Photon counting device and photon counting method

Abstract

A photon counting device includes: a plurality of pixels that are a plurality of pixels constituting an image sensor, and include a photoelectric conversion element that converts input light into electric charges and an amplifier that amplifies the electric charges converted by the photoelectric conversion element and converts the electric charges into a voltage; an A / D converter that converts the voltage output from the amplifier of the plurality of pixels into a digital value; and a conversion section that corrects the digital value by correcting a deviation of a gain and an offset value of each of the amplifiers constituting the plurality of pixels, and converts the corrected digital value into a photon number based on corresponding threshold data for distinguishing the photon number.

Description

[0001] This application was filed on [date]. June 18, 2018 Application number is 201880075244.2 The invention is named Photon meter Counting devices and photon counting methods A divisional application of the patent application. Technical Field

[0002] This disclosure relates to a photon counting device and a photon counting method. Background Technology

[0003] For example, Non-Patent Document 1 describes a photon counting technique using a CMOS image sensor. In this technique, by increasing the frame rate of the image sensor, images are captured under the condition that only one photon is incident on one pixel in one frame.

[0004] Existing technical documents

[0005] Non-patent literature

[0006] Non-Patent Literature 1: B Saleh Masoodian, Jiaju Ma, Dakota Starkey, Yuichiro Yamashita, and Eric R. Fossum, “A1Mjot 1040fps 0.22e-rms Stacked BSI Quanta Image Sensor with Cluster-Parallel Readout”, preprint from the 2017 International Image Sensor Workshop (IISW), May 30-June 2, 2017, pp. 230-233. Summary of the Invention

[0007] The problem that the invention aims to solve

[0008] For example, when considering photon counting using a CMOS image sensor, one might think of determining the number of photons based on the digital value output from an A / D converter. However, in a CMOS image sensor, each pixel that makes up the sensor has readout noise. Furthermore, the gain and offset values ​​of multiple pixels can deviate within a certain range. Therefore, there might be cases where the digital value for one incident photon represents the same value as the digital value for two incident photons, raising concerns about decreased photon counting accuracy.

[0009] One aspect of this disclosure is to provide a photon counting device and a photon counting method that can suppress the decrease in photon counting accuracy.

[0010] Technical solutions for solving the problem

[0011] One aspect of the photon counting device includes: a plurality of pixels comprising a photoelectric conversion element that converts input light into electrical charge and an amplifier that amplifies the charge converted by the photoelectric conversion element and converts it into voltage; an A / D converter that converts the voltage output from the amplifier of the plurality of pixels into digital values; and a conversion unit that converts the digital values ​​output from the A / D converter into photon counts per plurality of pixels, with reference to reference data created based on gain and offset values ​​for each of the plurality of pixels.

[0012] In such a photon counting device, an amplifier outputs a voltage corresponding to the photons input to the photoelectric conversion element. The output voltage is converted into a digital value by an A / D converter. Then, a conversion unit converts the digital value into a photon count. In the conversion unit, the digital value is converted into a photon count using reference data. The reference data is created based on the individual gain and offset values ​​of multiple pixels. Therefore, even if there are deviations in the gain and offset values ​​between pixels, the conversion unit can suppress the effects of these deviations and convert the digital value into a photon count. Consequently, the decrease in photon counting accuracy can be suppressed.

[0013] Furthermore, the reference data includes multiple threshold data created based on the gain and offset values ​​of each of the multiple pixels. The conversion unit can convert the digital values ​​of each of the multiple pixels into photon counts based on these multiple threshold data. In this configuration, even if there are deviations in the gain and offset values ​​of each pixel, the photon count can be derived using the threshold corresponding to the deviation.

[0014] Furthermore, the reference data includes preset parameters common to multiple pixels, corresponding to the gain and offset values. The conversion unit can correct the digital value of each of the multiple pixels based on the deviation of the gain and offset values ​​from the parameters, and convert the corrected digital value into a photon count. In this configuration, the digital value can be corrected based on the deviation of the gain and offset values; therefore, for example, the digital value can be converted into a photon count using the same threshold for all pixels.

[0015] Furthermore, the amplifier's readout noise can be below 0.2 [e-rms]. In this case, for example, the false detection rate can be suppressed to below 1%. Further, the amplifier's readout noise can be below 0.15 [e-rms]. In this case, for example, the false detection rate can be suppressed to below 0.1%.

