, image data acquired from the imaging section during non-emission of radiation, the processor acquires the plurality of dark image data in order at time intervals different from time intervals for a first period, which is a period of change in the value of the bias current, resulting from 1/f noise in each current source. On the basis of them, the processor creates the correction data.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a radiation imaging apparatus, a driving method thereof, and a program.

  The radiation imaging apparatus, for example, forms image data based on an imaging unit in which a plurality of pixels are arranged to form a plurality of rows and a plurality of columns, and a pixel signal read from each pixel of the imaging unit. And a processor for performing correction processing on the image data. The correction processing includes offset correction for removing a component corresponding to offset noise from each pixel signal constituting the image data.

  In Patent Document 1, a plurality of offset data (dark image data) is acquired from the imaging unit in a state where the imaging unit is not irradiated with radiation, and they are averaged to generate correction data for offset correction. Is described. The processor corrects the image data obtained from the imaging unit irradiated with radiation using the correction data.

JP 2006-075359 A

  By the way, the radiation imaging apparatus can further include a plurality of column signal lines for reading out pixel signals from a plurality of pixels, and a plurality of current sources corresponding thereto. Each column signal line is connected to each pixel in the corresponding column, and each current source supplies a bias current to the corresponding column signal line.

  Here, while the current source supplies a bias current to the column signal line, 1 / f noise (flicker noise) may be generated in the current source. 1 / f noise is considered to be caused by charge trapping due to crystal defects (for example, a channel region lattice defect when a MOS transistor is used as a current source), and is generated periodically. Observed as fluctuations in the current value (fluctuations, fluctuations).

  Therefore, different amounts of noise components are mixed in the above-described offset data depending on the timing of acquiring the offset data. In addition, since a different amount of noise components may be mixed for each column, striped spots may occur in the offset data. This leads to a decrease in the accuracy of offset correction and may cause a decrease in the quality of the radiation image.

  An object of the present invention is to provide a technique advantageous in improving the accuracy of offset correction and improving the quality of a radiographic image accompanying therewith.

  One aspect of the present invention relates to a radiation imaging apparatus, and the radiation imaging apparatus corresponds to an imaging unit in which a plurality of pixels are arranged to form a plurality of rows and a plurality of columns, and the plurality of columns, respectively. And a plurality of column signal lines respectively connected to the respective pixels of the corresponding column, and a plurality of currents respectively corresponding to the plurality of columns and supplying a bias current to the column signal lines of the corresponding column. Image data is acquired based on pixel signals output from the plurality of pixels of the imaging unit via the plurality of column signal lines after radiation is applied to the source and the imaging unit, and the image data is A processor for correcting based on the correction data, wherein the processor uses the image data obtained from the imaging unit not irradiated with radiation as dark image data, and converts a plurality of dark image data to each current source. The correction data is sequentially acquired at a time interval different from the time interval of the first period, which is a period of fluctuation of the bias current value caused by 1 / f noise, and the correction data is generated based on the plurality of dark image data It is characterized by that.

  According to the present invention, the quality of a radiographic image can be improved.

It is a figure for demonstrating the example of a structure of a radiation imaging device. It is a figure for demonstrating the example of a structure of the single image sensor chip | tip, and its drive method. It is a figure for demonstrating the example of the circuit structure of a single pixel (sensor), and its drive method. It is a figure for demonstrating the 1 / f noise which generate | occur | produces in a current source. It is a flowchart for demonstrating the example of the drive method of a radiation imaging device.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings. In the drawings, the same elements are denoted by the same reference numerals, and the description of the overlapping contents will be omitted below. Each drawing is only described for the purpose of explaining the structure or configuration, and the dimensions of each element shown in the drawings do not necessarily reflect actual ones.

(Example of configuration and driving method of radiation imaging apparatus)
FIG. 1 is a block diagram illustrating an example of the overall configuration of the radiation imaging apparatus IA. The radiation imaging apparatus IA is a fluoroscopic diagnosis apparatus for observing the inside of a subject (subject) such as a patient from various angles by moving image shooting or continuous shooting, such as a C-arm X-ray fluoroscopy device. The radiation imaging apparatus IA includes an image acquisition unit 100, a control unit 101, a display unit 102, a radiation source control unit 103, and a radiation source 104. Although details will be described later, the image acquisition unit 100 acquires image data indicating an image in the body of the subject by radiography (or imaging), and outputs the image data to the control unit 101.

  The control unit 101 exchanges instructions using control signals or control commands with each unit (including those shown in the figure as well as those not shown) to perform synchronous control of the entire radiation imaging apparatus IA. , Functioning as a system controller in the device IA. The control unit 101 can also function as a processing unit that receives image data from the image acquisition unit 100 and performs processing such as image processing and correction processing. An input terminal (not shown) can be connected to the control unit 101, and the user can use the input terminal to acquire shooting information (for example, setting information necessary for shooting, such as an operation mode, a frame rate, and other parameters. Can be input to the control unit 101. The control unit 101 controls each unit synchronously based on the input photographing information. The control unit 101 may be a personal computer storing a program or software for realizing each operation described in this specification, or may be an arithmetic device including an integrated circuit (for example, ASIC, FPGA, etc.).

  The display unit 102 is a liquid crystal display, for example, and displays a radiographic image based on image data from the control unit 101. A user such as a doctor can diagnose the subject while referring to the radiographic image displayed on the display unit 102.

  The radiation source control unit 103 can be controlled by the control unit 101 so as to be synchronized with the image acquisition unit 100, and outputs a signal for performing radiation irradiation to the radiation source 104 in response to a control signal from the control unit 101. To do. The radiation source 104 generates radiation (typically X-rays) for performing radiography in response to a signal from the radiation source control unit 103. The generated radiation passes through a subject (not shown) and enters the image acquisition unit 100.

  The start of radiation imaging may be executed by the user via the input terminal, but may be executed using an exposure switch that can be provided in the radiation source control unit 103. For example, when an instruction (command) to start radiation imaging is input, the control unit 101 controls each unit synchronously to cause the radiation source 104 to generate radiation, and the image acquisition unit 100 detects the generated radiation. The image data is output from the image acquisition unit 100.

  The image acquisition unit 100 includes an imaging unit 10 for acquiring a radiation image, a reading unit 20 that reads a signal from the imaging unit 10, and a processor 109 that processes the read signal. The processor 109 drives and controls each unit in the image acquisition unit 100 while exchanging control signals, commands, data, and the like with the control unit 101.

