JP2009171128A - Solid state imaging element and solid imaging apparatus using the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To accurately perform correction to remove a DC noise component even if a plurality of horizontal output lines are arranged in order to attain a parallel output. <P>SOLUTION: In a solid state imaging element, respective pixels are divided into four groups, i.e., the first to fourth groups and image signals outputted from the respective pixels are outputted in parallel by the use of the four horizontal output lines L21-L24 corresponding to the respective groups. The vertical output lines L11 of the first to fourth groups in an OB area are connected to the whole horizontal output lines L21-L24 via output circuits 30, respectively. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a solid-state imaging device in which a plurality of pixels are arranged in a matrix and a solid-state imaging device using the same.

  In recent years, a solid-state imaging device in which an optical black pixel (hereinafter referred to as an “OB pixel”) is provided around an effective pixel is known. The purpose is to obtain a correction value by actually measuring DC noise components such as dark current that changes with the element temperature in the OB pixel, and using this correction value, a pixel that is output from an effective pixel (hereinafter referred to as “UK pixel”). The purpose is to correct the signal and remove the DC noise component contained in the pixel signal.

  Since the pixel signal output from the OB pixel includes a random noise component, it is preferable to increase the number of OB pixels in order to calculate a highly accurate correction value that is less affected by the random noise component.

  FIG. 5 is a diagram showing a conventional solid-state imaging device. The solid-state imaging device illustrated in FIG. 5 is a solid-state imaging device having one horizontal output line L130, and includes an OB region, a UK region, a vertical output line L110, a horizontal output line L130, and an amplifier 200.

  The OB area is composed of a plurality of OB pixels arranged in a matrix of N (N is a positive integer) rows × 8 columns. The UK region is composed of a plurality of UK pixels arranged in a matrix with N rows × 8 columns.

  In the OB region, eight vertical output lines L110 exist corresponding to each column, and are connected to each of the N OB pixels in each column to output a pixel signal output from the OB pixel. In the UK region, eight vertical output lines L110 exist corresponding to each column, are connected to each of N UK pixels in each column, and output a pixel signal output from the UK pixel. The horizontal output line L130 is connected to each of the 16 vertical output lines L110 via an output circuit (not shown), and outputs a pixel signal to the amplifier 200.

  In the solid-state imaging device shown in FIG. 5, when each row is sequentially selected by a vertical scanning circuit (not shown), the eight OB pixels in each selected row output a pixel signal via the vertical output line L110. This pixel signal is held in the output circuit of each column. Next, when each column is sequentially selected by a horizontal scanning circuit (not shown), the output circuit of the selected column outputs the held pixel signal to the horizontal output line L130 and inputs it to the amplifier 200. The pixel signals of each column input to the amplifier 200 are taken into an image processing unit connected to the subsequent stage of the amplifier 200 and averaged to calculate a correction value.

  When calculation of the correction value is completed, pixel signals are sequentially output from the UK pixels in each column to the amplifier 200 via the vertical output line L110, an unillustrated output circuit, and the horizontal output line L130 as in the case of the OB pixel. Then, it is taken into the image processing unit and corrected using the correction value.

  In the solid-state imaging device shown in FIG. 5, since there is only one horizontal output line L130, pixel signals of each column cannot be output in parallel, and there is a problem that imaging time becomes long.

  Therefore, in recent years, for the purpose of shortening the imaging time, each pixel is divided into a plurality of groups, and a pixel signal output from each pixel is output in parallel using a horizontal output line corresponding to each group. An image sensor is known. FIG. 6 shows a solid-state imaging device that outputs pixel signals in parallel using four horizontal output lines L131 to L134.

  In the solid-state imaging device shown in FIG. 6, the OB pixel and the UK pixel are divided into four groups of first to fourth. Specifically, the OB pixels in the first and fifth columns from the left in the OB region and the UK pixels in the first and fifth columns from the left in the UK region belong to the first group. The OB pixels in the second and sixth columns from the left in the OB region and the UK pixels in the second and sixth columns from the left in the UK region belong to the second group. The OB pixels in the third and seventh columns from the left in the OB region and the UK pixels in the third and seventh columns from the left in the UK region belong to the third group. The OB pixels in the fourth and eighth columns from the left in the OB region and the UK pixels in the fourth and eighth columns from the left in the UK region belong to the fourth group.

