JPH066599A - Solid-state image pickup device and its signal processing method - Google Patents

Abstract

PURPOSE:To reduce color mixture due to leaked light and stored charge in other pixel by reading a signal comprising plural pixel strings each comprising plural pixels and applying arithmetic operation and correction to an output from a pixel in a pixel string in response to an output from a pixel corresponding to a pixel in other pixel string. CONSTITUTION:A charge storage section is provided beneath a shift gate 4, and a transfer gate 5 is arranged between registers CK1, 2 of CCD registers 9, 11 in which the registers CK1, 2 are arranged. Furthermore, pixels are shielded by a light shield aluminum 20. Through the constitution above, each signal output based on the signal charge from 2nd color pixels 2A-D is corrected by a coefficient obtained from the signal output from 3rd color pixels 3A-D or 1st color pixels 1A-D. On the other hand, outputs of the pixels B, C are averaged and a pixel 2C is corrected by a coefficient and the method is implemented by an arithmetic operation circuit 0C provided to an output circuit 19. Thus, when a signal charge is read, color mixture due to leaked light or charge storage in other pixel is decreased.

Description

Detailed Description of the Invention

[0001]

[Object of the Invention]

[0002]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a solid-state image pickup device and a signal processing method thereof, and more particularly to a solid-state image pickup device suitable for a three-line linear image sensor used for reading an original in a copying machine and the signal processing method thereof. It is about.

[0003]

2. Description of the Related Art FIG. 5 is a diagram showing a device configuration of a general image reading sensor. In particular, in a 3-line linear image sensor having a pixel row interval in the sub-scanning direction of 1 line, that is, the same size as the pixel repeating pitch in the main scanning direction, the signal charge of the central pixel row is read out between the pixel rows at both ends. 1 illustrates an example of a transfer type configuration. As shown in the figure, the first color pixels 1A, 1B, 1C, 1D to which a filter of one of the three primary colors (R, G, B) or the three complementary colors (Cy, Mg, Ye) is applied, and the remaining A second color pixel 2A provided with a filter for one of the two colors
2B, 2C, and 2D and the third color pixels 3A, 3B, 3C, and 3D to which the filter of the remaining one color is provided are arranged at one line pitch. The shift gate 4 includes a second color pixel 2A,
2B, 2C, a part of 2D and the third color pixels 3A, 3B,
In order to transfer the signal charges of 3C and 3D to the CCD register 9, or to transfer the signal charges of the first color pixels 1A, 1B, 1C and 1D, and the remaining part of the second color pixels 2A, 2B, 2C and 2D. Used to transfer to register 8. The transfer gate 5 is used to transfer the signal charge of the CCD register 9 to the CCD register 11 and the signal charge of the CCD register 8 to the CCD register 10. Further, the output circuit 19 derives the first pixel row ODD output signal 13 and the first pixel row EVEN output signal 14 from the CCD register 8, and further outputs the third pixel row ODD output signal 17 and the third pixel row from the CCD register 9. E
The VEN output signal 18 is derived, and further the CCD register 1
The second pixel column EVEN output signal 16 is derived from 0, and the second pixel column ODD output signal 1 is further derived from the CCD register 11.
Derive 5

In the above structure, the signal charges of the second color pixels 2A, 2B, 2C and 2D in the center are the same as the first charge at both ends.
The color pixels 1A, 1B, 1C, 1D and the third color pixel 3A,
It passes through the pixels 3B, 3C, and 3D, and is read once by the shift gate 4 to the CCD registers 8 and 9 arranged inside. After that, by opening the transfer gate 5, the charges of the CCD registers 8 and 9 are transferred to the CCD register 11 arranged outside, and the second pixel column ODD (odd number) output signal 15 and Second pixel column EVEN (even number) output signal 1
It is read as 6.

