JP2020198586A - Imaging apparatus and control method thereof - Google Patents

Abstract

To correct the bias of a pixel signal caused by clipping a pixel signal without changing the black level of an image.SOLUTION: An imaging apparatus includes a plurality of first pixels (111) that output a pixel signal with a light receiving unit shielded from light, a plurality of second pixels (110) that output pixel signals on the basis of photoelectric conversion without shading the light receiving unit, an offset setting unit (230) that sets a first offset for the pixel signal of the first pixel, sets a second offset smaller than the first offset for the pixel signal of the second pixel, and clips the pixel signal of the second pixel to which the second offset has been set at the lower limit, and a correction unit (300) that corrects some of the pixel signals of the second pixel for which the second offset has been set.SELECTED DRAWING: Figure 1

Description

The present invention relates to an image pickup apparatus and a control method for the image pickup apparatus.

The image pickup apparatus reads an optical black level (OB level) signal from the shaded pixels to determine a reference black level. The signal value of each pixel is represented by a positive value. The image pickup device adds the offset level to the signal value of the pixel and replaces the value below the offset level with zero. If there are many pixels replaced with zeros in a specific area of the image, the signal value will be biased in that area, and it may appear colored as magenta.

In Patent Document 1, the pixel signal level of the dark part of the effective pixel is compared with the OB level, and when the pixel signal level of the dark part of the effective pixel is low, the reference black level is changed to suppress the replacement with zero. The method of doing so is disclosed.

Patent Document 2 discloses a method of suppressing replacement with zero by detecting a black level and switching the reference black level when the black level is equal to or less than a threshold value.

JP-A-2018-182366 JP-A-2007-336477

However, Patent Documents 1 and 2 change the reference black level of the entire image, and when the reference black level is set high, the range of data expressing the pixel signal becomes narrow.

An object of the present invention is to make it possible to correct the bias of the pixel signal caused by clipping the pixel signal without changing the black level of the image.

According to one aspect of the present invention, in the imaging device, a plurality of first pixels whose light receiving portion is shielded from light and outputs a pixel signal, and a plurality of first pixels whose light receiving portion is not shielded from light and outputs a pixel signal based on photoelectric conversion. A first offset is set for the second pixel and the pixel signal of the first pixel, and a second offset smaller than the first offset is set for the pixel signal of the second pixel. , One of the offset setting unit that clips the pixel signal of the second pixel in which the second offset is set at the lower limit value and the pixel signal of the second pixel in which the second offset is set. It has a correction unit that corrects the pixel signal of the unit.

According to the present invention, it is possible to correct the bias of the pixel signal caused by clipping the pixel signal without changing the black level of the image.

It is a figure which shows the configuration example of the image pickup apparatus. It is a figure which shows the histogram of the output signal of the offset setting part. It is a figure which shows the structural example of the zero clip correction part. It is a figure which shows the target pixel selection. It is a flowchart which shows the target pixel selection. It is a flowchart which shows the zero clip correction operation. It is a figure which shows the effect of a zero clip correction operation. It is a figure which shows the structural example of the zero clip correction part. It is a flowchart which shows the zero clip correction operation.

(First Embodiment)
FIG. 1 is a diagram showing a configuration example of the image pickup apparatus 160 according to the first embodiment. The image pickup device 160 includes an image pickup element 100 and a signal processing unit 200. The image pickup device 100 has a plurality of pixels 112, a control unit 120, a row circuit 130, a transfer circuit 140, and a serial I / F 150. The plurality of pixels 112 are arranged in a two-dimensional matrix and are divided into a plurality of aperture pixels 110 and a plurality of optical black pixels (OB pixels) 111. Each of the plurality of pixels 112 has a color filter of one of red (R), green (G), and blue (B). The OB pixel 111 is the pixel 112 in the upper row. The aperture pixels 110 are the central and lower pixels 112.