  The horizontal output lines L131 to L134 correspond to the first to fourth groups, respectively, and the vertical output lines L110 of the first to fourth groups are connected to the horizontal output lines L131 to L134, respectively.

In the solid-state imaging device shown in FIG. 6, a total of four columns are selected from the first to fourth groups simultaneously by a horizontal scanning circuit (not shown), and held by the selected four columns of output circuits. Pixel signals are output in parallel through horizontal output lines L131 to L134. By providing the four horizontal output lines L131 to L134 in this way, the readout time for all the pixels can be shortened to ¼ compared with the case of reading with one horizontal output line.
JP 2000-12819 A

  However, in the solid-state imaging device shown in FIG. 6, since only two vertical output lines L110 are connected to each of the horizontal output lines L131 to L134 in the OB region, the correction values of the horizontal output lines L131 to L134 are , Respectively, using the pixel signals of two OB pixels.

  On the other hand, in the solid-state imaging device shown in FIG. 5, since the eight vertical output lines L110 are connected to the horizontal output line L130 in the OB region, the correction value is calculated using pixel signals of eight OB pixels. Will be.

  Therefore, the correction value calculated by the solid-state imaging device shown in FIG. 6 is greatly affected by the random noise component as compared with the correction value of the solid-state imaging device shown in FIG. There is a problem that the pixel signal cannot be corrected with high accuracy.

  An object of the present invention is to provide a solid-state imaging device and a solid-state imaging device capable of accurately performing correction to remove a DC noise component even when a plurality of horizontal output lines are provided in order to realize parallel output. It is to be.

  (1) A solid-state imaging device according to the present invention includes a plurality of pixels arranged in a matrix, and the pixels are divided into m (m is an integer of 2 or more) groups for each of one or a plurality of columns. The pixel portion in which the pixels in the column are optical black pixels and the pixels in the remaining columns are effective pixels, and the vertical output corresponding to each column that outputs the optical black pixels and the pixel signals output from the effective pixels A line and m horizontal output lines corresponding to each group to which the pixel signal is input via the vertical output line and output the input pixel signal, and the vertical output line of the effective pixel corresponds to The vertical output line of the optical black pixel is connected to the horizontal output line of the corresponding group, and is connected to the horizontal output line of any one or a plurality of groups other than the corresponding group. Connected It is characterized in.

  According to this configuration, the vertical output line of the optical black pixel is connected to the horizontal output line of the corresponding group, and is connected to the horizontal output line of any one or a plurality of groups other than the corresponding group. Therefore, the number of optical black pixels employed when calculating the correction value for each group is increased compared to the conventional solid-state imaging device shown in FIG. 6, and the correction values for each group can be calculated with high accuracy. Even when a plurality of horizontal output lines are provided in order to realize parallel output, correction for removing DC noise components can be performed with high accuracy.

  (2) It is preferable that the vertical output lines of the optical black pixels are connected to the horizontal output lines of the corresponding group and are connected to the horizontal output lines of all the groups other than the corresponding group.

  According to this configuration, the number of optical black pixels employed when calculating the correction value for each group is the same as that of the solid-state imaging device shown in FIG. 5, and a more accurate correction value can be obtained while realizing parallel output. Obtainable.

  (3) In the pixel portion, it is preferable that each pixel is arranged in one row × multiple columns. According to this configuration, accurate correction can be realized while realizing parallel output in a solid-state imaging device in which pixels are arranged one-dimensionally.

  (4) In the pixel portion, each pixel is preferably arranged in a plurality of rows and a plurality of columns. According to this configuration, accurate correction can be realized while realizing parallel output in a solid-state imaging device in which each pixel is two-dimensionally arranged.

  (5) A solid-state imaging device according to the present invention is a pixel output from an effective pixel based on the solid-state imaging device according to any one of claims 1 to 4 and a pixel signal output from an optical black pixel of each group. A correction value calculation unit for obtaining a correction value of each group for removing a DC noise component included in the signal, and a group corresponding to the correction value by using the correction value of each group calculated by the correction value calculation unit And a correction unit that corrects a pixel signal output from the effective pixel.