On the other hand, the first color pixels 1A, 1B, 1 at both ends
The signal charges of C, 1D and the third color pixels 3A, 3B, 3C, 3D are read out to the corresponding inner CCD registers 8, 9 through the shift gate 4, respectively. After this,
The charges are transferred in the horizontal direction, and are separated into charges of odd-numbered pixels and charges of even-numbered pixels on the way. After this,
Through the output circuit 19, the first pixel column ODD output signal 1
3, first pixel column EVEN output signal 14, third pixel column OD
The D output signal 17 and the third pixel column EVEN output signal 18 are output.

FIG. 6 shows another example of the device configuration of a general image reading sensor. In particular, with a 3-line linear image sensor in which the pixel row interval in the sub-scanning direction is 1 line, that is, the same size as the pixel repeating pitch in the main scanning direction, the signal charge of the pixel row at either end is read out, It illustrates an intra-pixel transfer type configuration for reading out a signal charge of a pixel column. As shown in FIG. 6, the signal charges of the first color pixels 1A, 1B, 1C and 1D are transferred from the shift gate 7 through the CCD register 8 and the output circuit 19 to the first pixel column ODD output signal 13 and the first pixel column EVEN. It is derived as the output signal 14.
On the other hand, the signal charges of the second color pixels 2A, 2B, 2C and 2D are transferred to the shift gate 6 and the third color pixels 3A, 3B, 3C and 3D.
D, shift gate 4, CCD register 12, output circuit 1
9 to be output as the second pixel column ODD output signal 15 and the second pixel column EVEN output signal 16. The signal charges of the third color pixels 3A, 3B, 3C, and 3D pass through the shift gate 4, the CCD register 12, the transfer gate 5, the CCD register 9, and the output circuit 19 to the third pixel column ODD output signal 17 and the third pixel column ODD output signal 17. It is derived as the pixel column EVEN output signal 18.

In the above configuration, the third color pixel 3
The signal charges of A, 3B, 3C, and 3D are read out to the CCD register 12 inside by opening the shift gate 4. Further, by opening the transfer gate 5, the data is transferred to the outside CCD register 9. After that, the charges are transferred in the horizontal direction, and are separated into charges of odd-numbered pixels and charges of even-numbered pixels on the way. Then, respectively, the output circuit 1
9 through the third pixel column ODD output signal 17 and the third
It is output as the pixel column EVEN output signal 18. Second
The signal charges of the color pixels 2A, 2B, 2C and 2D are opened by opening the shift gate 6 so that the third color pixels 3A, 3B and 3D.
Pass through C and 3D. Further, by opening the shift gate 4, the data is once read out to the CCD register 12. Further, the charges are transferred in the horizontal direction, and are separated into charges of odd-numbered pixels and charges of even-numbered pixels on the way. Then, through the output circuit 19, respectively, the second pixel column ODD output signal 15,
It is output as the second pixel column EVEN output signal 16. On the other hand, the first color pixels 1A, 1B, 1C and 1D are read out to the CCD register 8 through the shift gate 7. Further, the charges are transferred in the horizontal direction, and are separated into charges of odd-numbered pixels and charges of even-numbered pixels on the way. Then, they are respectively output as the first pixel column ODD output signal 13 and the first pixel column EVEN output signal 14 through the output circuit 19.

[0008]

As described above, the conventional image reading sensor uses the inter-pixel transfer or the intra-pixel transfer,
It is configured to transfer the signal charge. Therefore, there is a problem that color mixing between pixels is likely to occur. That is, the problem with the configuration of FIG. 5 is that although light is shielded by aluminum, it leaks light when exposed to strong light, and for example, originally the third color pixels 3A, 3B, 3C,
The color information to be read in 3D is the second color pixels 2A, 2B, 2
It leaks into C and 2D, causing color mixing.
On the other hand, if an attempt is made to increase the light-shielding margin by increasing the area of the light-shielding portion, there is a problem that the pixel apertures of the pixel rows at both ends become small and the sensitivity decreases. On the other hand, the problem with the configuration shown in FIG. 6 is that the third color pixels 3A, 3B, 3C,
Charges are accumulated in the third color pixels 3A, 3B, 3C, and 3D until the 3D signal charge is transferred to the outer CCD register 9, and signals of the second color pixels 2A, 2B, 2C, and 2D are stored. This means that the color is mixed with the charge and then output, resulting in color mixing. From the above, the development of an effective image read signal processing method for reducing color mixture has been a major issue.