Each of the plurality of aperture pixels 110 does not block the light receiving portion, and photoelectrically converts the light image formed by the lens and outputs a pixel signal based on the photoelectric conversion. Each of the plurality of OB pixels 111 has the same configuration as the aperture pixel 110, the light receiving portion is shielded from light by aluminum wiring or the like, and a black level pixel signal is output. The control unit 120 causes the column circuit 130 to output the pixel signals of the pixels 112 of each row in order on a row-by-row basis. The column circuit 130 converts the pixel signal of the pixel 112 in each column from analog to digital. The transfer circuit 140 sequentially transfers the digital pixel signals of the pixels 112 in each row to the signal processing unit 200. The image sensor 100 outputs the pixel signals of all the pixels 112 to the signal processing unit 200 in order.

The signal processing unit 200 includes an optical black clamp processing unit (OB clamp processing unit) 210, a gain processing unit 220, an offset level setting unit 230, a storage unit 240, and a zero clip correction unit 300.

The OB clamp processing unit 210 sequentially inputs the pixel signals of all the pixels 112 from the transfer circuit 140. The pixel signal of the pixel 112 includes the pixel signal of the aperture pixel 110 and the pixel signal of the OB pixel 111. The OB clamp processing unit 210 calculates the average value of the pixel signals of the plurality of OB pixels 111, and determines the average value as the reference black level. Then, the OB clamping processing unit 210 clamps the black level of the pixel signals of the OB pixel 111 and the opening pixel 110 to "0" by subtracting the reference black level from the pixel signals of the OB pixel 111 and the opening pixel 110. .. The pixel signal output by the OB clamp processing unit 210 is, for example, a 16-bit signal with a positive or negative sign.

The gain processing unit 220 performs gain processing on the pixel signal output by the OB clamp processing unit 210, and outputs the gain-processed pixel signal.

The storage unit 240 has an SRAM or the like, and stores the correction threshold value TH, the aperture pixel offset, and the OB pixel offset read from the ROM of the image pickup device 160 when the image pickup device 160 is started.

The offset setting unit 230 sets the OB pixel offset stored in the storage unit 240 with respect to the pixel signal of the OB pixel 111 output by the gain processing unit 220, and outputs a positive pixel signal. Further, the offset setting unit 230 sets the opening pixel offset stored in the storage unit 240 with respect to the pixel signal of the opening pixel 110 output by the gain processing unit 220, and outputs a positive pixel signal.

Specifically, the offset setting unit 230 adds the OB pixel offset to the pixel signal of the OB pixel 111, and adds the aperture pixel offset to the pixel signal of the aperture pixel 110. The offset setting unit 230 determines the pixel signals of the OB pixel 111 and the aperture pixel 110 based on the pixel signal of the first aperture pixel 110 or the timing of the read start line of the aperture pixel 110. This timing is the timing of the synchronization signal used when the image sensor 100 reads out the pixel signal for each row, the timing of counting the number of pixel signals input to the offset setting unit 230, and the like. When the pixel signal after the addition becomes negative, the offset setting unit 230 clips the pixel signal to zero (lower limit value) by replacing the pixel signal with zero (lower limit value). The pixel signal of the pixel 112 has variations due to heat, noise on the circuit, and the like even in a completely shielded state.

2 (a) and 2 (b) are diagrams showing a histogram of the pixel signal of each pixel 112 output by the offset setting unit 230 when the noise has a normal distribution and the image is taken in a light-shielded state. FIG. 2A shows a histogram when a sufficiently large offset t1 is added to the pixel signal of the pixel 112 with respect to the variation in noise. FIG. 2B shows a histogram when a small offset t2 is added to the pixel signal of the pixel 112 with respect to the variation in noise. The offset t2 is smaller than the offset t1. In FIGS. 2A and 2B, the horizontal axis is the pixel signal value of the pixel 112, and the vertical axis is the frequency indicating the frequency of appearance.

In FIG. 2A, when the pixel signals have a normal distribution, the frequency of appearance gradually decreases around the offset t1. The average value u1 is an average value for the pixel signal output by the offset setting unit 230, and indicates an average value of the black level.