  According to this configuration, the correction value of each group for removing the DC noise component is obtained based on the pixel signal output from the optical black pixel of each group, and is output from the effective pixel using this correction value. Therefore, the direct current noise component included in the pixel signal output from the effective pixel can be accurately removed.

  (6) The correction value calculation unit calculates an average value of pixel signals output from the optical black pixels of each group for each row as a correction value, and the correction unit subtracts the corresponding correction value from the pixel signal. It is preferable to correct the pixel signal with

  According to this configuration, the correction value of each group is calculated for each row, and the correction value is subtracted from the pixel signal to correct the pixel signal. Therefore, the DC noise component is removed from the pixel signal, and the horizontal output line It is possible to remove variations in output characteristics for each pixel signal.

  According to the present invention, even in the case of a solid-state imaging device provided with a plurality of horizontal output lines in order to realize parallel output, it is possible to accurately perform correction for removing a DC noise component.

  Hereinafter, a solid-state imaging device according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is an overall configuration diagram of a solid-state imaging device according to an embodiment of the present invention. As shown in FIG. 1, the solid-state imaging device includes a pixel unit 10, a vertical scanning circuit 20, a vertical output line L11, horizontal output lines L21 to L24, an output circuit (OC: output circuit) 30, a pair of upper and lower horizontal scanning circuits 40, 40, an amplifier 50, and an analog / digital converter (A / D) 60, an image processing unit 70, a timing generator (TG) 80, and a control unit 90. The solid-state imaging device shown in FIG. 1 divides each pixel into a plurality of groups, for example, first to fourth groups, and uses four horizontal output lines L21 to L24 corresponding to each group. The pixel signals output from the pixels are output in parallel.

  In the present embodiment, the solid-state imaging device is constituted by a CMOS image sensor made into one chip, but may be constituted by a CCD image sensor instead of the CMOS image sensor. The image processing unit 70 and the control unit 90 are generally not included in the solid-state image sensor, but are not limited thereto, and may be included in the solid-state image sensor.

  The pixel unit 10 includes an OB area composed of optical black pixels (hereinafter, optical black pixels are referred to as “OB pixels”), adjacent to the OB area in the column direction, and effective pixels (hereinafter referred to as “UK pixels”). UK region consisting of “pixels”).

  The OB area is composed of a plurality of OB pixels arranged in a matrix of N (N is a positive integer) rows × 8 columns. The UK region is composed of a plurality of UK pixels arranged in a matrix with N rows × M columns (M is a positive integer of 4 or more).

  In the above description, the number of columns in the OB area is 8 and the number of columns in the UK area is M. However, since each pixel is divided into the first to fourth groups, the OB area and the UK area The number of columns may be four or more. In addition, when the number of groups is 3, the number of columns in the OB area and the UK area is 3 or more, and when the number of groups is 5, the number of columns in the OB area and the UK area is 5 or more. As described above, the number of columns in the OB area and the UK area may be appropriately changed according to the number of groups.

  The OB pixel includes a photodiode and an amplification circuit that amplifies the signal charge photoelectrically converted by the photodiode, measures a DC noise component such as a dark current that changes depending on the element temperature, and determines a DC noise component from the pixel signal of the effective pixel. Is used to calculate a correction value for removing. Similar to the OB pixel, the UK pixel includes a photodiode, an amplifier circuit, and the like, and is used to capture an optical image.

  Here, the OB pixels in the columns of the first and fifth rows from the left in the OB region and the UK pixels in the columns of the 4s + 1 (s is an integer of 0 or more) row from the left in the UK region belong to the first group. The OB pixels in the second and sixth columns from the left in the OB area and the UK pixels in the 4s + 2 column from the left in the UK area belong to the second group. In addition, the OB pixels in the third and seventh rows from the left in the OB region and the UK pixels in the 4s + 3 rows from the left in the UK region belong to the third group. Further, the OB pixels in the columns of the fourth and eighth rows from the left in the OB region and the UK pixels in the columns of the 4s + 4th row from the left in the UK region belong to the fourth group.