[0016] Furthermore, the gain can be 10 [DN / e] or higher. By increasing the gain, it is possible to reproduce the analog values ​​output from the amplifier with high accuracy.

[0017] Furthermore, one aspect of the photon counting method includes: a step of converting light input to each photoelectric conversion element constituting a plurality of pixels into charge; a step of amplifying the converted charge by an amplifier constituting the plurality of pixels and converting it into voltage; a step of converting the voltage output from each amplifier into a digital value by an A / D converter and outputting it; and a step of converting the digital value of each of the plurality of pixels output from the A / D converter into a photon count based on each gain and offset value in the plurality of pixels.

[0018] In this photon counting method, the voltage output from the amplifier based on the input photons is converted into a digital value by an A / D converter. Then, the digital value is converted into a photon count based on the individual gain and offset values ​​of multiple pixels. Therefore, even if there are deviations in the gain and offset values ​​between pixels, the effects of these deviations can be suppressed when converting the digital value into a photon count. Consequently, the decrease in photon counting accuracy can be suppressed.

[0019] Furthermore, in the step of converting digital values ​​to photon counts, threshold data created for each of the multiple pixels based on individual gain and offset values ​​in the multiple pixels can also be used to convert the digital values ​​to photon counts. In this configuration, even if there are deviations in the gain and offset values ​​of each pixel, the photon count can be derived using a threshold corresponding to the deviation.

[0020] Furthermore, in the step of converting digital values ​​to photon counts, the digital values ​​of each of multiple pixels are corrected based on the deviation of the gain and offset values ​​from the parameters. The corrected digital values ​​are then converted into photon counts. The parameters correspond to the gain and offset values ​​and can also be preset in a manner common to multiple pixels. In this configuration, the digital values ​​can be corrected based on the deviation of the gain and offset values; therefore, for example, the same threshold can be used to convert digital values ​​into photon counts across all pixels.

[0021] Invention Effects

[0022] Based on a photon counting device and photon counting method, it is possible to suppress the decrease in photon counting accuracy. Attached Figure Description

[0023] Figure 1 This is a diagram illustrating the configuration of a photon counting device according to one embodiment.

[0024] Figure 2 It is a graph showing the relationship between the number of electrons and the probability density.

[0025] Figure 3 It is a graph showing the relationship between readout noise and false detection rate.

[0026] Figure 4 (a) is a graph showing the relationship between the number of electrons and the probability density. Figure 4(b) indicates that it is based on Figure 4 (a) is a graph showing the results of the simulation experiment.

[0027] Figure 5 (a) is a graph showing the relationship between the number of electrons and the probability density. Figure 5 (b) is based on Figure 5 (a) The results of the simulation experiment are converted into a graph of numerical values.

[0028] Figure 6 (a) is a graph showing the relationship between the number of electrons and the probability density. Figure 6 (b) is based on Figure 6 (a) The results of the simulation experiment are converted into a graph of numerical values.

[0029] Figure 7 (a) is a graph showing the relationship between the number of electrons and the probability density. Figure 7 (b) is based on Figure 7 (a) The results of the simulation experiment are converted into a graph of numerical values.

[0030] Figure 8 It is a diagram that schematically illustrates the process of converting measured digital values ​​into photon counts.

[0031] Figure 9 It is a diagram that schematically illustrates the process of converting measured digital values ​​into photon counts.

[0032] Figure 10 This is a diagram that schematically illustrates the process of deriving offset values.

[0033] Figure 11 This is a diagram that schematically illustrates the process of deriving the gain.

[0034] Figure 12 This is a graph showing the correspondence between gain and offset values ​​and threshold values.

[0035] Figure 13 It is a diagram that schematically illustrates the process of converting measured digital values ​​into photon counts.

[0036] Figure 14 This is a flowchart illustrating the operation of a photon counting device according to one embodiment.

[0037] Figure 15 It is a graph showing the correspondence between the measured digital values ​​and the corrected digital values.