In FIG. 2B, when the pixel signals have a normal distribution, the frequency of appearance gradually decreases around the offset t2. The average value u2 is an average value for the pixel signal output by the offset setting unit 230, and indicates an average value of the black level.

In FIG. 2A, the offset t1 and the average value u1 coincide with each other. On the other hand, in FIG. 2B, the average value u2 is larger than the offset t2. This is because the offset setting unit 230 clips the negative pixel signal at zero as described above. If the average value u2 deviates from the offset t2, the image processing based on the offset t2 is adversely affected. For example, when the gain processing unit 220 performs different gain processing on the pixel signal shown in FIG. 2B for each color of the color filter of the pixel 112, such as white balance gain, the pixel signal appears as a color shift in the image. ..

The offset setting unit 230 uses the offset t1 as the OB pixel offset and the offset t2 as the aperture pixel offset. In that case, FIG. 2A corresponds to the histogram of the pixel signal of the OB pixel 111 in the shaded state, and FIG. 2B corresponds to the histogram of the pixel signal of the aperture pixel 110 in the shaded state.

The zero clip correction unit 300 of FIG. 1 performs a process of suppressing a deviation between the average value of the black level output by the offset setting unit 230 and the offset due to the offset setting unit 230 clipping at zero.

FIG. 3 is a diagram showing a configuration example of the zero clip correction unit 300. The zero clip correction unit 300 includes a demultiplexer 310 that distributes the pixel signals of the OB pixel 111 and the opening pixel 110, a correction value calculation unit 320, a buffer memory 340, a target pixel selection unit 350, and a zero clip correction calculation unit 360. The storage unit 240 stores the correction threshold value TH, the OB pixel offset t1, and the aperture pixel offset t2.

The demultiplexer 310 inputs the pixel signal output by the offset setting unit 230, outputs the pixel signal of the OB pixel 111 to the correction value calculation unit 320, and targets the pixel signals of the OB pixel 111 and the opening pixel 110 with the buffer memory 340. Output to the pixel selection unit 350.

The correction value calculation unit 320 reads the OB pixel offset t1 and the opening pixel offset t2 from the storage unit 240, and subtracts (t1-t2) from the pixel signals of all the OB pixels 111 to obtain the average value of the negative values. Ask. (T1-t2) is the difference between the OB pixel offset t1 and the aperture pixel offset t2. The correction value calculation unit 320 outputs the average value of the negative values as the correction value to the zero clip correction calculation unit 360.

The buffer memory 340 stores the pixel signals of the OB pixel 111 and the aperture pixel 110. The target pixel selection unit 350 detects a “0” value (lower limit value) from the pixel signals of the OB pixel 111 and the opening pixel 110, and sets the pixels around the “0” value pixel as the correction target pixel, and the correction target pixel. The coordinate data of is output to the zero clip correction calculation unit 360.

FIG. 4 is a diagram for explaining the processing performed by the target pixel selection unit 350. FIG. 4 shows the pixel signal input by the target pixel selection unit 350 in two-dimensional coordinates. The pixel signals are assigned to the coordinates of i, i + 1, i + 2, ... In the horizontal direction in the order of input, and are further assigned to the coordinates of j, j + 1, j + 2, ... In the vertical direction for each pixel signal for one line.

The color filter of each pixel 112 is a red (R), green (G) and blue (B) Bayer array. In FIG. 4, the white pixel 112 indicates a green color filter pixel, the shaded pixel 112 indicates a red color filter pixel, and the dot pixel 112 indicates a blue color filter pixel.

The target pixel selection unit 350 determines whether or not there is a pixel signal having a “0” value in the area Sk of 6 × 6 pixels. When a pixel signal having a “0” value exists in the area Sk, the target pixel selection unit 350 does not have a “0” value among the pixel 112 of the color filter having the same color as the pixel 112 having the “0” value in the area Sk. The pixel of the pixel signal is selected as the correction target pixel.