  There are eight vertical output lines L11 in the OB region corresponding to each column in the OB region toward the upper horizontal scanning circuit 40 and eight toward the lower horizontal scanning circuit 40, and N lines in each column. Connected to each of the OB pixels, and outputs a pixel signal output from the OB pixel. There are M vertical output lines L11 in the UK region corresponding to each column of the UK region, and each of the vertical output lines L11 is connected to each of N UK pixels in each column and outputs a pixel signal output from the UK pixel.

  In the UK region, the first group of vertical output lines L11 is connected to the horizontal output line L21 via the output circuit 30, and the second group of vertical output lines L11 is output horizontally via the output circuit 30. The third group of vertical output lines L11 connected to the line L22 is connected to the horizontal output line L23 via the output circuit 30, and the fourth group of vertical output lines L11 is connected to the horizontal output line 30 via the output circuit 30. It is connected to the line L24. Here, in FIG. 1, since the horizontal output lines L21 and L23 are arranged on the upper side of the pixel unit 10, the odd-numbered vertical output lines L11 in the UK region are connected to the upper horizontal scanning circuit 40. Further, since the horizontal output lines L22 and L24 are arranged on the opposite side of the horizontal output lines L21 and L23 across the pixel unit 10, the even-numbered vertical output lines L11 are connected to the lower horizontal scanning circuit 40. Yes.

  On the other hand, in the OB region, the vertical output lines L11 of the first to fourth groups are connected to all the horizontal output lines L21 to L24 via the output circuit 30, respectively. Specifically, the upper vertical output line L11 of the first and second groups is connected to the horizontal output lines L21 and L23, and the lower vertical output line L11 is connected to the horizontal output lines L22 and L24. Has been. The upper vertical output line L11 of the third and fourth groups is connected to the horizontal output lines L21 and L23, respectively, and the lower vertical output line L11 of the third and fourth groups is connected to the horizontal output lines L22, L22. L24 is connected to each. This makes it possible to calculate correction values for each group using pixel signals from eight OB pixels, and to obtain correction values with higher accuracy than the conventional solid-state imaging device shown in FIG. .

  The vertical scanning circuit 20 includes, for example, a shift register, and is connected to OB pixels and UK pixels in each row via N row selection lines L51 corresponding to each row, and sequentially selects each row in accordance with the control of the TG 80. To do.

  There are 8 × 2 = 16 output circuits 30 corresponding to the vertical output lines L11 above and below each column in the OB region, and M corresponding to the vertical output lines L11 in each column in the UK region. The pixel signals output from the OB pixels or UK pixels are held, and the held pixel signals are output to the amplifier 50 via the horizontal output lines L21 to L24 under the control of the horizontal scanning circuit 40. Here, in the UK region, the odd-numbered output circuits 30 are arranged above the pixel unit 10, and the even-numbered output circuits 30 are arranged below the pixel unit 10. Therefore, it can be seen that the wiring density of the output circuit 30 in the UK region is smaller than that in the OB region, and there is a margin in the wiring space.

  FIG. 2 shows a circuit diagram of the output circuit 30. As shown in FIG. 2, the output circuit 30 includes a constant current circuit 31, a signal sample hold circuit 32, and a noise sample hold circuit 33. The constant current circuit 31 is composed of a field effect transistor, and a load voltage VD is applied to the gate to function as a load.

  The signal sample hold circuit 32 includes a switch S1, a capacitor C1, and an amplifier A1, and the switch S1 is turned on / off under the control of the TG 80, and the pixel signal output from the vertical output line L11 is held by the capacitor C1. The noise sample and hold circuit 33 includes a switch S2, a capacitor C2, and an amplifier A2. The switch S2 is turned on / off according to the control of the TG 80, and the noise signal output from the vertical output line L11 is held by the capacitor C2.

  The signal sample hold circuit 32, when selected by the horizontal scanning circuit 40, outputs the pixel signal held by the capacitor C1 to the amplifier 50 via the amplifier A1 and one line L11-1 of the pair of vertical output lines L11. To do. Here, the signal sample hold circuit 32 is selected by the selection signal input to the amplifier A1 by the horizontal scanning circuit 40 via one line L41-1 of the pair of column selection lines L41.