The present invention has been made in view of the above, and an object thereof is a solid-state image pickup device and its signal capable of correcting the color mixture even when the signal charge transfer occurs and reducing the influence on the output image. It is to provide a processing method.

[0010]

A solid-state image pickup device according to the present invention includes a plurality of pixel columns having a plurality of pixels, a signal reading unit for reading a signal from each pixel column, and the plurality of pixel columns. An arithmetic unit that corrects an output from a pixel in a pixel row according to an output from a pixel in the other pixel row corresponding to the certain pixel.

The signal processing method of the solid-state image pickup device of the present invention is
In reading a signal from each of a plurality of pixel columns having a plurality of pixels, the output of a pixel in a certain pixel column of the plurality of pixel columns corresponds to the certain pixel in another pixel column. It is configured to be corrected according to the output of the pixel.

[0012]

The output is obtained from each pixel of the plurality of pixel columns. Then, the output of the pixel of the specific pixel column is corrected based on the output of the corresponding pixel of the other pixel column. For example, an amount obtained by multiplying the R output by a certain count is subtracted from the G output. This reduces the effect of color mixing.

[0013]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a detailed enlarged block diagram of a device integrated on the same semiconductor substrate for explaining one embodiment of the present invention. In particular, the case where it is applied to the basic configuration as shown in FIG. 5 is illustrated. As shown in FIG. 1, a charge storage section is provided below the shift gate 4. C
The CD registers 9 and 11 have registers CK1 and CK1, respectively.
It is configured by arranging a register CK2. The transfer gate 5 is configured by arranging a register TR between the registers CK2 and CK2 of the CCD registers 9 and 11. A light-shielding aluminum 20 that shields light between the pixels is arranged.

In the configuration as described above, the following operation will be described with reference to the time chart of FIG. By the way, FIG. 3A shows the register C of the CCD registers 9 and 11.
K2 clock, CCD register 9 and 11 in FIG.
The clock of the register CK1 is shown in FIG. 6C, the clock is given to the shift gate 4, and the clock shown in FIG. The time TINT from one opening of the shift gate 4 to the next opening corresponds to the integration time of the signal charge.

At time t = t1, the third color pixel 3
The signal charges generated in B and 3C are accumulated in the shift gate 4, and the signal charges generated in the second color pixel 2C pass through the shift gate 4 and the register CK2 of the CCD register 9.
Is carried to. Transfer gate 5 at time t = t2
By opening, the electric charges are all transferred to the register CK2 of the outside CCD register 11 by the time t = t4. At time t = t5, the adjacent register CK1
Is transferred to the output circuit 19 by the transfer pulse given to the CCD register 11. On the other hand, the signal charges generated in the third color pixels 3B and 3C and accumulated in the shift gate 4 remain in place until time t = t4, and are stored in the register CK1 of the inner CCD register 9 at time t = t5. Read out. The signal charge generated in the third color pixels 3A and 3D behaves similarly to the signal charge generated in the third color pixels 3B and 3C. The signal charge generated in the second color pixel 2A is
It behaves the same as that of the second color pixel 2C.

Now, what has been a problem in the conventional example is light leakage to the transfer path of the signal of the second color pixel 2C passing between the pixels of the third color pixels 3B and 3C. That is, FIG.
In the above, if the distances a and b between the transfer path and the pixel openings of the third color pixels 3B and 3C are small, the light leakage increases due to the wraparound, and if the distances a and b are large, the pixel opening is small and the sensitivity is lowered. That is.

Therefore, in the embodiment of the present invention, the output of the second color pixel 2C is corrected by multiplying the output of the third color pixel 3B by a coefficient, so that the influence of the color mixture on the output of the second color pixel 2C is corrected. Has been reduced. That is, the second color pixels 2A, 2
Coefficients obtained from the signal outputs from the corresponding third color pixels 3A, 3B, 3C, 3D or the first color pixels 1A, 1B, 1C, 1D, respectively, based on the signal charges from B, 2C, and 2D. The second color pixel 2A,
The color mixture components included in the 2B, 2C, and 2D outputs are reduced.