  In this example, the imaging unit 10 includes a sensor array 105 in which a plurality of sensor chips 106 are arranged. For each sensor chip 106, for example, a CMOS image sensor chip manufactured by a known semiconductor manufacturing process using a semiconductor wafer such as a silicon wafer can be used. Each sensor chip 106 has a plurality of sensors (referred to as “sensors”) arranged in an array (to form a plurality of rows and a plurality of columns). With such a configuration, the sensor array 105 can be increased in size. Here, for ease of explanation, the configuration in which the sensor chips 106 are arranged so as to form 2 rows × 14 columns is illustrated, but the number of rows and the number of columns is not limited to these numbers. Absent.

  The sensor chips 106 adjacent to each other may be physically separated by dicing, but may not be separated. For example, each sensor chip 106 formed on the semiconductor wafer is inspected before dicing, and only the sensor chips 106 whose inspection results satisfy a predetermined standard may be arranged to form the sensor array 105. The sensor array 105 can be formed by arranging a plurality of sensor chips 106 on a plate-like member such as a base or a support plate. From this viewpoint, the sensor array 105 is also expressed as a sensor panel, a sensor substrate, or the like. Can be done.

  The imaging unit 10 may further include a scintillator (not shown) arranged on the radiation incident side with respect to the sensor array 105. The scintillator converts radiation into light. Each sensor s detects light from the scintillator and obtains an electric signal corresponding to the light by photoelectric conversion. Here, a case of a so-called indirect conversion type configuration in which radiation is converted into light by a scintillator and the light is photoelectrically converted is considered, but the present invention is a so-called direct conversion type that converts (directly) radiation into an electric signal. The present invention can also be applied to the case of the configuration.

  A plurality of electrodes (not shown) for signal input / output or supply of power supply voltage can be arranged in the imaging unit 10, and the imaging unit 10 is externally connected to the flying lead type printed wiring board through the plurality of electrodes. It can be connected to a circuit. With such a configuration, for example, a signal from the imaging unit 10 can be read out by the reading unit 20 through the electrodes, and a control signal from the processor 109 can be supplied to the imaging unit 10 through the electrodes.

  The reading unit 20 includes, for example, multiplexers 131 to 138, signal amplification units 141 to 148, and AD conversion units 151 to 158. The multiplexer 131 or the like functions as a selection unit that selects a sensor from which signals are to be read in a predetermined unit. For example, the multiplexer 131 or the like selects the sensor s that is a signal reading target for each sensor chip 106 or each column. The signal amplification unit 141 and the AD conversion unit 151 and the like function as an output unit that outputs a signal (sensor signal) of each sensor s to be selected. For example, the signal amplifying unit 141 or the like amplifies the signal with a differential amplifier or the like, and the AD converting unit 151 or the like performs analog-digital conversion (AD conversion) on the amplified signal.

  The processor 109 exchanges instructions using control signals, control commands, and the like with the control unit 101 via various interfaces, and drives and controls the imaging unit 10 and the reading unit 20 when performing radiography. Radiography. A known communication means may be used for the various interfaces, a wired communication means such as a LAN, or a wireless communication means such as Wi-Fi may be used. The processor 109 reads sensor signals from the sensors s of the imaging unit 10 and combines them to generate image data for one frame (may be expressed as “frame data”). The processor 109 outputs the image data to the control unit 101.

  The control interface 110 is an interface for realizing control of the image acquisition unit 100 by the control unit 101 by exchanging device information such as shooting information input by the user and an operation state of the image acquisition unit 100. . The image data interface 111 is an interface for outputting image data from the image acquisition unit 100 to the control unit 101. Further, the processor 109 notifies the control unit 101 by supplying a READY signal via the interface 112 that the image acquisition unit 100 is ready to start radiation imaging. While the READY signal from the processor 109 is at the active level, the control unit 101 sets the timing of starting radiation imaging for one time to the processor 109 via the interface 113 as a synchronization signal (synchronization signal SYNC). Notify by supplying The processor 109 requests the start of radiation irradiation by supplying an exposure permission signal to the control unit 101 via the interface 114. While the exposure permission signal is at the active level, the control unit 101 causes the radiation source 104 to generate radiation via the radiation source control unit 103.

  The image acquisition unit 100 further includes a nonvolatile memory 360. The processor 109 can read out information necessary for executing radiation imaging (information stored in advance such as eigenvalues and parameters for specifying the operation of each unit) from the memory 360 as necessary. . Further, the processor 109 can update the information held in the memory 360 based on new information received from the outside, and can also cause the memory 360 to hold data obtained during radiation imaging. The image acquisition unit 100 further includes a temperature sensor 370 for detecting the temperature of the imaging unit 10. The processor 109 can also change the control method of the operation in the image acquisition unit 100 based on the detection result by the temperature sensor 370.

  In this example, the control unit 101 controls the entire system of the radiation imaging apparatus IA, whereas the processor 109 is based on an instruction (for example, a signal, a command, etc.) from the control unit 101. Each unit in 100 is driven and controlled. Thus, the sensor signal is output from each sensor s, and the processor 109 generates image data based on the sensor signal and outputs the image data to the control unit 101. The control unit 101 performs processing such as correction processing on the image data. Part or all of the processing may be performed by the processor 109.

  Although the configuration in which the control unit 101 and the processor 109 are separated is shown here, these functions may be configured by a single unit. For example, the control unit 101 can be built in the image acquisition unit 100, and a part or all of the functions of the control unit 101 can be realized by the processor 109. The processor 109 and the control unit 101 may be collectively referred to as a processor, a control unit, or the like, or may be referred to by other expressions such as a calculation unit or a shooting control unit.

  Radiation imaging can be realized with the above configuration. In summary, the image acquisition unit 100 generates image data based on the sensor signal read from each sensor s of the imaging unit 10 and outputs it to the control unit 101. The control unit 101 performs correction processing or the like on the image data, and causes the display unit 102 to display a radiation image based on the image data.

  Each unit in the radiation imaging apparatus IA is not limited to the above configuration, and the configuration of each unit may be appropriately changed according to the purpose and the like. For example, as described above, the control unit 101 and the processor 109 can be configured by a single unit, but the same applies to other units. That is, each function of two or more units may be realized by one unit, and a part of functions of a unit may be realized by another unit.

  FIG. 2A shows a configuration example of a single sensor chip 106. Each sensor chip 106 includes a plurality of sensors s, a vertical scanning circuit 303 for driving the plurality of sensors s, and a horizontal scanning circuit 304 for reading signals from the plurality of sensors s.