  When selected by the horizontal scanning circuit 40, the noise sample hold circuit 33 outputs the noise signal held by the capacitor C2 to the amplifier 50 via the line L11-2. Here, the noise sample hold circuit 33 is selected when the horizontal scanning circuit 40 inputs a selection signal to the amplifier A2 via the line L41-2. The amplifier 50 subtracts the noise signal output via the line L11-2 from the pixel signal output via the line L11-1 and outputs the result to the A / D 60.

  Returning to FIG. 1, the horizontal output lines L21 and L23 are each connected to eight vertical output lines L11 in the OB region, and the horizontal output lines L22 and L24 are respectively connected to eight vertical output lines L11 in the OB region. The pixel signals held in the output circuit 30 in the column selected by the horizontal scanning circuit 40 are output in parallel to the amplifier 50. In the UK region, the horizontal output line L21 is connected to the first group of vertical output lines L11, the horizontal output line L22 is connected to the second group of vertical output lines L11, and the horizontal output line L23 is connected to the third group. The vertical output line L11 of the group is connected, and the horizontal output line L24 is connected to the vertical output line L11 of the fourth group, and outputs the pixel signals of the first to fourth groups in parallel.

  Here, the horizontal output lines L21 and L23 are arranged above the pixel unit 10 in parallel with the column direction, and the horizontal output lines L22 and L24 are arranged below the pixel unit 10 and in parallel with the column direction. However, the present invention is not limited to this, and any two of the horizontal output lines L21 to L24 may be arranged on the upper side of the pixel unit 10 and the remaining two may be arranged on the lower side of the pixel unit 10. In this case, the wiring patterns of the first to fourth groups of vertical output lines L11 in the UK region may be appropriately changed according to the arrangement pattern of the horizontal output lines L21 to L24. Further, all of the horizontal output lines L21 to L24 may be arranged on the upper side or the lower side of the pixel unit 10. In this case, the number of vertical output lines L11 in the OB region can be set to 8, and the number of horizontal scanning circuits 40 can be set to 1.

  The horizontal scanning circuit 40 is constituted by a shift register, for example, and there is a pair of upper and lower sides. The upper horizontal scanning circuit 40 is connected to the eight vertical output lines L11 corresponding to each column in the OB area via the output circuit 30, and outputs to the vertical output lines L11 of the first and third groups in the UK area. Connected via the circuit 30, a column selection signal is output to each output circuit 30 according to the control of the TG 80, and each column is selected according to a predetermined sequence. The lower horizontal scanning circuit 40 is connected to the eight vertical output lines L11 corresponding to each column in the OB region via the output circuit 30, and the second and fourth group vertical output lines in the UK region. L11 is connected to the output circuit 30 and outputs a column selection signal to each output circuit 30 according to the control of the TG 80, and selects each column according to a predetermined sequence.

  Specifically, when the horizontal scanning circuit 40 reads pixel signals from the OB area, the upper and lower horizontal scanning circuits 40 sequentially select the output circuit 30 in the OB area, for example, in the column direction, and the pixels in the selected column The signal is output in parallel to the horizontal output lines L21 to L24. On the other hand, when the pixel signal is read from the UK area, the upper horizontal scanning circuit 40 simultaneously selects the first and third columns, and the lower horizontal scanning circuit 40 simultaneously selects the second and fourth columns. The first horizontal scanning circuit 40 simultaneously selects the fifth and seventh columns of the UK region, and the lower horizontal scanning circuit simultaneously selects the sixth and eighth columns of the UK region. The group 4 pixel signals are output in parallel to the horizontal output lines L21 to L24.

  There are four amplifiers 50 corresponding to the horizontal output lines L21 to L24. The amplifier 50 removes noise components from the pixel signals output from the horizontal output lines L21 to L24, amplifies them with a predetermined gain, and performs A / D60. Output to.