For example, the pixel 112 of the red color filter at the coordinates (i + 2, j + 2) is a pixel signal having a “0” value. In that case, the target pixel selection unit 350 has coordinates (i, j), (i + 2, j), (i + 4, j), (i, j + 2), (i + 4, j + 2), (i, j + 4), (i + 2,). Eight pixels of the red color filters j + 4) and (i + 4, j + 4) are selected as the correction target pixels. Although the processing of the pixel 112 of the red color filter has been described, the processing of the pixel 112 of the green and blue color filters is also the same.

The “0” value pixel 112 may include an error because it includes pixels in which the pixel signal has been replaced from a negative value to zero. The target pixel selection unit 350 selects the surrounding pixels 112 as the correction target pixels in order to distribute the error to the peripheral pixels 112 of the "0" value pixel 112.

In FIG. 4, the pixels 112 do not overlap in the adjacent area Sk, but the pixels 112 may overlap in the adjacent area Sk. Further, the size of the area Sk is not limited to 6 × 6 pixels.

Further, the shape of the area Sk is not limited to a rectangle. For example, the area Sk may be a rhombus having four coordinates (i + 2, j), (i, j + 2), (i + 4, j + 2) and (i + 2, j + 4) as vertices. Further, the area Sk may be set to a different area for each color of the color filter of the pixel 112.

FIG. 5 is a flowchart showing a control method of the target pixel selection unit 350. In step S501, the target pixel selection unit 350 determines whether or not the processing of all areas Sk has been completed. The target pixel selection unit 350 proceeds to step S502 when the processing of all area Sks is not completed, and ends the processing of the flowchart of FIG. 5 when the processing of all area Sks is completed.

In step S502, the target pixel selection unit 350 detects a pixel signal having a “0” value in the area Sk, and proceeds to step S503.

In step S503, the target pixel selection unit 350 determines whether or not there is a pixel signal having a “0” value in the area Sk. The target pixel selection unit 350 proceeds to step S504 when there is a pixel signal having a “0” value in the area Sk, and returns to step S501 when there is no pixel signal having a “0” value in the area Sk. ..

In step S504, the target pixel selection unit 350 uses the pixel 112 of the pixel signal of the pixel signal of the same color as the pixel 112 of the pixel signal of the “0” value in the area Sk and the pixel 112 of the pixel signal of the pixel signal of the same color as the color filter. The coordinates are detected, and the process proceeds to step S505.

In step S505, the target pixel selection unit 350 outputs the coordinates detected in step S504 as the coordinates of the correction target pixel to the zero clip correction calculation unit 360, and returns to step S501.

The target pixel selection unit 350 may convert the coordinates of the correction target pixel into coordinates corresponding to the rows and columns of the pixels 112 of the image sensor 100 and output the coordinates. Further, when there are overlapping pixels 112 in the adjacent area Sk, the target pixel selection unit 350 masks the coordinates indicating the same pixel 112 so as not to be output a plurality of times, and then outputs the coordinates.

In step S501, when the processing of all the areas Sk is completed, the target pixel selection unit 350 ends the processing of the flowchart of FIG.

FIG. 6 is a flowchart showing a control method of the zero clip correction calculation unit 360 of FIG. In step S601, the zero clip correction calculation unit 360 acquires the correction value CR output by the correction value calculation unit 320, and proceeds to step S602.

In step S602, the zero clip correction calculation unit 360 acquires the correction threshold TH from the storage unit 240, and proceeds to step S603.

In step S603, the zero clip correction calculation unit 360 acquires the coordinates of the correction target pixel from the target pixel selection unit 350, and proceeds to step S604.