  In each sensor chip 106, the plurality of sensors s are arranged so as to form, for example, m rows × n columns. In the figure, for example, the sensors in the first row and the second column are indicated as “s (1, 2)”. As will be described in detail later, each sensor s holds an S signal corresponding to a signal component and an N signal corresponding to a noise component, and the S signal and the N signal are individually output from each sensor s. The

  The vertical scanning circuit 303 and the horizontal scanning circuit 304 are constituted by shift registers, for example, and operate based on a control signal from the processor 109. The vertical scanning circuit 303 functions as a drive unit that drives the sensor s to be read out for each row based on a control signal from the processor 109. Specifically, the vertical scanning circuit 303 supplies drive signals to the plurality of sensors s via the control line 305, and drives the plurality of sensors s in units of rows based on the drive signals. Further, the horizontal scanning circuit 304 sequentially outputs the signals of the sensors s in each column based on a control signal from the processor 109 (also referred to as “horizontal transfer”). Specifically, the horizontal scanning circuit 304 sequentially outputs the signals (S signal and N signal) of the sensor s driven by the vertical scanning circuit 303 via the column signal lines 306 and 307 and the analog output lines 308 and 309. Output to the outside.

Sensor chip 106 has a terminal E _S for reading the S signal held by the sensor s, a terminal E _N for reading the N signal held by the sensor s. The sensor chip 106 further has a select terminal _{E CS,} by signal received at terminal _{E CS} becomes active level, the signal of each sensor s of the sensor chip 106, via the terminal _{E S} and _{E N} Read out.

More specifically, each sensor s has a terminal ts for outputting an S signal and a terminal tn for outputting an N signal, and the terminal ts is connected to the column signal line 306. The terminal tn is connected to the column signal line 307. Column signal lines 306 and 307, through the switch SW _H becomes conductive in response to a control signal from the horizontal scanning circuit 304, respectively, are connected to the analog output line 308 and 309. Signal of the analog output line 308 and 309, through the switch _{SW CS} becomes conductive in response to a signal terminal _{E CS} receive, respectively, output from the terminal _{E S} and _{E N.}

  Here, for each column, the column signal lines 306 and 307 are each provided with a current source 350 for giving an operation reference point thereto. That is, the current source 350 supplies a bias current for reading a sensor signal from each sensor s in the corresponding column to the column signal line 306 (or 307).

  The sensor chip 106 further includes a terminal VST that receives control signals for controlling the vertical scanning circuit 303 and the horizontal scanning circuit 304, and the like. Terminal VST receives a start pulse input to vertical scanning circuit 303. Terminal CLKV receives a clock signal input to vertical scanning circuit 303. Terminal HST receives a start pulse input to horizontal scanning circuit 304. Terminal CLKH receives a clock signal input to horizontal scanning circuit 304. Each of these control signals is supplied from the processor 109.

  With the above configuration, in the sensor chip 106, each sensor s is controlled in units of rows, and the signals (S signal and N signal) of the sensors s in each column are sequentially output to perform signal reading.

FIG. 2B shows a part of the circuit configuration of the reading unit 20. Signal from the terminal E _S is the inverted input terminal of the signal amplifier 141 (in FIG. "-" indicates a) is input, the signal from the terminal E _N during non-inverting input terminal (FIG signal amplifying section 141 Is indicated by “+”). The signal amplifier 141, the difference between the signal from the signal and terminal E _N from the terminal E _S (difference between the signal value) is amplified, the signal corresponding to said difference is output to the AD conversion unit 151. The AD conversion unit 151 receives a clock signal at the CLKAD terminal, and AD converts the signal from the signal amplification unit 141 based on the clock signal. The AD converted signal is output as a sensor signal to the processor 109 via the ADOUT terminal. Here, for ease of explanation, the signal amplification unit 141 and the AD conversion unit 151 have been described with reference to the above (although the multiplexer 131 is not shown), the same applies to the case where the multiplexer 131 is further included. .

FIG. 2C shows an example of a timing chart for explaining a reading operation (reading operation RO) for reading a signal from the image acquisition unit 100. The horizontal axis represents a time axis, and the vertical axis represents each control signal. Here, for ease of explanation, a case where signal reading is performed from four sensor chips 106 (referred to as sensor chips 106 _{0 to} 106 ₃ ) will be described.

The selection signal Sel (Sel0 to Sel3) is a control signal for selecting the sensor chip 106 to be read. The selection signals Sel0 to Sel3 correspond to the sensor chips 106 _{0 to} 106 ₃ and are input to the terminals E _CS of the corresponding sensor chips 106, respectively. For example, when the sensor chip 106 ₁ subject to signal readout is to a high level (H level) and Sel1, other selection signal Sel0, the Sel2 and Sel3, to a low level (L level).

  The other control signals VST and the like indicate control signals input to the respective terminals. For example, a control signal input to the terminal VST is indicated as a signal VST. The same applies to other control signals.

  The signal VST is a row selection start pulse signal. Based on this signal, each sensor s in the first row in the sensor chip 106 selected by the selection signal Sel is selected by the vertical scanning circuit 303. The signal CLKV is a clock signal, and each time the clock signal is received at the terminal CLKV, the selected row is sequentially shifted from the first row to the m-th row (that is, each sensor s is changed from the first row to the first row). (Up to m rows are selected in order for each row.)

  The signal HST is a column selection start pulse signal. Based on this signal, each sensor s in the first column in the sensor chip 106 selected by the selection signal Sel is selected by the horizontal scanning circuit 304. The signal CLKH is a clock signal, and each time the clock signal is received at the terminal CLKH, the selected column is sequentially shifted from the first column to the nth column (ie, each sensor s is shifted from the first column to the first column). (Sequentially selected row by row up to n columns).

  The signal CLKAD is a clock signal. As described above, based on this signal, a signal corresponding to the difference between the S signal and the N signal of each sensor s is AD-converted by the AD conversion unit 108.

First, the signal VST and the signal CLKV becomes H level, the selection signal Sel0~Sel3 becomes H level sequentially, the sensor chip ₁₀₆ 0-106 ₃ are sequentially selected. At a timing when a certain selection signal Sel becomes H level (or after becoming H level), the clock signal CLKH and the signal HST are changed to H level, and then the next selection signal Sel becomes H level. CLKAD is input.