  There are four A / Ds 60 corresponding to the horizontal output lines L21 to L24, respectively, and converts the pixel signals output from the amplifier 50 into digital pixel signals. The image processing unit 70 is configured by a dedicated hardware circuit, and receives digital pixel signals output from the four A / Ds 60. The image processing unit 70 includes a correction value calculation unit 71 and a correction unit 72.

  The correction value calculation unit 71 obtains a correction value for each group for removing a DC noise component included in the pixel signal output from the UK pixel, based on the pixel signal output from the OB pixel of each group. Here, it is preferable to calculate a correction value for the average value of the pixel signals output from the OB pixels of each group for each row.

Specifically, the correction value calculation unit 71 takes in pixel signals from eight OB pixels belonging to the row selected by the vertical scanning circuit 20 via the four horizontal output lines L21 to L24, and outputs the horizontal output line. The average value of the eight pixel signals output by each of L21 to L24 is calculated as the correction values of the first to fourth groups. For example, an average of pixel signals output from eight OB pixels in the i-th (i = 1, 2,..., N) row through the horizontal output line L2k (k = 1, 2, 3, 4). When the value is A _ik , A _ik is calculated as the correction value H _ik for the k-th group in the i-th row.

  Therefore, the first to fourth correction values are each composed of N correction values corresponding to each row. That is, the correction value is composed of a total of 4 × N correction values. The calculated correction value is temporarily stored in the memory.

The correction unit 72 corrects the pixel signal output from the UK pixel of the group corresponding to the correction value, using the correction value calculated by the correction value calculation unit 71. Here, for example, for the pixel signal from the first group of UK pixels in the first row, the correction unit 72 subtracts H ₁₁ from the pixel signal and applies the pixel signal from the first group of UK pixels in the second row. Corrects the pixel signal by subtracting the corresponding correction value from the pixel signal of the UK pixel, such as subtracting H ₂₁ from the pixel signal.

  The TG 80 outputs control signals to the vertical scanning circuit 20, the output circuit 30, the horizontal scanning circuit 40, and the like under the control of the control unit 90, and controls these circuits. The control unit 90 includes, for example, a CPU, a ROM, a RAM, and the like, and governs overall control of the solid-state imaging device.

The solid-state imaging device configured as described above operates as follows. First, the first row is selected by the vertical scanning circuit 20, and pixel signals are simultaneously output from the eight OB pixels in the first row and held in the output circuit 30. Next, each column in the OB area is sequentially selected in the column direction by the upper and lower horizontal scanning circuits 40, and the pixel signal held by the output circuit 30 in each selected column is parallel to the horizontal output lines L21 to L24. Output. The pixel signals output in parallel are input to the image processing unit 70 via the amplifier 50 and the A / D 60. When all of the eight pixel signals in the row selected by the vertical scanning circuit 20 are input to the image processing unit 70 via the horizontal output lines L21 to L24, the correction value calculation unit 71 outputs the horizontal output line L21. the average value _a 11 _{to a 14} of the input pixel signal via the ~L24 calculated as the correction value _H 11 _{to H 14,} is held in the memory.

  Next, pixel signals are simultaneously output from the M UK pixels in the first row and held in the output circuit 30. Next, the first to fourth columns are simultaneously selected by the horizontal scanning circuit 40, and the selected output circuit 30 outputs the held pixel signals to the corresponding horizontal output lines L21 to L24 in parallel. The pixel signals output in parallel are input to the image processing unit 70 via the amplifier 50 and the A / D 60.

Then, the pixel signal of the fourth column, respectively, is corrected using the correction value _H 11 _{to H 14.} Then, the selected 5-8 row by the horizontal scanning circuit 40 at the same time, the pixel signal from the output circuit 30 of the 5-8 line is parallel output to the image processing unit 70, using the respective correction values _H 11 _{to H 14} Corrected. In this manner, the pixel signals of all the UK pixels in the first row of the UK region is read, the second line by the vertical scanning circuit 20 is selected, in the same manner as the first row correction value H ₂₁ to H ₂₄ After the calculation, the pixel signals output from the UK pixels in the second row are corrected using the correction values H _{21 to} H ₂₄ . The above operation is repeatedly executed up to the Nth row, and pixel signals output from all UK pixels are corrected.