In step S604, the zero clip correction calculation unit 360 reads the pixel signals of the pixels 112 in order from the buffer memory 340, and determines whether or not the pixel signal Ad of the pixels 112 corresponding to the coordinates of the pixels to be corrected is smaller than the correction threshold TH. To do. When the pixel signal Ad of the pixel 112 corresponding to the coordinates of the pixel to be corrected is smaller than the correction threshold TH, the zero clip correction calculation unit 360 is a pixel signal corresponding to the coordinates of the pixel to be corrected, as in Equation (1). Correct Ad. The zero-clip correction calculation unit 360 adds the product of the correction value CR and the coefficient α to the pixel signal Ad corresponding to the coordinates of the pixel to be corrected to obtain the corrected pixel signal Ae. The zero clip correction calculation unit 360 outputs the corrected pixel signal Ae. Here, the coefficient α is a coefficient for adjusting the correction amount.
Ae = Ad + CR × α ・ ・ ・ (1)

As described above, the zero clip correction calculation unit 360 is the peripheral pixels 112 in which the pixel signal Ad is clipped to the “0” value and the pixel signal Ad is not the “0” value. Correct the pixel signal of. Specifically, the zero clip correction calculation unit 360 corrects the pixel signal of the pixel 112 having a color filter of the same color as the color filter of the pixel 112 whose pixel signal Ad is clipped to a “0” value.

The correction threshold value TH may be set to a different value for each color of the color filter of the pixel 112, or may be a single value. The coefficient α may be 1. The coefficient α is set for each color of the color filter of the pixel 112, and the zero clip correction calculation unit 360 may control the correction amount for each color according to the coefficient α. For example, when a reddish color shift occurs in the region near the black level of the corrected image, the zero clip correction calculation unit 360 sets a large coefficient α corresponding to the pixel 112 of the red color filter. By doing so, the color shift can be corrected.

Since the correction value CR is a negative value, the equation (1) is shown by an addition equation, but the equation (1) may be a subtraction equation depending on the positive and negative signs of the correction value CR. The zero clip correction calculation unit 360 corrects the corrected pixel signal Ae so that it is smaller than the uncorrected pixel signal Ad. When the corrected pixel signal Ae is negative, the zero clip correction calculation unit 360 replaces the corrected pixel signal Ae with zero. The zero-clip correction calculation unit 360 performs the above processing on all pixel signals.

As described above, the zero clip correction unit 300 corrects a part of the pixel signals of the plurality of aperture pixels 110.

7 (a) and 7 (b) are diagrams for explaining the effect of the correction process of the zero clip correction calculation unit 360. FIG. 7A is a diagram showing a histogram of the pixel signal Ad input from the buffer 340 by the zero clip correction calculation unit 360 when the image is taken in a light-shielded state. FIG. 7B is a diagram showing a histogram of the pixel signal Ae output by the zero clip correction calculation unit 360 when the image is taken in a light-shielded state.

The aperture pixel offset t2 in FIGS. 7 (a) and 7 (b) is the aperture pixel offset t2 stored in the storage unit 240. In FIG. 7A, the average value u2 is the average value of the pixel signals Ad before correction. In FIG. 7B, the average value u3 is the average value of the corrected pixel signals Ae. The correction threshold value TH is a correction threshold value input by the zero clip correction calculation unit 360 from the correction value calculation unit 320.

In FIG. 7B, the decrease in the appearance frequency due to the correction process is shown by a black block, and the increase in the appearance frequency due to the correction process is shown by a shaded block. The zero-clip correction calculation unit 360 performs the correction processing of the equation (1) on the pixel signal Ad smaller than the correction threshold TH, and obtains the corrected pixel signal Ae. The histogram of the pixel signal Ae smaller than the correction threshold value TH changes with respect to the pixel signal Ad before correction. The appearance limit of the large pixel signal Ae decreases, and the appearance frequency of the small pixel signal Ae increases. As a result, the average value u3 of the corrected pixel signal Ae becomes a value closer to the aperture pixel offset t2 than the average value u2 of the pixel signal Ad before correction. As a result, the image pickup apparatus 160 can suppress the deviation between the aperture pixel offset t2 and the average value u2 caused by clipping the pixel signal at zero without changing the black level of the image. As a result, it is possible to prevent the clipped image area from appearing as colored magenta or the like.