By such a driving method, for example, in the first period T1 in the figure, signal reading from each sensor s in the first row is performed from each of the sensor chips 106 _{0 to} 106 ₃ . Specifically, first, for each sensor s in the first row of the sensor chip 106 ₀ from the first column to the n-th column, sequentially, AD conversion of the signal of each of the sensor s is made. Next, AD conversion of the signal of each sensor s in the first row of the sensor chip 106 ₁ is made in the same manner. Thereafter, AD conversion of the signal of each sensor s in the first row of the sensor chip 106 ₂ is made similarly, Thereafter, AD conversion of the signal of each sensor s in the first row of the sensor chip 106 ₃ are performed in the same manner The Second period T2 (reading signals from each sensor s in the second row of each sensor chip 106), third period T3 (reading signals from each sensor s in the third row of each sensor chip 106), and thereafter (After T4 (not shown)) is the same as the first period T1.

  The read operation RO is performed as described above. Note that the read operation RO is performed from the viewpoint of the vertical scanning circuit 303 that functions as a drive unit, the processor 109 that controls the operation, or the control unit 101 that controls the control. It may be referred to as “drive”.

  FIG. 3A illustrates the circuit configuration of a single sensor s. The sensor s includes a first part ps1, a second part ps2, and a third part ps3.

The first portion ps1 includes a photodiode PD, transistors M1 and M2, a floating diffusion capacitor C _FD (hereinafter, FD capacitor C _FD ), and a sensor sensitivity switching capacitor C _FD '.

The photodiode PD is a photoelectric conversion element, and converts light (scintillator light) generated by the above-described scintillator in accordance with irradiated radiation into an electric signal. Specifically, an amount of charge corresponding to the amount of scintillator light is generated in the photodiode PD, and the voltage of the FD capacitor C _FD corresponding to the amount of generated charge is output to the second portion ps2. Here, the configuration in which the photodiode PD is used as the detection element for detecting radiation while assuming the indirect conversion type imaging unit 10 is illustrated, but other photoelectric conversion elements may be used.

The sensor sensitivity switching capacitor C _FD ′ is used to switch the sensitivity of the sensor s to the radiation, and is connected to the photodiode PD via the transistor M1 (switch element). When the signal WIDE becomes active level, the transistor M1 becomes conductive, and the voltage of the combined capacitance of the FD capacitor _CFD and the capacitor _CFD ′ is output to the second part ps2. That is, the sensor s is in a low sensitivity mode when the signal WIDE is at an H level, and is in a high sensitivity mode when the signal WIDE is at an L level. With such a configuration, the sensitivity to radiation can be changed.

  The transistor M2 resets (initializes) the charge of the photodiode PD when the signal PRES becomes active level, and resets the voltage output to the second portion ps2.

The second portion ps2 includes a transistor M3～M7, and the clamp capacitor _{C CL,} and a constant current source (e.g. transistors of the current mirror configuration). The transistor M3, the transistor M4, and the constant current source are connected in series so as to form a current path. When the enable signal EN input to the gate of the transistor M3 becomes an active level, the transistor M4 receiving the voltage from the first part ps1 performs a source follower operation, and a voltage corresponding to the voltage from the first part ps1 is output. Is done.

The subsequent stage, clamp circuit formed of a transistor M5~7 and the clamp capacitor C _CL is arranged. Specifically, one terminal n1 of the clamp capacitor C _CL, are connected to a node between the transistors M3 and M4 of the first portion ps1, the other terminal n2, the clamp voltage through a transistor M5 Connected to VCL. The transistors M6, M7, and the constant current source are directly connected to form a current path, and the terminal n2 is connected to the gate of the transistor M7.

With such a configuration, kTC noise (so-called reset noise) generated in the photodiode PD of the first portion ps1 is removed. Specifically, a voltage corresponding to the voltage from the first portion ps1 during the aforementioned reset is input to the terminal n1 of the clamp capacitor C _CL. Further, the clamp signal PCL is the transistor M5 is in a conducting state by an active level, the clamp voltage VCL is supplied to the terminal n2 of the clamp capacitor C _CL. Thus, the potential difference between both terminals n1-n2 of the clamp capacitor _{C CL} is clamped as a noise component. In other words, the second portion ps2 functions as a holding unit that holds the voltage corresponding to the charge generated in the photodiode PD and corresponding to the kTC noise by the clamp capacitor _CCL . In this configuration, in the second portion ps2, a voltage obtained by removing the clamped noise component from the voltage output according to the electric charge generated in the photodiode PD from the transistor M4 performing the source follower operation is held.

  The enable signal EN is supplied to the gate of the transistor M6. When the enable signal EN becomes active level, the transistor M7 performs a source follower operation, and a voltage corresponding to the gate voltage of the transistor M7 is output to the third portion ps3. . For example, when the charge is generated in the photodiode PD, the gate voltage of the transistor M7 changes, and a voltage corresponding to the changed voltage is output to the third portion ps3.

The third part ps3 includes transistors M8, M10, M11, and M13, analog switches SW9 and SW12, and capacitors (capacitors) CS and CN. A unit formed by the transistors M8 and M10, the analog switch SW9, and the capacitor CS is referred to as a “first unit U _SHS ”.

In the first unit _USSHS , the transistor M8 and the capacitor CS form a sample and hold circuit. Specifically, by switching the state (conductive state or nonconductive state) of the transistor M8 using the control signal TS, the signal from the second portion ps2 is held in the capacitor CS as the S signal. In other words, the first unit U _SHS functions as a first sampling unit that samples the S signal. Further, under the state where the analog switch SW9 is turned on by the control signal VSR, the transistor M10 performs a source follower operation to amplify the S signal, and the amplified S signal is connected to the terminal via the analog switch SW9. is output from ts.

Similarly to the first unit U _SHS , the transistors M11 and M13, the analog switch SW12, and the capacitor CN form a “second unit U _SHN ” that outputs a signal from the terminal tn. In the second unit _USHN , the N signal is held by the capacitor CN. In other words, the second unit _USHN functions as a second sampling unit that samples N signals.

  With such a configuration, the reading unit 20 reads the difference between the S signal and the N signal via the terminals ts and tn. Thereby, fixed pattern noise (FPN: Fitted Pattern Noise) resulting from the second portion ps2 is removed.

  With the above configuration, in the sensor s, the S signal and the N signal are held in the capacitors CS and CN, respectively, and the held S signal and N signal make the analog switches SW9 and SW12 conductive, respectively. Thus, reading is performed by so-called nondestructive reading. That is, while the transistors M8 and M11 are in the non-conductive state, the held S signal and N signal can be read at an arbitrary timing.

  When reading the S signal or the N signal, a bias current is supplied from the current source 350 to the column signal line 306 or 307. Although details will be described later, 1 / f noise (flicker noise) may be generated in the current source 350, and the current value of the bias current may periodically fluctuate (periodic fluctuation may occur).