[0038] Figure 16 This is a flowchart illustrating the operation of a photon counting device in other embodiments. Detailed Implementation

[0039] Hereinafter, the embodiments will be described in detail with reference to the accompanying drawings. For convenience, substantially identical elements may be labeled with the same symbols and their descriptions may be omitted. Furthermore, the photon count in the embodiments includes both the count of the number of photoelectrons generated by each pixel of the image sensor and the count of the number of photons taking into account the quantum efficiency (QE) of the image sensor.

[0040] [First Implementation Method]

[0041] Figure 1 This is a diagram showing the configuration of a photon counting device. (For example...) Figure 1 As shown, the photon counting device 1 includes a CMOS image sensor 10 and a computer 20 connected to the CMOS image sensor 10. The CMOS image sensor 10 includes: a plurality of pixels 11; and an A / D converter 15. The plurality of pixels 11 are configured in a two-dimensional shape, arranged along the row and column directions. Each pixel 11 has a photodiode (photoelectric conversion element) 12 and an amplifier 13. The photodiode 12 stores electrons (photoelectrons) generated by the input of photons as charge. The amplifier 13 converts the charge stored in the photodiode 12 into voltage and amplifies it. The amplified voltage is transmitted to the vertical signal line 16 row by row by switching the selection switch 14 of each pixel 11. A CDS (correlated double sampling) circuit 17 is arranged on each vertical signal line 16. The CDS circuit 17 removes noise that is misaligned between pixels and temporarily stores the transmitted voltage.

[0042] The A / D converter 15 converts the voltages output from each amplifier 13 in the plurality of pixels 11 into digital values. In this embodiment, the A / D converter 15 converts the voltage stored in the CDS circuit 17 into digital values. The converted digital values ​​are then output to the computer 20. For example, the digital values ​​can also be transmitted to a horizontal signal line (not shown) via column selection switching and then output to the computer 20. In this way, in the CMOS image sensor 10, when photons are input to each pixel 11, a digital value corresponding to the number of input photons is output to the computer 20. Alternatively, the A / D converter 15 can also be provided in each pixel 11.

[0043] When reading out the voltage amplified by amplifier 13, readout noise is generated within amplifier 13 as random noise. Figure 2 This is a graph representing the probability distribution of electrons, with the horizontal axis representing the number of electrons and the vertical axis representing the probability density. For example... Figure 2 As shown, the number of electrons generated from the input photon follows a Poisson distribution. Figure 2The diagram shows the probability distribution of electrons for an average input of 2 photons per pixel, with each readout noise value representing a different probability. Examples of readout noise values ​​are 0.12, 0.15, 0.25, 0.35, 0.40, 0.45, and 1.0 [e-rms]. Figure 2 As shown, the lower the readout noise, the more pronounced the peaks of the probability distribution waveform, and the clearer the distinction between the distributions of each electron. On the other hand, when the readout noise increases, the distributions of adjacent electrons overlap, making it difficult to distinguish the distribution of each electron. For example, when the readout noise is below 0.40 [e-rms], the peaks of each electron can be identified. Conversely, when the readout noise is above 0.45 [e-rms], it is difficult to identify the peaks of each electron. In this embodiment, the magnitude of the readout noise that can distinguish electrons is determined based on whether the peaks can be identified. Thus, in the CMOS image sensor 10 of this embodiment, the readout noise is below 0.4 [e-rms]. Alternatively, the magnitude of the readout noise that can distinguish electrons can also be determined by detecting the inflection point through the second derivative of the probability distribution.

[0044] Furthermore, when a threshold is set to distinguish the number of adjacent electrons from each other, the false detection rate of the detected number of electrons varies depending on the readout noise. Figure 3 This is a graph showing the relationship between readout noise and false detection rate when thresholds are set at values ​​between electron numbers such as 0.5e, 1.5e, 2.5e, etc. Furthermore, the false detection rate, the proportion of erroneous electrons detected, is caused by the spread of the electron probability distribution. Figure 3 As shown, to achieve a false detection rate of less than 1%, the readout noise needs to be less than 0.2 [e-rms]. Furthermore, to achieve a false detection rate of less than 0.1%, the readout noise needs to be less than 0.15 [e-rms].