(Second Embodiment)
FIG. 8 is a diagram showing a configuration example of the zero clip correction unit 300 according to the second embodiment. The image pickup apparatus 160 of the second embodiment is different from the image pickup apparatus 160 of the first embodiment in the zero clip correction unit 300. The zero clip correction unit 300 of FIG. 8 is obtained by adding a correction value storage unit 330 to the zero clip correction unit 300 of FIG. Hereinafter, the difference between the second embodiment and the first embodiment will be described.

The correction value calculation unit 320 reads the OB pixel offset t1 and the opening pixel offset t2 from the storage unit 240, and subtracts (t1-t2) from the pixel signals of all the OB pixels 111, and N negative values are obtained. The correction value is sequentially output to the correction value calculation unit 320. The correction value storage unit 330 stores N correction values CR (k) output by the correction value calculation unit 320. The correction value CR (k) is the kth correction value. k is an integer value from 0 to N-1.

FIG. 9 is a flowchart showing a control method of the zero clip correction calculation unit 360 according to the second embodiment. In step S901, the zero clip correction calculation unit 360 acquires N correction values CR (k) stored in the correction value calculation unit 330, and proceeds to step S902.

In step S902, the zero clip correction calculation unit 360 acquires the correction threshold value TH stored in the storage unit 240, and proceeds to step S903.

In step S903, the zero clip correction calculation unit 360 acquires the coordinates of the correction target pixel from the target pixel selection unit 350, and proceeds to step S904.

In step S904, the zero clip correction calculation unit 360 reads out the pixel signals of the pixels 112 in order from the buffer memory 340, and determines whether or not the pixel signal Ad of the pixels 112 corresponding to the coordinates of the pixels to be corrected is smaller than the correction threshold TH. To do. When the pixel signal Ad of the pixel 112 corresponding to the coordinates of the pixel to be corrected is smaller than the correction threshold TH, the zero clip correction calculation unit 360 has the pixel 112 corresponding to the coordinates of the pixel to be corrected as in the equation (2). The pixel signal Ad of is corrected. The zero-clip correction calculation unit 360 adds the product of the correction value CR (k) and the coefficient α to the pixel signal Ad of the pixel 112 corresponding to the coordinates of the pixel to be corrected, and obtains the corrected pixel signal Ae. .. The zero clip correction calculation unit 360 outputs the corrected pixel signal Ae. Here, the coefficient α is a coefficient for adjusting the correction amount.
Ae = Ad + CR (k) × α ・ ・ ・ (2)

k is incremented from 0 to N-1, and then reset to 0, and the process is repeated. In the first cycle from 0 to N-1, the zero clip correction calculation unit 360 adds the product of the kth correction value CR (k) and the coefficient α to the kth pixel signal Ad. The k-th pixel signal Ae after correction is obtained. After that, the zero clip correction calculation unit 360 adds the product of the correction value CR (0) and the coefficient α to the Nth pixel signal Ad to obtain the corrected Nth pixel signal Ae. After that, the zero clip correction calculation unit 360 adds the product of the correction value CR (1) and the coefficient α to the N + 1th pixel signal Ad to obtain the corrected N + 1th pixel signal Ae.

As described above, the zero clip correction calculation unit 360 repeatedly uses N correction values CR (k) in order to correct a plurality of pixel signals Ad corresponding to the coordinates of the pixel to be corrected in order, and a plurality of corrected values. To obtain the pixel signal Ae of. Since the N correction values CR (k) have random variations, the corrected plurality of pixel signals Ae can disperse the bias. As a result, the zero-clip correction calculation unit 360 can obtain a natural correction result.

Similar to the first embodiment, the image pickup apparatus 160 can suppress the deviation between the aperture pixel offset t2 and the average value u2 caused by clipping the pixel signal at zero without changing the black level of the image. it can. Further, the zero clip correction calculation unit 360 can obtain an image having a small bias of the corrected pixel signal Ae by correcting the pixel signal Ae using N correction values CR (k).

Although the first and second embodiments have been described above, the present invention is not limited to these embodiments, and various modifications and modifications can be made within the scope of the gist thereof.

The image pickup device 160 can be applied to smartphones, tablets, industrial cameras, medical cameras, in-vehicle cameras, and the like, in addition to digital cameras and video cameras.