  FIG. 3B is a timing chart for explaining an example of a driving method of the sensor chip 106 in a case where one radiography is performed (when image data for one frame is acquired). A series of operations described below is an example of a driving method in the case of still image shooting, and moving image shooting and continuous shooting can be realized by repeatedly executing the series of operations. Here, the case of the high sensitivity mode (that is, the control signal WIDE is at the L level) will be described, but the case of the low sensitivity mode (that is, the control signal WIDE is at the H level) may be considered similarly.

At time t50, shooting information (setting information necessary for shooting such as operation mode, frame rate, and other parameters) is set. Next, at time T51～t56, in response to the synchronization signal SYNC from the control unit 101, a reset driving RD for resetting each sensor s and the clamp capacitor _{C CL.} Then, from time t60 to t69, sampling drive SD for reading the image signal is performed. Thereafter, the aforementioned read operation RO (see FIG. 2) is performed.

As shown in the enlarged portion at times t51 to t56, in the reset driving RD, the reset operation for resetting the photodiode PD in response to the synchronization signal SYNC and the voltage corresponding to the kTC noise are held in the clamp capacitor _CCL . To perform actions. First, at time t51, the enable signal EN is set to H level, and the transistors M3 and M6 are turned on. As a result, the transistors M4 and M7 enter a state in which the source follower operation is performed.

At time t52, the signal PRES is set to H level, so that the transistor M2 is turned on. Thus, the photodiode PD is connected to the reference voltage VRES, the voltage of the capacitor C _FD with the photodiode PD is reset is reset. Further, a voltage corresponding to the gate voltage of the transistor M4 during the reset is supplied to one terminal n1 (transistor M4 side terminal) of the clamp capacitor C _CL.

At time t53, the signal PCL is set to H level, and the transistor M5 is turned on. Accordingly, the clamp voltage VCL is supplied to the terminal n2 (transistors M7 side terminal) of the clamp capacitor _{C CL.}

At time t54, the signal PRES is set to L level, and the transistor M2 is turned off. Thus, terminal n1 of the clamp capacitor C _CL is set to a voltage corresponding to the gate voltage of the transistor M4 during the reset.

At time t55, the signal PCL is set to L level, and the transistor M5 is turned off. Thus, kTC noise charge corresponding to a potential difference between the terminal n1 and the terminal n2 (charge based on the potential difference between the reference voltage VRES the clamp voltage VCL) is held by the clamp capacitance C _CL, due to heat or the like of the photodiode PD Is clamped.

  At time t56, the enable signal EN is set to L level, and the transistors M3 and M6 are turned off. This puts the transistors M4 and M7 into a non-operating state. Thereafter, the above-mentioned exposure permission signal is set to H level.

As described above, a series of operations of the reset driving RD is completed. That is, the reset driving RD, resets the photodiode PD, and resets the clamp capacitor C _CL, the said reset the clamp capacitor C _CL a voltage corresponding to the kTC noise is maintained. Thereafter, radiation is irradiated, and charges corresponding to the radiation dose are generated in the photodiode PD. Note that the reset driving RD is performed for all the sensors at once, and the continuity of data between the sensor chips and between the sensors is maintained by preventing a shift in control timing.

  Next, as shown in the enlarged portion at times t60 to t68, the sampling drive SD performs an operation of sampling a signal level corresponding to the amount of charge generated in the photodiode PD as an S signal and holding it in the capacitor CS. . In the sampling drive SD, an operation is performed in which a noise level corresponding to fixed pattern noise caused by the configuration of the sensor s, manufacturing variations of each element, and the like is sampled as an N signal and held in the capacitor CN.

At time t60, the enable signal EN is set to the H level, the transistors M3 and M6 are turned on, and the transistors M4 and M7 perform the source follower operation. The gate voltage of the transistor M4 changes according to the amount of electric charge generated in the photodiode PD accumulating a voltage corresponding to said change the gate voltage is input to the terminal n1 of the clamp capacitor C _CL, the potential of the terminal n1 Change. Then, in accordance with the potential change of the terminal n1, the potential of the terminal n2 of the clamp capacitor _{C CL} is changed.

  At time t61, the signal TS is set to H level, and the transistor M8 is turned on. As a result, a voltage corresponding to the potential of the terminal n2 (the above-described changed potential of the terminal n2) is charged in the capacitor CS.

  At time t62, the signal TS is set to L level, and the transistor M8 is turned off. As a result, the voltage is fixed to the capacitor CS (S signal sampling). At time t62, the exposure permission signal is set to the L level. Note that the period from time t54 to t62 corresponds to the charge accumulation time in the photodiode PD. In this period, in addition to the amount of charge corresponding to the amount of irradiated radiation, the charge is caused by dark current or the like. An amount of charge corresponding to the period is accumulated in the photodiode PD.

At time t63, the signal PCL is set to H level, and the transistor M5 is turned on. Accordingly, the clamp voltage VCL is supplied to the terminal n2 (transistors M7 side terminal) of the clamp capacitor _{C CL.}

At time t64, the transistor M2 in a conductive state by a signal PRES to H level, while the reference voltage VRES resets the voltage of the FD capacitor _{C FD,} the voltage of the terminal n1 also reset. After that, at time t65, the signal PRES is set to L level, and the transistor M2 is turned off. Thus, terminal n1 of the clamp capacitor C _CL is set to a voltage corresponding to the gate voltage of the transistor M4 during the reset.

  At time t66, the signal TN is set to H level, and the transistor M11 is turned on. As a result, a voltage according to the potential of the terminal n2 (the above-mentioned supplied voltage VCL) is charged in the capacitor CN. After that, at time t67, the signal TN is set to L level, and the transistor M11 is turned off. As a result, the voltage is fixed to the capacitor CN (N signal sampling).

  Finally, at time t68, the signal PCL is set to L level to turn off the transistor M5, and at time t69, the enable signal EN is set to L level to turn off the transistors M3 and M6 (transistors M4 and M7 are in a non-operating state). ).

  In summary, in the sampling drive SD, the S signal is sampled at times t61 to t62, the photodiode PD is reset and the N signal is sampled at times t63 to t68.

  As described above, a series of operations of the sampling drive SD is completed. That is, in the sampling drive SD, the signal level corresponding to the amount of electric charge generated in the photodiode PD is sampled as an S signal and held in the capacitor CS, and the noise level corresponding to the fixed pattern noise is sampled as an N signal to measure the capacitance. Hold on CN. Note that the sampling drive SD can be performed for all the sensors at the same time as the reset drive RD described above.