[0045] Figure 4 (a) is a graph showing the relationship between the number of electrons and the probability density. Figure 4 (b) indicates that it is based on Figure 4 (a) A graph showing the results of the simulation experiment. Figure 4 In (a), the probability distribution of electrons with an average input of 2 photons per pixel is shown, given a readout noise of 0.15 [e-rms]. Furthermore, in Figure 4 In (b), the distribution of the number of electrons for each measurement is shown using analog values. In the A / D converter 15, the... Figure 4 The analog values ​​shown in (b) are converted to digital values ​​and output. The digital values ​​output from each pixel 11 are represented using the following formula.

[0046] Digital value [DN] = Gain [DN / e] × Number of electrons [e] + Offset value [DN]

[0047] Figure 5 (b) Figure 6 (b) and Figure 7 (b) are respectively the Figure 4 (b) A graph showing the conversion of analog values ​​to digital values. Figure 5 (a) Figure 6 (a) and Figure 7 In (a), all are related to Figure 4 (a) Similarly, it shows the probability distribution of electrons when an average of 2 photons are input to one pixel, with a readout noise of 0.15 [e-rms]. Figures 5-7 In this context, threshold values, such as 0.5e, 1.5e, 2.5e, etc., are used as a reference to distinguish between different electron numbers. In the accompanying diagram, these threshold values ​​are represented by dashed lines. Figure 5 In (b), the gain is 2 [DN / e], and the offset is 100 [DN]. For example... Figure 5 As shown in (b), with a gain of 2 [DN / e], it is difficult to reflect the deviation of the measured values ​​observed through analog values ​​in the graph. In addition, the proportion of the output representing the same value as the threshold becomes higher.

[0048] exist Figure 6 In (b), the gain is 10 [DN / e], and the offset is 100 [DN]. For example... Figure 6 As shown in (b), with a gain of 10 [DN / e], the distribution of digital values ​​approximates the distribution of analog values. On the other hand, because the gain is even, as shown in the attached figure, there are also cases where the digital value corresponds to the threshold. Figure 7 In (b), the gain is 11 [DN / e], and the offset is 100 [DN]. For example... Figure 7 As shown in (b), with a gain of 11 [DN / e], the distribution of digital values ​​is more similar to that of analog values. Furthermore, because the gain is odd, it is possible to suppress the occurrence of digital values ​​corresponding to the threshold. In this way, by increasing the value of the gain, the output digital values ​​can be made to approximate analog values. In this embodiment, the CMOS image sensor 10 may also have a gain of, for example, 10 [DN / e] or more.

[0049] Refer again Figure 1The computer 20 physically comprises storage devices such as RAM (Random Access Memory) and ROM (Read-Only Memory), a processor (arithmetic circuit) such as a CPU, and a communication interface. Examples of such computers 20 include personal computers, cloud servers, smart devices (smartphones, tablet computers, etc.), microcomputers, and FPGAs (field-programmable gate arrays). The computer 20 functions as a storage unit 21, a conversion unit 22, a data processing unit 23, and a control unit 24 by executing programs stored in the storage devices via the computer system's CPU. The computer 20 can be configured either inside a camera containing a CMOS image sensor 10 or outside the camera. The computer 20 can be connected to a display device 25 and an input device 26. The display device 25 is, for example, a monitor capable of displaying photon counting results obtained by the computer 20. The input device 26 is a keyboard, mouse, etc., for users to input measurement conditions. Alternatively, a shared touchscreen can be used for both the display device 25 and the input device 26.

[0050] Storage unit 21 stores a table (reference data) for converting digital values ​​output from CMOS image sensor 10 into photon counts. The table is, for example, a lookup table. The table is created based on the gain and offset values ​​of each of the multiple pixels 11. In this embodiment, threshold data corresponding to each pixel 11 is stored as a table.

[0051] The conversion unit 22, referring to a table stored in the storage unit 21, converts the digital value of each of the multiple pixels 11 output from the A / D converter 15 into a photon count. The data processing unit 23 creates a two-dimensional image representing the photon count of each pixel 11 based on the photon count output from the conversion unit 22. Furthermore, the data processing unit 23 can also create a histogram or similar layout as a plot of the number of pixels relative to the photon count. The created two-dimensional image or similar can be output to the display device 25. The control unit 24 can uniformly control the various functions of the computer 20 or the CMOS image sensor 10.

[0052] Next, details of the photon counting device 1 will be described, focusing on the processing of the conversion unit 22. For the sake of simplicity, the CMOS image sensor 10 of the photon counting device will be described as a CMOS image sensor having pixels 11 arranged in 3 rows × 3 columns.