  Next, effects of the present solid-state imaging device will be described. In the conventional solid-state imaging device shown in FIG. 6, the number of OB pixel columns connected to one horizontal output line is two even though the number of OB pixel columns is eight. Usually, the pixel signal output from the OB pixel is averaged by each horizontal output line, and the correction value is calculated. In this case, if the direct current component (including the low frequency component) of the pixel signal Vobi of the OB pixel in the i-th column in a certain horizontal output line is vdi and the random noise component is vri, the OB pixel connected to this horizontal output line The average value Vod_av of the pixel signal is expressed as follows.

Vod_av = ((vd1 + vr1) + (Vd2 + vr2)) / 2
= (Vd1 + vd2) / 2 + (vr1 + vr2) / 2 (1)
Here, assuming that the amplitude of the random noise component vr is Δvr, the amplitude of (vr1 + vr2) / 2 in (1) is √Δvr because vr is random noise.

  When the horizontal output line shown in FIG. 5 is a single solid-state imaging device, the average value Vod_av of the OB pixels of a certain horizontal output line is expressed as follows.

vod_av = ((vd1 + vr1) + (vd2 + vr2) +... + (vd8 + vr8)) / 8
= (Vd1 + vd2 + ... + vd8) / 8 + (vr1 + vr2 + ... + vr8) / 8 (2)
Therefore, the amplitude of the random noise component vr is 8√Δvr. That is, when a pixel signal output from an OB pixel is output in parallel by four horizontal output lines and averaged to obtain a correction value, the pixel signal output from the OB pixel is averaged by one horizontal output line. As a result, the influence of the random noise component vr is stronger than when the correction value is obtained.

  That is, in the conventional solid-state imaging device that outputs in parallel as shown in FIG. 6, the random noise component vr included in the correction value calculated for each horizontal output line is larger than that in the solid-state imaging device shown in FIG. An accurate correction value cannot be obtained.

  Each horizontal output line is provided with an amplifier at the final stage, and the output characteristics differ for each horizontal output line due to variations in characteristics. At the time of shooting, a correction value for removing the variation in output characteristics is calculated based on the result of comparing the average value of the output of the OB pixels in each horizontal output line. However, as described above, the OB pixels connected to the respective horizontal output lines are different, and the random noise component vr is also large. Therefore, even if the variation in the output characteristics between the horizontal output lines is corrected, the accuracy is improved. High correction cannot be realized.

  In the present solid-state imaging device, as shown in FIG. 1, each OB pixel is connected to all the horizontal output lines L21 to L24. Therefore, the average value of the pixel signal of the OB pixel in each of the horizontal output lines L21 to L24 is expressed by the following formula.

vod_av = ((vd1 + vr1) + (vd2 + vr2) +... + (vd8 + vr8)) / 8
= (Vd1 + vd2 + ... vd8) / 8 + (r1 + vr2 + ... + vr8) / 8 (3)
This is the same expression as the average output value of the OB pixels in the solid-state imaging device shown in FIG. 6, and it can be seen that the amplitude of the random noise component vr is suppressed to 8√Δvr. That is, in the UK pixel, the parallel output by the four horizontal output lines L21 to L24 reduces the readout time, and the random noise component vr in the pixel signal output from the OB pixel also has one horizontal output line. Therefore, it is possible to obtain a highly accurate correction value, and to accurately remove DC component noise from the pixel signal output from the UK pixel by using the output of the OB pixel. It becomes possible. In addition, since the average value of the pixel signals output from the OB pixels of each group for each row is adopted as the correction value, it is possible to correct the variation between horizontal output lines and the variation of the amplifier 50 with high accuracy. .

  In FIG. 1, the eight vertical output lines L11 in the OB area are connected to all the horizontal output lines. However, the present invention is not limited to this, and connected to the horizontal output lines of the corresponding group and other than the corresponding group. A group may be connected to one or two groups of horizontal output lines. FIG. 3 is a diagram showing another connection example of the vertical output line L11 and the horizontal output lines L21 to L24 in the OB region.