  Thereafter, as described above, in the read operation RO performed after the sampling drive SD, the S signal and the N signal are read, a signal corresponding to the difference between them is AD-converted, and the group of the AD-converted signals Based on the image data, image data is formed.

(About offset correction)
When radiation imaging is performed using the radiation imaging apparatus IA, image data obtained from the imaging unit 10 irradiated with radiation (hereinafter referred to as “radiation image data” for distinction) may include an offset component. The offset component is a noise component that can appear in the radiation image data separately from the signal component corresponding to the radiation dose, and the sensor signal even when radiation is not irradiated (when the signal component is substantially 0). It is a noise component contained in. In other words, the offset component also appears in image data obtained from the imaging unit 10 that has not been irradiated with radiation (hereinafter referred to as “dark image data” for distinction). That is, each signal constituting the dark image data has a noise component that gives a signal value that is not substantially zero. On the other hand, each signal constituting the radiation image data has this noise component and And a signal component corresponding to the radiation dose.

  Therefore, in order to remove the offset component from the radiographic image data, the radiographic image data needs to be corrected by the control unit 101 (or the processor 109) described with reference to FIG. This correction may be referred to as “offset correction”. For example, radiation image data obtained from the imaging unit 10 irradiated with radiation and dark image data obtained from the imaging unit 10 not irradiated with radiation. This can be done by taking the difference. The dark image data may also be referred to as “offset data”, “offset image data”, or the like.

(Effects of radiographic images due to 1 / f noise)
By the way, when reading the S signal or the N signal from each sensor s, a bias current is supplied from the current source 350 to the column signal line 306 or 307. As described above, 1 / f noise may be generated in the current source 350, and the current value of the bias current is periodically (depending on the configuration of the current source 350, for example, several [kHz] to several [MHz] Can vary (at lower frequencies). The 1 / f noise is considered to be caused by trapping of charges due to crystal defects. For example, the current source 350 may be a MOS transistor that can act as a constant current source by, for example, a current mirror structure. In this case, charge traps and releases occur in lattice defects in the channel region of the MOS transistor. In the bias current, 1 / f noise is observed as a fluctuation (fluctuation) in the periodicity of the current value caused by the trapping and releasing of the charge.

FIG. 4A is a diagram for explaining the time dependency of the current value I _REF of the bias current of a certain current source 350, and the horizontal axis indicates the time axis. The current value I _REF may vary over time due to 1 / f noise, and the variation may have substantial periodicity. For example, the current value I _REF may gradually change from the median value I _A to the value I _B (> I _A ) or the value I _C (<I _A ) over time. Here, the S signal and the N signal forming the sensor signal are respectively read out through the column signal lines 306 and 307 supplied with the bias current of the current value I _REF as the operation reference point, and the current value I _REF Fluctuations cause fluctuations in the sensor signal.

Incidentally, the median I _A in the figure, obtained corresponds to the design value (target value), the average value may be expressed as an effective value or the like. The value I _B and I _C are relatively large current value or a relatively small current value when varied to give, it does not necessarily indicate the maximum or minimum value.

  Here, referring to FIGS. 4B1 and 4B2, it may be caused by different timings of acquiring two image data (referred to as “first data D1” and “second data D2”, respectively). The influence of the 1 / f noise will be described. Actually, the data D1 corresponds to the radiation image data, and the data D2 corresponds to the dark image data. Here, for the sake of easy explanation, it is assumed that both the data D1 and D2 are dark image data. That is, it is assumed that the data D1 and D2 have only an offset component and substantially no signal component. Therefore, when the difference between the data D1 and the data D2 is taken (when the data D1 is corrected using the data D2), the signal value of the data (D1−D2) obtained thereby may be substantially zero. Shall.

FIG. 4B1 shows signal values when the two image data D1 and D2 are acquired ideally simultaneously (at the same timing) as the first comparative example. The horizontal axis in the figure indicates the column number, and illustrates the (K-2) th column, the (K-1) th column, the Kth column, the (K + 1) th column, and the (K + 2) th column. Yes. As described above, the current value I _REF may periodically change with time. However, in the first comparative example of FIG. 4B1, the data D1 and D2 are ideally acquired simultaneously, so the data D1 and the data D2 Are compared for each column, the signal values are equal to each other. Therefore, when the difference between the data D1 and the data D2 is taken, the difference (D1-D2) in the signal value is substantially 0 in any column.

  However, in practice, there may be a time difference in the timing of acquisition of the two data D1 and D2, and therefore the signal values of each column may be different between the data D1 and the data D2.

  FIG. 4B2 shows signal values when two image data D1 and D2 are acquired at different timings as a second comparative example (for comparison, the data D1 is shown in FIG. 4). 4 (b1) is the same). In the second comparative example of FIG. 4 (b2), the signal value of each column is different between the data D1 and the data D2, so there are many columns in which the difference (D1-D2) in the signal value is not substantially zero. In the signal value difference (D1−D2), “variation” occurs between the columns.

As described above, the current source 350 is disposed in each of the column signal lines 306 and 307 in each column, and the current value I _REF of the bias current may be different between the columns, but the column signal lines 306 and 307 in the same column are different. It can be different from the column signal line 307. Therefore, according to the second comparative example of FIG. 4 (b2), in the corrected image data (that is, D1-D2), a difference in signal value due to 1 / f noise occurs between the columns. Indicates that, for example, striped spots appear in the radiographic image, and the quality of the image can be degraded.

Here, the example (this example) of FIG.4 (c) has shown the signal value of data D1 and D2 when the averaging process is performed about data D2 among two image data D1 and D2 (Note that For comparison, the data D1 is the same as that shown in FIGS. 4B1 and 4B2.) In the present example, as compared to the second comparative example of FIG. 4 (b2), the signal value of each column of the data D2 has become close to the median value I _A, as a result, the difference between the signal value (D1-D2) The above-mentioned variation is reduced. Therefore, according to this example, striped spots that can appear in the radiation image are reduced or suppressed. Hereinafter, specific examples according to this example will be described with reference to some embodiments and their modifications.

(First embodiment)
FIG. 5 is a flowchart illustrating an example of a driving method of the radiation imaging apparatus IA. As described with reference to FIG. 1, the control sequence according to this flowchart can be substantially executed by the processor 109 that receives an instruction from the control unit 101, but a part of the control sequence is executed by the control unit 101. May be.