[0053] First, it describes a method for converting digital values ​​into photon numbers, assuming that the gain and offset values ​​are unbiased. Figure 8 The process of converting measured numerical values ​​into electron counts is illustrated schematically. Figure 8In this example, we assume that the offset is 100 [DN] and the gain is 11 [DN / e] in each pixel. Furthermore, we assume the readout noise is 0.15 [e-rms].

[0054] like Figure 8 As shown, in such a CMOS image sensor 10, when photons are input to each pixel 11, charge is stored in each pixel 11 according to the number of photons. In the illustrated example, five electrons are shown stored in all pixels 11. The stored charge is converted into voltage by amplifier 13 and then into a digital value by A / D converter 15. Figure 8 In this process, the digital value of each pixel is displayed within the pixel. Then, the converted digital value is converted into the electron count. In this case, a threshold range is used, for example, to convert the digital value into the electron count. Figure 8 In the example above, it is assumed that the gain and offset values ​​are unbiased. Therefore, when the upper and lower limits of the threshold range are taken as the midpoint of the number of electrons, the threshold range corresponding to 5 electrons is 4.5e or higher and less than 5.5e. When expressed in numerical values, this threshold range is 150–160 [DN]. Referring to this threshold range, [the following will be considered]. Figure 8 In the example shown, when the numerical value is converted to the number of electrons, the numerical value is converted to 5 electrons in all pixels 11. As mentioned above, the number of electrons generated from the input photons follows a Poisson distribution; therefore, the average number of photons can be obtained by dividing the average number of electrons per pixel by the quantum efficiency. At a quantum efficiency of 100%, the number of electrons and the number of photons are the same.

[0055] Next, consider the case where the digital value is converted to the number of electrons using the same threshold range, even when the gain and offset values ​​are biased. Figure 9 These are numerical values ​​representing the deviations in gain and offset. In this example, the average gain is 11 [DN / e], and the gain deviation σ is 10%. That is, the gain ±σ can take values ​​from 9.9 to 12.1. Furthermore, the average offset is 100 [DN], and the offset deviation σ is 3%. That is, the offset ±σ can take values ​​from 97 to 103. Figure 9 Examples also related to Figure 8 This model also stores 5 electrons in all pixels. When using... Figure 8 For example, at the same threshold, there are 4 electrons in the case of 139–149, 5 electrons in the case of 150–160, and 6 electrons in the case of 161–171. Thus, when there are deviations in the gain and offset values, it can be difficult to convert the digital value into the correct number of electrons.

[0056] Therefore, in this embodiment, the conversion unit 22, referring to a table stored in the storage unit 21, converts the digital value of each of the plurality of pixels output from the A / D converter 15 into the number of electrons. The table has threshold data created based on the respective offset values ​​and gains of the plurality of pixels 11. Figure 10 This is a schematic diagram illustrating the process of obtaining the offset value. As described above, the digital value is represented using the following formula. Therefore, the offset value is represented as the digital value output in the absence of input light. Thus, in this embodiment, as... Figure 10 As shown, multiple digital values ​​are obtained from multiple dark images acquired by the CMOS image sensor 10 in the absence of input light. An offset value can be obtained by averaging the acquired digital values ​​for each pixel.

[0057] Digital value [DN] = Gain [DN / e] × Number of electrons [e] + Offset value [DN]

[0058] Figure 11 This is a schematic diagram illustrating the process of obtaining gain. When obtaining the gain of each pixel, multiple frames are acquired by the CMOS image sensor 10 with sufficient light intensity. Then, the average optical signal value S[DN] and standard deviation N[DN] of the digital values ​​of each pixel are obtained. The gain can be expressed as N. 2 / S, therefore, the gain can be derived from the average optical signal value S and the standard deviation N.

[0059] In this embodiment, the storage unit 21 stores the threshold data for each pixel derived based on the gain and offset values ​​as a table. When the median value of the number of electrons is used as the threshold, the threshold representing the lower limit of the number of electrons and the threshold representing the upper limit are respectively expressed by the following formulas, and the range of these lower and upper thresholds is the threshold range corresponding to the number of electrons.