On the other hand, the influence of light leakage can be similarly reduced by averaging the outputs of the third color pixels 3B and 3C and correcting the second color pixel 2C by multiplying them by a coefficient. Therefore, among the first color pixels 1A, 1B, 1C, 1D or the third color pixels 3A, 3B, 3C, 3D, the output of the pixels on both sides through which the outputs of the second color pixels 2A, 2B, 2C, 2D pass. A coefficient obtained from the average value of the corresponding second color pixel 2A,
Even if correction is added to the outputs of 2B, 2C, and 2D,
Similarly, color mixture can be reduced.

The above method can be easily realized by the simple arithmetic circuit 0C provided in the output stage of the image reading device. Further, even if a circuit for reducing color mixture is provided inside the image reading device, the same effect can be obtained.

FIG. 2 is a detailed enlarged block diagram of a device for explaining a case where the present invention is applied to another image reading sensor. In particular, the case where it is applied to the basic configuration as shown in FIG. 6 is illustrated. As shown in FIG. 6, the CCD registers 12 and 9 are configured by arranging a register CK1 and a register CK2, respectively. Transfer gate 5
Is configured by arranging the register TR between the registers CK2CK2 of the CCD registers 9 and 12. Light-shielding aluminum 20 is arranged between each pixel.

The operation of the above arrangement will be described with reference to the time chart of FIG. By the way,
FIG. 4A shows the register CK2 of the CCD registers 9 and 12.
, The clock of the register CK1 of the CCD registers 9 and 12 is shown in FIG. 7B, and the shift gate 4 is shown in FIG.
To the shift gate 6, and FIG. 7E shows the clock to the transfer gate 5.
The time TINT 3 is the third color pixels 3A, 3B, 3C, 3
The accumulation time of the signal charge of D, the time TINT 2 is the second color pixel 2
The signal charge accumulation time of A, 2B, 2C, and 2D corresponds to the charge accumulation time of the second color pixels 2A, 2B, 2C, and 2D mixed in the second color pixels 2A, 2B, 2C, and 2D. To do.

The signal charges generated in the third color pixels 3C and 3D pass through the shift gate 4 at time t = T1 and the inner CC
It is read to the register CK1 of the D register 12. At time t = T2, it moves to the adjacent register CK2. The signal charge generated in the third color pixel 3C is transferred to the transfer gate 5
Is opened, the data is transferred to the register CK2 of the outer CCD register 9 by the time t = T5. However, the signal charge generated in the third color pixel 3D is time t = T
It stays at the location of the register CK2 of the CCD register 12 which has moved to 2 until time t = T5. At time t = T6, the signal charges of the third color pixels 3C and 3D move to the register CK1 of the adjacent CCD register 12, and further move to the adjacent register CK2 at time t = T7. Next, transfer gate 5
By opening, the signal charge generated in the third color pixel 3D is transferred to the register CK2 of the outside CCD register 9 by time t = T10. At this time, the signal charge generated in the third color pixel 3C stays at the place where it was at the time t = T7. When there is no signal charge in the inner CCD register 12, the shift gate 6 is opened at time t = T11, the signal charge of the second color pixel 2C is transferred to the pixel of the third color pixel 3C, and the signal charge of the second color pixel 2D is transferred to the pixel of the third color pixel 3C. It transfers to each pixel of the 3rd color pixel 3D. At this time, the signal charge generated in the third color pixel 3C remains at the position of the register CK2 that was at the time t = T7, and the signal charge generated in the third color pixel 3D is the time t = T1.
It stays in the place of register CK2 which was 0.