  In step S1100 (hereinafter simply referred to as S1100; the same applies to other steps), dark image data acquisition conditions are set. In S1100, for example, necessary parameters and the like are set for each unit that is driven and controlled when dark image data is acquired. For example, the frame rate for acquiring dark image data (time required to acquire image data for one frame, specifically, charge accumulation time in the photodiode PD, time required for vertical scanning and horizontal scanning, Data transfer time). Further, for example, an operation mode (for example, a high sensitivity mode or a low sensitivity mode) for acquiring dark image data is set.

  When acquiring radiographic image data, which will be described later, a condition different from the condition set in S1100 may be adopted. Therefore, a predetermined condition (typical condition) is set in S1100. Alternatively, it may be set based on the use history by the user. For example, in the example of the frame rate, a typical one can be selected from the range of 10 to 200 [FPS], and can be set as the frame rate for dark image data acquisition.

  In S1110, a plurality of dark image data is acquired based on the conditions set in S1100. The plurality of dark image data are image data for a plurality of frames. Further, the number of dark image data to be acquired in S1110 only needs to be set in S1100, and at least two are necessary, but it may be larger.

  In the present embodiment, in the above-described S1100, a plurality of dark image data is set to be acquired in order at a time interval different from the time interval of the 1 / f noise cycle (first cycle) in the current source 350. Shall be. In order to realize this, a timer (not shown) can be arranged in the image acquisition unit 100, and the processor 109 can acquire (start acquisition) each of a plurality of dark image data based on the measurement value by the timer. Timing can be adjusted.

  The period of the 1 / f noise is measured in advance by spectrum analysis, for example, before shipment of the radiation imaging apparatus IA (inspection process of the manufactured apparatus IA, evaluation process of the formed current source 350, etc.), and stored in the memory 360. It only has to be stored. In step S1110, the processor 109 refers to the memory 360 so that the acquisition timing of each of the plurality of dark image data is set to a time interval different from the time interval of the 1 / f noise cycle. Adjust it. Note that the period of the 1 / f noise stored in the memory 360 may be updated during maintenance of the device IA.

  In the present embodiment, the timing of obtaining each of the plurality of dark image data is assumed to be a time interval larger than the time interval of the 1 / f noise period, but the time interval of the 1 / f noise period Smaller time intervals may be used. Note that the time difference between the acquisition timing of a certain dark image data and the acquisition timing of the next dark image data may be different from the time interval of the cycle of the 1 / f noise. It is not necessary to acquire image data periodically (at the same interval).

  Here, it is assumed that the period of fluctuation of the signal value due to the 1 / f noise is one kind, but two or more kinds of periods may be mixed, in which case What is necessary is just to acquire several dark image data in order by the time interval different from the time interval of this period. In particular, a plurality of dark image data may be sequentially acquired at a time interval larger than the maximum time interval of the two or more types of cycles.

  In S1120, it is determined whether or not there is an instruction to start radiography. If the start instruction is given, imaging conditions (radiation dose (intensity, time), necessary parameters for each unit driven and controlled when acquiring radiation image data) included in the start instruction are set. Then, the process proceeds to S1130. Further, since the environment such as the temperature of the radiation imaging apparatus IA may change with time, if there is no start instruction, the process returns to S1110 and a plurality of dark image data is acquired again.

  In S1130, it is determined whether correction data used for offset correction is ready. If correction data is not ready, the process proceeds to S1140. If correction data is ready, the process proceeds to S1150.

  In S1140, correction data is generated. The correction data is generated based on the plurality of dark image data obtained in S1110. Specifically, the correction data is an average of a plurality of dark image data (more specifically, each of the plurality of dark image data is configured. The signal value to be added is averaged for each pixel).

  Here, as described above, the plurality of dark image data obtained in S1110 are sequentially acquired at a time interval different from the time interval of the 1 / f noise period in the current source 350. For this reason, in S1140, by averaging the plurality of dark image data, fluctuations in the signal value (substantially only the offset component) due to 1 / f noise in each dark image data are appropriately adjusted. It is thought that it will be canceled. In other words, each signal value of the dark image data that may vary depending on the acquisition timing is averaged so as to approach the median value. If a plurality of dark image data is obtained at the same time interval as the time interval of the 1 / f noise cycle, the fluctuation caused by the 1 / f noise is not canceled appropriately even if the addition average of them is taken ( Each signal value of dark image data does not approach the median value.)

  In this embodiment, the correction data is generated by adding and averaging a plurality of dark image data. However, if necessary, weighted averaging may be performed, and standard deviation and the like are taken into consideration. Other arithmetic processes may be performed incidentally.

  In S1150, it is determined whether or not the correction data needs to be updated. If necessary, the process returns to S1110, and if not necessary, the process proceeds to S1160. The determination in S1150 can be made based on, for example, that a predetermined period has elapsed since the correction data was first generated or the correction data was last updated, and / or that the temperature has changed accordingly. The temperature here is the temperature of the imaging unit 10 and can be detected by the temperature sensor 370 described with reference to FIG.

  In S1160, radiography is started. The operation of the radiation imaging apparatus IA at the time of radiation imaging is as described with reference to FIGS. 2 to 3, whereby radiation image data is obtained.

  In S1170, the above-described offset correction is performed on the radiation image data. The offset correction is performed using the correction data generated in S1140 (specifically, by taking the difference between the radiation image data and the correction data), thereby removing the offset component from the radiation image data. Is done.

  In S1180, the radiographic image data corrected in S1170 is output to the display unit 102 described with reference to FIG. 1, and the radiographic image is displayed on the display unit 102.

  As described above, in the correction data generated in S1140, the fluctuation of the signal value due to the 1 / f noise is appropriately canceled. Therefore, in S1170, the radiation image data can be appropriately offset-corrected. Therefore, according to the corrected radiographic image data, striped spots that may appear in the radiographic image can be suppressed or reduced, and the quality of the image is improved.

  In S1190, it is determined whether or not to continue radiation imaging. If radiation imaging is to be continued (when not stopped or interrupted), the process proceeds to S1200, and if not (when stopped or interrupted), the process proceeds to S1210. For example, when performing moving image shooting or continuous shooting, the series of operations from S1160 to S1180 can be repeatedly executed until an instruction to stop or interrupt the shooting is input.

  In S1200, it is determined whether or not the correction data needs to be updated. If necessary, the process returns to S1110, and if not necessary, the process returns to S1160. That is, even if it is determined in S1190 that radiography is to be continued, if it is determined in S1200 that the correction data needs to be updated due to the passage of time or the accompanying temperature change, the imaging is stopped and the process returns to S1110. . On the other hand, if it is not necessary to update the correction data, a series of operations from S1160 to S1180 are executed. S1200 may be performed similarly to S1150.