[0060] Threshold (lower limit) = (Number of electrons - 0.5) × Gain + Offset value

[0061] Threshold (upper limit) = (Number of electrons + 0.5) × Gain + Offset value

[0062] Therefore, for example, the threshold for determining the number of electrons to be 5 can be as follows: Figure 12 Export as shown. Figure 12 This is a graph showing the correspondence between gain and offset values ​​and the threshold. Figure 12The obtained gain and offset values ​​are displayed for each pixel. Furthermore, the threshold values ​​for determining the number of electrons to be 5 are displayed for each pixel. For example, with a gain of 10.9 [DN / e] and an offset value of 97.7 [DN], the lower threshold is 146.8 [DN] and the upper threshold is 157.7 [DN]. In the storage unit 21, the threshold data corresponding to the number of electrons for each pixel is displayed as a table. Figure 13 This diagram schematically illustrates the process of converting the measured digital value of each pixel into the number of electrons. In the conversion unit 22, the correct number of electrons can be derived from the digital values ​​by referring to a table stored in the storage unit 21. For example, in... Figure 9 In the example, with a numerical value of 162[DN], it is determined to have 6 electrons. Figure 13 In the example, it is determined to be 5 electrons. The conversion unit 22 can obtain the average number of photons for each pixel by dividing the average number of electrons by the quantum efficiency.

[0063] Next, the operation of photon counting device 1 will be explained. Figure 14 This is a flowchart illustrating the operation of the photon counting device. In this embodiment, when measurement begins while the photon counting device 1 is operating, firstly, photons incident on the pixel 11 of the CMOS image sensor 10 are converted into electrical charge by the photodiode 12 (step S11). Then, the converted charge is converted into voltage by the amplifier 13 (step S12). This voltage is converted into a digital value by the A / D converter 15 and output to the computer 20 (step S13). The conversion unit 22 of the computer 20 compares the digital value with a threshold set for each pixel 11 (step S14), and converts the digital value into a photon count based on the comparison result (step S15). Thus, the number of photons input to each pixel can be measured. The measurement result can also be displayed on the display device 25 as image data, for example.

[0064] As explained above, in the photon counting device 1, digital values ​​are converted into photon counts by the conversion unit 22. In the conversion unit 22, the digital values ​​are converted into electron counts using a threshold set for each pixel, with reference to a table stored in the storage unit 21. The table is created taking into account the individual gain and offset values ​​of the multiple pixels 11. Therefore, even if there are deviations in the gain and offset values ​​between pixels 11, the conversion unit 22 can suppress the effects of these deviations and convert the digital values ​​into electron counts. Consequently, a decrease in photon counting accuracy can be suppressed.

[0065] The table contains multiple threshold data corresponding to multiple pixels 11. In this configuration, appropriate threshold data is created for each pixel 11 based on the deviation of the gain and offset values, thus enabling high-precision determination of the number of electrons. Furthermore, for example, digital values ​​can be corrected without considering deviations.

[0066] The readout noise of amplifier 13 can be 0.2 [e-rms] or less. In this case, for example, the false detection rate can be suppressed to 1% or less. Furthermore, the readout noise of amplifier 13 can be 0.15 [e-rms] or less. In this case, for example, the false detection rate can be suppressed to 0.1% or less.

[0067] The gain can be 10 [DN / e] or higher. The CMOS image sensor 10 has a high gain, enabling the analog values ​​output from the amplifier 13 to be reproduced digitally with high accuracy.

[0068] [Second Implementation]

[0069] The photon counting device of this embodiment differs from the photon counting device of the first embodiment in the configuration of the storage unit 21 and the conversion unit 22. Hereinafter, the differences from the first embodiment will be mainly explained. Furthermore, in the device configuration, because... Figure 1 The first embodiment shown is the same, so it is omitted.

[0070] The storage unit 21 stores a table (reference data) for converting digital values ​​output from the CMOS image sensor 10 into electron counts. The table is created based on the individual gain and offset values ​​of the plurality of pixels 11. In this embodiment, the storage unit 21 stores the gain and offset values ​​of each pixel 11 as a table. Furthermore, the storage unit stores threshold data for each electron count common to all pixels 11 as a table.