  In the connection example of FIG. 3, the vertical output lines L11 of the first and third groups are connected to the horizontal output lines L21 and L23, and the vertical output lines L11 of the second and fourth groups are connected to the horizontal output lines L22 and L24. It is connected. This eliminates the need for the even column vertical output lines L11 above the pixel portion 10 in the OB region and eliminates the odd column vertical signal lines L11 below the pixel portion 10 in the OB region. Therefore, the wiring space of the output circuit 30 on the upper side and the lower side of the pixel unit 10 can be increased.

  In FIG. 1, the number of horizontal output lines is four, but the number is not limited to this, and other numbers may be adopted. FIG. 4 is a diagram illustrating an example of connection between the vertical output line L11 and the horizontal output lines L21 and L22 of the solid-state imaging device when the number of horizontal output lines is two. As shown in FIG. 4, in the OB region, eight vertical output lines L11 corresponding to each column are arranged on the upper and lower sides across the pixel unit 10, and the upper eight vertical output lines L11 are arranged on the upper side of the pixel unit 10. The eight vertical output lines L11 on the lower side are connected to the horizontal output line L22 arranged on the lower side of the pixel unit 10.

  In this case, the OB pixel and the UK pixel are divided into first and second groups, and the odd-numbered columns are, for example, the first group, the even-numbered columns are, for example, the second group, and the odd-numbered columns are formed in the UK region. The vertical output line L11 is connected to the horizontal output line L21, and the even-numbered vertical output lines L11 are connected to the horizontal output line L22. In the UK region, pixel signals are output in parallel in units of two columns.

1 shows an overall configuration diagram of a solid-state imaging device according to an embodiment of the present invention. FIG. The circuit diagram of the output circuit is shown. It is the figure which showed the other example of a connection of the vertical output line of an OB area | region, and a horizontal output line. It is the figure which showed the example of a connection of the vertical output line and horizontal output line of a solid-state image sensor at the time of using two horizontal output lines. It is the figure which showed the conventional solid-state image sensor. It is the figure which showed the conventional solid-state image sensor.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Pixel part 20 Vertical scanning circuit 30 Output circuit 31 Constant current circuit 32 Signal sample hold circuit 33 Noise sample hold circuit 40 Horizontal scanning circuit 50 Amplifier 70 Image processing part 71 Correction value calculation part 72 Correction part 90 Control part L11 Vertical output line L21 ~ L24 Horizontal output line

Claims (6)

A plurality of pixels arranged in a matrix, wherein the pixels are divided into m (m is an integer of 2 or more) groups for each of a plurality of columns, and pixels in a predetermined column are optical black pixels; A pixel portion in which the pixels in the remaining columns are effective pixels;
A vertical output line corresponding to each column for outputting a pixel signal output from the optical black pixel and the effective pixel;
The pixel signals are input via the vertical output lines, and m horizontal output lines corresponding to each group for outputting the input pixel signals,
The vertical output lines of the effective pixels are connected to the corresponding group of horizontal output lines;
A vertical output line of the optical black pixel is connected to a horizontal output line of a corresponding group, and is connected to a horizontal output line of any one or a plurality of groups other than the corresponding group. Image sensor.   The vertical output line of the optical black pixel is connected to a horizontal output line of a corresponding group and is connected to a horizontal output line of all groups other than the corresponding group. Solid-state image sensor.   3. The solid-state imaging device according to claim 1, wherein each pixel of the pixel unit is arranged in one row × multiple columns.   The solid-state imaging device according to claim 1, wherein each pixel of the pixel unit is arranged in a plurality of rows and a plurality of columns. A solid-state imaging device according to any one of claims 1 to 4,
Based on the pixel signal output from the optical black pixel of each group, a correction value calculation unit for obtaining a correction value of each group for removing the DC noise component included in the pixel signal output from the effective pixel;
A solid-state imaging device comprising: a correction unit that corrects a pixel signal output from an effective pixel of the group corresponding to the correction value using the correction value of each group calculated by the correction value calculation unit. . The correction value calculation unit calculates an average value of pixel signals output from each group of optical black pixels for each row as a correction value,
The solid-state imaging device according to claim 5, wherein the correction unit corrects the pixel signal by subtracting a corresponding correction value from the pixel signal.

Images