Next, at time t = T12, the shift gate 4 is opened, and the charges generated in the second color pixels 2C and 2D are read into the register CK1 of the CCD register 12. At this time, the signal charges of the third color pixels 3C and 3D are at time t = T1.
It is transferred to the register CK1 next to the register CK2 that was at the time of 1. As a result, the read operation from a series of pixels is completed, and the transfer pulse from the CCD register 12 is carried to the output circuit 19.

Here, the third problem has heretofore been a problem.
CCD register 9 where the signal charges of color pixels 3C and 3D are outside
The signal charge is accumulated in each pixel of the third color pixels 3C and 3D by the time until it is transferred to the second color pixel 2C,
It was to be added to the 2D signal charge and then output.

However, according to the embodiment of the present invention, the outputs of the second color pixels 2C and 2D are corrected by multiplying the output of the third color pixel 3C by the coefficient and the output of the third color pixel 3D by the coefficient. Therefore, color mixture can be prevented. Similarly, the second color pixels 2A, 2B, 2 are determined by the coefficients obtained from the signal outputs of the respective pixels of the third color pixels 3A, 3B, 3C, 3D.
By correcting the signal output of each pixel of C and 2D, it is possible to obtain the signal output of which the influence of color mixture is reduced from the second color pixels 2A, 2B, 2C, and 2D.

In FIG. 4, the second color pixels 2A, 2
Accumulation time of B, 2C, 2D is time TINT, third color pixel 3
Let TINT be the accumulation time of A, 3B, 3C, and 3D, and let T be the accumulation time of charges of the second color pixel 2A and the third color pixels 3A, 3B, 3C, and 3D mixed in 2B, 2C, and 2D. Output 100 of the third color pixels 3A, 3B, 3C, 3D
The ratio of color mixture can also be reduced by subtracting / TINT (%) from the outputs of the second color pixels 2A, 2B, 2C, and 2D.

[0028]

As described above, according to the present invention, after the output is obtained from each pixel of a plurality of pixel columns, the output of each pixel of a specific pixel column is corresponded to other pixel columns. By correcting with a coefficient obtained based on the output of a pixel, when signal charges are read out from a plurality of line sensors arranged in parallel, light leakage and color mixture due to charge accumulation in other pixels are reduced. There is an effect that a possible image reading signal processing method can be realized.

[Brief description of drawings]

FIG. 1 is a detailed enlarged configuration diagram of a device for explaining an embodiment of the present invention.

FIG. 2 is a detailed enlarged configuration diagram of a device for explaining a case where the present invention is applied to another image reading sensor.

FIG. 3 is a time chart for explaining the operation of the configuration of FIG.

FIG. 4 is a time chart for explaining the operation of the configuration of FIG.

FIG. 5 is a device configuration diagram of a general image reading sensor.

FIG. 6 is a device configuration diagram of another example of a general image reading sensor.

[Explanation of symbols]

 1A to 1D First color pixel 2A to 2D Second color pixel 3A to 3D Third color pixel 4,6,7 Shift gate 5 Transfer gate 8,9 CCD register 11,12 CCD register 13 First pixel column ODD output signal 14 First pixel column EVEN output signal 15 Second pixel column ODD output signal 16 Second pixel column EVEN output signal 17 Third pixel column ODD output signal 18 Third pixel column EVEN output signal 19 Output circuit 20 Light-shielding aluminum CK1, 2 register

Claims (3)

[Claims] 1. A plurality of pixel columns having a plurality of pixels, a signal reading unit for reading a signal from each pixel column, an output from a pixel in a pixel column among the plurality of pixel columns, Solid-state image pickup device comprising: a calculation unit that corrects the output in accordance with an output from a pixel corresponding to the certain pixel in the pixel column. 2. The solid-state imaging device according to claim 1, wherein the plurality of pixel columns and the signal reading means are integrated on the same semiconductor substrate. 3. When reading a signal from each of a plurality of pixel columns having a plurality of pixels, an output of a pixel in a pixel column of the plurality of pixel columns is output from another pixel column in the other pixel column. A signal processing method for a solid-state imaging device, comprising: performing correction according to an output of a pixel corresponding to a certain pixel.