[0071] The conversion unit 22 can also correct the digital value of each of the multiple pixels 11 based on the individual gain and offset values ​​of the multiple pixels 11, in a way that suppresses the influence of the deviation of the gain and offset values ​​among the multiple pixels 11, and convert the corrected digital value into a photon count. In this embodiment, by correcting the digital value of each pixel 11 by the conversion unit 22, the apparent gain and offset values ​​(parameters) of each pixel 11 become the same. The corrected digital value can be derived based on the deviation of the gain and offset values ​​of each pixel from the apparent gain and offset values ​​common to all pixels. In this case, the corrected digital value can be derived using the following correction formula. In addition, the apparent gain and apparent offset values ​​are preset and can also be stored in the storage unit 21 or the conversion unit 22.

[0072] Corrected numerical value = ((numerical value - offset value) / gain) × apparent gain + apparent offset value

[0073] Figure 15 This is a graph showing the correspondence between the measured numerical values ​​and the corrected numerical values. In Figure 15In the example shown, the CMOS image sensor in the first embodiment is modified by the above-described modification. Figure 9 Refer to the example of the digital value measured in (refer to) 10. In this example, the conversion unit 22 corrects the digital value so that the apparent gain in all pixels is 11 [DN / e] and the apparent offset value is 100 [DN]. That is, the corrected digital value is derived from the following correction formula.

[0074] Corrected numerical value = ((numerical value - offset value) / gain) × 11 + 100

[0075] In the conversion unit 22, the electron count is obtained using threshold data common to all pixels for the corrected digital value. For example, the storage unit 21 may also maintain a table containing the threshold range derived from the following formula. The conversion unit 22 can convert the corrected digital value into the electron count by referring to the threshold data maintained in the table. Furthermore, in Figure 15 In the example, the apparent gain is 11 [DN / e], and the apparent offset is 100 [DN]. Therefore, when the corrected digital value is 150 to 160, it is determined to be 5 electrons. The conversion unit 22 can obtain the average number of photons by dividing the average number of electrons per pixel by the quantum efficiency.

[0076] Threshold (lower limit) = (Number of electrons - 0.5) × Apparent gain + Apparent offset

[0077] Threshold (upper limit) = (Number of electrons + 0.5) × Apparent gain + Apparent offset

[0078] Figure 16 This is a flowchart illustrating the operation of the photon counting device. In this embodiment, when measurement begins while the photon counting device 1 is operating, firstly, light incident on the pixels of the CMOS image sensor 10 is converted into electrical charge by the photodiode 12 (step S21). Then, the converted charge is converted into voltage by the amplifier 13 (step S22). This voltage is converted into a digital value by the A / D converter 15 and output to the computer 20 (step S23). The digital value is corrected by the conversion unit 22 of the computer 20 for each pixel (step S24). The corrected digital value is compared with a set threshold data (step S25), and based on the comparison result, the corrected digital value is converted into a photon count (step S26). Thus, the number of photons input to each pixel can be measured.

[0079] In this embodiment, the digital value is corrected in a way that suppresses the effect of the deviation between the gain and offset values. Therefore, as described above, the corrected digital value can be converted into the number of electrons using the same threshold data for all pixels.

[0080] The embodiments have been described in detail above with reference to the accompanying drawings, but the specific configuration is not limited to these embodiments.

[0081] For example, in the conversion unit, the number of electrons per pixel can be calculated by dividing the value obtained by subtracting the offset value from the measured digital value by the gain, as shown in the following formula. In this case, a common threshold range corresponding to all pixels can be used for the calculated number of electrons. For example, the upper and lower limits of the threshold range can be used as the middle value of the number of electrons to set a common threshold range, as described above. In this case, the threshold range corresponding to 5 electrons is 4.5e or higher and less than 5.5e.

[0082] Number of electrons = ((digital value - offset value) / gain)

[0083] Furthermore, in the CMOS image sensor 10 of this embodiment, an example is shown where the readout noise of each pixel is 0.4 [e-rms] or less. However, even in a sensor configuration where the readout noise is 0.4 [e-rms], there are cases where the noise of some pixels is greater than 0.4 [e-rms]. In such cases, the pixels with readout noise of 0.4 [e-rms] or less can be identified in advance through measurement or the like, and photon counting can be performed using only the pixels with readout noise of 0.4 [e-rms] or less.