100 image sensor, 110 aperture pixel, 111 OB pixel, 120 control unit, 130 row circuit, 140 transfer circuit, 150 serial I / F, 160 image pickup device, 200 signal processing unit, 210 OB clamp processing unit, 220 gain processing unit, 230 offset setting unit, 240 storage unit, 300 zero clip correction unit

Claims (13)

The light receiving part is shielded from light, and a plurality of first pixels that output pixel signals and
A plurality of second pixels that output pixel signals based on photoelectric conversion without shading the light receiving part, and
A first offset is set for the pixel signal of the first pixel, a second offset smaller than the first offset is set for the pixel signal of the second pixel, and the second offset is set. An offset setting unit that clips the pixel signal of the second pixel in which is set at the lower limit value,
An imaging device including a correction unit that corrects a part of the pixel signals of the second pixel to which the second offset is set. It further has a clamping processing unit that clamps the pixel signal of the first pixel and the pixel signal of the second pixel based on the black level based on the pixel signal of the first pixel.
The offset setting unit sets the first offset with respect to the pixel signal of the clamped first pixel, and sets the second offset with respect to the pixel signal of the clamped second pixel. The imaging device according to claim 1, wherein the image pickup apparatus is used. It further has a gain processing unit that performs gain processing on the pixel signals of the clamped first and second pixels.
The offset setting unit sets the first offset with respect to the pixel signal of the gain-processed first pixel, and the second offset with respect to the pixel signal of the gain-processed second pixel. The imaging device according to claim 2, wherein the image pickup apparatus is set. The imaging device according to any one of claims 1 to 3, wherein the correction unit corrects a part of the pixel signals so as to be small. The correction unit is characterized in that a part of the pixel signal is corrected based on the result of subtracting the difference between the first offset and the second offset from the pixel signal of the first pixel. The imaging device according to any one of claims 1 to 4. The correction unit is based on the average value of the negative values of the result of subtracting the difference between the first offset and the second offset from the pixel signal of the first pixel, and the partial pixel signal. The imaging apparatus according to any one of claims 1 to 5, wherein the image pickup device is characterized in that. The correction unit sequentially uses the negative value of the result of subtracting the difference between the first offset and the second offset from the pixel signal of the first pixel, and sequentially uses some of the pixel signals. The imaging device according to any one of claims 1 to 5, wherein the image pickup apparatus is corrected. The correction unit according to any one of claims 1 to 7, wherein the correction unit corrects the pixel signal of the second pixel around the second pixel whose pixel signal is clipped to the lower limit value. Imaging device. The imaging device according to claim 8, wherein the correction unit corrects a pixel signal of the surrounding second pixel whose pixel signal is not the lower limit value. The plurality of second pixels have a color filter and
The correction unit 8 or 9 is characterized in that the correction unit corrects a pixel signal of a second pixel having a color filter of the same color as the color filter of the second pixel whose pixel signal is clipped to the lower limit value. The imaging apparatus according to. Any one of claims 8 to 10, wherein the correction unit corrects the pixel signal of the surrounding second pixel when the pixel signal of the surrounding second pixel is smaller than the threshold value. The imaging apparatus according to. When the pixel signal of the surrounding second pixel is smaller than the threshold value, the correction unit corrects the pixel signal of the surrounding second pixel.
The imaging device according to claim 10, wherein the threshold value is a value different for each color of the color filter of the second pixel. The light receiving part is shielded from light, and a plurality of first pixels that output pixel signals and
It is a control method of an image pickup apparatus having a plurality of second pixels that output a pixel signal based on photoelectric conversion without shading the light receiving portion.
A first offset is set for the pixel signal of the first pixel, a second offset smaller than the first offset is set for the pixel signal of the second pixel, and the second offset is set. The offset setting step of clipping the pixel signal of the second pixel in which is set at the lower limit value, and
A control method for an image pickup apparatus, which comprises a correction step for correcting a part of a pixel signal of the pixel signals of the second pixel in which the second offset is set.

Images