  In S1210, it is determined whether or not to end radiation imaging. If the radiography is to be terminated, this sequence is terminated. If not, the process returns to S1100, and, for example, setup for the next radiography is performed.

  According to the present embodiment, a plurality of dark image data is sequentially acquired at a time interval different from the time interval of the 1 / f noise period in the current source 350 (S1110), and correction data is generated based on them. Therefore, it is possible to obtain correction data in which the fluctuation of the signal value due to the 1 / f noise is appropriately canceled. Therefore, since the radiographic image data is appropriately offset-corrected using this correction data (S1170), according to the radiographic image data after correction, striped spots that may appear in the radiographic image can be suppressed or reduced. , Improve the image quality.

  The above flowchart is merely an example of a driving method of the radiation imaging apparatus IA for canceling the fluctuation of the signal value caused by the 1 / f noise, and even if a part thereof is changed without departing from the gist thereof. Good. For example, not all of the plurality of dark image data obtained in S1110 need be used in S1140. Specifically, in S1110, a group of dark image data is acquired at a predetermined cycle, and in S1140, a part of the group of dark image data (obtained at a time interval different from the time interval of the 1 / f noise cycle). Correction data may be generated. That is, a plurality of dark image data appropriate for canceling the fluctuation of the signal value due to the 1 / f noise is selected in the subsequent step (S1140) from the group of dark image data obtained in S1110. Are extracted, and correction data is generated based on the extracted data. By this method, the same effect as described above can be obtained.

(Second Embodiment)
As described in the first embodiment described above, when acquiring radiographic image data, a condition different from the condition set in S1100 may be adopted. Therefore, in S1100, a predetermined condition ( Typical conditions) can be set. The second embodiment is different from the first embodiment in that a plurality of dark image data is acquired for each of a plurality of imaging conditions that can be specified at the time of acquisition of radiation image data in S1100.

  That is, in the second embodiment, for example, when the high sensitivity mode / low sensitivity mode (signal WIDE is L level / H level) and the frame rate is selected from within a predetermined range, for each frame rate, and a combination thereof, Acquire a plurality of dark image data. If a radiation imaging start instruction is received in S1120, correction data is generated in S1130 using a plurality of dark image data corresponding to the imaging conditions included in the start instruction.

  According to the second embodiment, the same effects as those of the first embodiment can be obtained, and offset correction that can cope with arbitrary shooting conditions can be performed, which is further advantageous in improving the quality of an image.

(Other)
The present invention supplies a program that realizes one or more functions of the above-described embodiment to a system or apparatus via a network or a storage medium, and one or more processors in a computer of the system or apparatus read and execute the program May be realized. For example, the present invention may be realized by a circuit (for example, ASIC) that realizes one or more functions.

  As mentioned above, although some suitable aspects were illustrated, this invention is not limited to these examples, The one part may be changed in the range which does not deviate from the meaning of this invention. In addition, it is needless to say that each term described in this specification is merely used for the purpose of describing the present invention, and the present invention is not limited to the strict meaning of the term. The equivalent can also be included.

  IA: radiation imaging apparatus, s: pixel, 10: imaging unit, 100: image acquisition unit, 101: control unit, 109: processor, 306 to 307: column signal line, 303: drive unit, 350: current source, 360: Memory, 370: Temperature sensor.

Claims (11)

An imaging unit in which a plurality of pixels are arranged to form a plurality of rows and a plurality of columns;
A plurality of column signal lines respectively corresponding to the plurality of columns and connected to each pixel of the corresponding column;
A plurality of current sources respectively corresponding to the plurality of columns, each supplying a bias current to a column signal line of the corresponding column;
After the imaging unit is irradiated with radiation, image data is acquired based on pixel signals output from the plurality of pixels of the imaging unit via the plurality of column signal lines, and the image data is used as correction data. A processor for correcting based on,
The processor is
The image data obtained from the imaging unit in a state in which no radiation is irradiated is used as dark image data, and a plurality of dark image data is converted into a cycle of fluctuations in the value of the bias current caused by 1 / f noise in each current source. In order at a time interval different from the time interval of the first cycle,
The radiation imaging apparatus, wherein the correction data is generated based on the plurality of dark image data. A memory for storing the first period;
2. The processor according to claim 1, wherein the processor sequentially acquires the plurality of dark image data at a time interval different from the time interval of the first period with reference to the first period stored in the memory. The radiation imaging apparatus described. The radiation imaging apparatus according to claim 1, wherein an average of the plurality of dark image data is acquired as the correction data. A drive unit for driving the imaging unit;
The drive unit drives the imaging unit that is not irradiated with radiation in a predetermined cycle to output a group of dark image data from the imaging unit,
The radiographic imaging according to any one of claims 1 to 3, wherein the processor selects a part of the group of the output dark image data as the plurality of dark image data. apparatus. The processor selects, as the plurality of dark image data, a group obtained at a time interval different from the time interval of the first period from the group of the output dark image data. Item 5. The radiation imaging apparatus according to Item 4. The radiation imaging apparatus according to claim 4, wherein the predetermined period is different from the first period. The radiation processor according to any one of claims 1 to 6, wherein the processor acquires the plurality of dark image data at a time interval larger than the time interval of the first period. When a predetermined period has elapsed since the generation of the correction data, the processor newly acquires a plurality of dark image data in order at a time interval different from the time interval of the first period, and newly adds correction data. The radiation imaging apparatus according to any one of claims 1 to 7, wherein: A temperature sensor for detecting the temperature of the imaging unit;
The processor is configured to newly acquire a plurality of dark image data in order at a time interval different from the time interval of the first period based on a detection result by the temperature sensor, and newly generate correction data. The radiation imaging apparatus according to any one of claims 1 to 8. An imaging unit in which a plurality of pixels are arranged to form a plurality of rows and a plurality of columns;
A plurality of column signal lines respectively corresponding to the plurality of columns and connected to each pixel of the corresponding column;
A plurality of current sources respectively corresponding to the plurality of columns and supplying a bias current to the column signal lines of the corresponding columns,
Obtaining a period of variation in the value of the bias current due to 1 / f noise in each current source as a first period;
Using the image data obtained from the imaging unit not irradiated with radiation as dark image data, a plurality of dark image data is sequentially acquired at a time interval different from the time interval of the first period, and the plurality of dark images Generating correction data based on the data;
And a step of correcting image data obtained from the imaging unit based on the correction data after the imaging unit is irradiated with radiation. A method for driving a radiation imaging apparatus, comprising:   A program for causing a computer to execute the driving method according to claim 10.

Images