[0084] Symbol Explanation

[0085] 1… Photon counting device, 11… Pixel, 12… Photodiode (photoelectric conversion element), 13… Amplifier, 15… A / D converter, 21… Storage unit, 22… Conversion unit.

Claims

1. A photon counting device, characterized in that, include: Multiple pixels, which constitute an image sensor, and include a photoelectric conversion element that converts input light into electrical charge and an amplifier that amplifies the charge converted by the photoelectric conversion element and converts it into voltage; An A / D converter that converts the voltage output from the amplifier of the plurality of pixels into a digital value; and The conversion unit, for each of the plurality of pixels, corrects the digital value by adjusting the deviations in the gain and offset values ​​of each of the amplifiers constituting the plurality of pixels, and converts the corrected digital value into a photon count based on corresponding threshold data for distinguishing photon counts. The threshold data is data representing a threshold range, which includes an upper limit and a lower limit of the corrected digital value set relative to each photon number.

2. The photon counting device as described in claim 1, characterized in that, Also includes: The data processing unit generates a two-dimensional image representing the photon count of each of the plurality of pixels based on the photon count converted by the conversion unit.

3. The photon counting device as described in claim 1, characterized in that, Also includes: The data processing unit generates a histogram of the layout of the number of pixels relative to the number of photons, based on the number of photons converted by the conversion unit.

4. The photon counting device as described in claim 2, characterized in that, Also includes: The data processing unit generates a histogram of the layout of the number of pixels relative to the number of photons, based on the number of photons converted by the conversion unit.

5. The photon counting device according to any one of claims 1 to 4, characterized in that, The A / D converter is configured for each of the plurality of pixels.

6. The photon counting device according to any one of claims 1 to 4, characterized in that, The amplifier has a readout noise of less than 0.2 e-rms.

7. The photon counting device according to any one of claims 1 to 4, characterized in that, The amplifier has a readout noise of less than 0.15 e-rms.

8. The photon counting device according to any one of claims 1 to 4, characterized in that, The gain is 10 [DN / e] or higher.

9. The photon counting device according to any one of claims 1 to 4, characterized in that, The number of photons is the number of photons that takes into account the quantum efficiency of the image sensor.

10. The photon counting device according to any one of claims 1 to 4, characterized in that, The number of photons refers to the number of photoelectrons.

11. A photon counting method, characterized in that, include: The step of converting light into electrical charge by the photoelectric conversion elements of each of the multiple pixels that make up an image sensor; The step of amplifying the converted charge by an amplifier constituting the plurality of pixels and converting it into voltage; The step of converting the voltage output from each of the amplifiers into a digital value using an A / D converter and then outputting it; and The steps involve, for each of the plurality of pixels, correcting the digital value converted by the A / D converter by correcting the deviations in the gain and offset values ​​of each of the amplifiers constituting the plurality of pixels, and converting the corrected digital value into a photon count based on corresponding threshold data for distinguishing photon counts. The threshold data is data representing a threshold range, which includes an upper limit and a lower limit of the corrected digital value set relative to each photon number.

12. The photon counting method as described in claim 11, characterized in that, Also includes: The step of generating a two-dimensional image representing the photon count of each of the plurality of pixels based on the converted photon count.

13. The photon counting method as described in claim 11, characterized in that, It also includes the step of generating a histogram of the layout of the number of pixels relative to the number of photons, based on the converted number of photons.

14. The photon counting method as described in claim 12, characterized in that, It also includes the step of generating a histogram of the layout of the number of pixels relative to the number of photons, based on the converted number of photons.

15. The photon counting method according to any one of claims 11 to 14, characterized in that, The amplifier has a readout noise of less than 0.2 e-rms.

16. The photon counting method according to any one of claims 11 to 14, characterized in that, The amplifier has a readout noise of less than 0.15 e-rms.

17. The photon counting method according to any one of claims 11 to 14, characterized in that, The gain is 10 [DN / e] or higher.

18. The photon counting method according to any one of claims 11 to 14, characterized in that, The number of photons is the number of photons that takes into account the quantum efficiency of the image sensor.

19. The photon counting method according to any one of claims 11 to 14, characterized in that, The number of photons refers to the number of photoelectrons.

20. The photon counting method according to any one of claims 11 to 14, characterized in that, The A / D converter is configured for each of the plurality of pixels.

Images