JP2000358250A - Color image pickup device and image pickup system using it - Google Patents

Abstract

PROBLEM TO BE SOLVED: To photograph a picture with high definition and a moving picture with low resolution in excellent image quality by permitting the line interval of pixels where a first color filter is arranged to be narrower than that of the pixels where the second and the third color filters are arranged. SOLUTION: A G filter is arranged in a half of a checkered pattern and an R filter and a B filter are respectively arranged by half in the remaining half. In the case of high definition reading(system 1), the whole pixels are read and chrominance signals are generated from the two lines. In the case of low resolution reading(system 2), addition and thinning-reading are executed. The chrominance signals are generated from the four lines. A G signal is obliquely added from the two center lines among the four ones and the R and B signals are added in a vertical direction and in a horizontal direction from the four lines. Besides, the line interval of the pixels to be added, where the G filter is arranged, is narrower than that of the pixels where the R and B filters are arranged. Thus, the G signal becomes the one with higher resolution than the others and a high band luminance signal is generated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color image pickup apparatus and an image pickup system using the same, and more particularly, to a color image pickup apparatus capable of selecting an operation of reading each pixel signal with an original signal and an operation of adding and reading each pixel signal. The present invention relates to an imaging system using the same.

[0002]

2. Description of the Related Art Digital still cameras have come to use an image sensor having 2 million pixels. This is the result of pursuing silver-lead photographic image quality and was used exclusively for still images. A conventional NTSC video camera has about 400,000 pixels, and the reading speed is about 13.5 MHZ in interlaced scanning.
z, about 27 MHz for progressive scanning.

When an image sensor having 2 million pixels is used for a moving image, the reading speed is five times as high as that in the case of 400,000 pixels.

[0004] When reading is performed at such a reading speed, power consumption is greatly increased, noise is degraded due to the increased power consumption, and the cost is increased due to an increase in an image processing memory. there were.

To solve such a problem, there is a color image pickup apparatus disclosed in Japanese Patent Application Laid-Open No. 9-247689. In the embodiment shown in the publication (FIG. 3 of the publication),
The same color is thinned out and read and added in units of 4 × 4 pixels.

[0006]

In this case, the problem is that the effective pixels used in 4 × 4 pixels are １ ／, and the number of effective pixels is 1/16.
It is becoming. Therefore, in the case of an element having 2 million pixels, only a resolution equivalent to 2,000,000 / 16 ≒ 125,000 pixels can be obtained. In other words, the utilization efficiency becomes very poor, and practically only the monitor can be used.

Further, in the embodiment of Japanese Patent Application Laid-Open No. 9-247689 (FIG. 2 of the Japanese Patent Publication), it is described that a plurality of pixel signals are mixed and read out. (It is difficult to read XY scanning because of charge transfer in the CCD), and there is a problem that the KTC noise of the vertical signal line is large and a good S / N cannot be obtained when using a semiconductor switch and a photodiode.

As described above, in the prior art, pixel signals are thinned out, and reading is performed in units of 4 × 4 pixels, so that there is a problem that a sufficient resolution cannot be obtained and S / N is poor.

In the case of reading signals from pixels, conventionally, pixel signals in the first row are transferred to an upper memory, and pixel signals in a second row are transferred to a lower memory. Therefore, addition in the horizontal direction was easy, but addition in the vertical and oblique directions was difficult.

An object of the present invention is to solve the problems of the prior art, and to provide a color image pickup apparatus capable of shooting a high-definition image and a low-resolution moving image with good image quality, and an image pickup system using the same. And

[0011]

A color image pickup apparatus according to the present invention has a plurality of pixels arranged in a matrix, and an image pickup device provided with a plurality of color filters for the plurality of pixels, and an image pickup device of the same color. A color image pickup device comprising: an addition unit that adds a signal from a pixel provided with a color filter, wherein a line interval between the pixels provided with the first color filter is set to a second value.
This is a color imaging device that is narrower than the line spacing of the pixels on which the color filters of the third and third colors are arranged.

An image pickup system according to the present invention includes the above color image pickup apparatus according to the present invention, an optical system for imaging light onto the color image pickup apparatus, and a signal processing circuit for processing an output signal from the color image pickup apparatus. It is characterized by the following.

[0013]

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

FIG. 1 is a schematic explanatory view showing a pixel signal reading method by the color image pickup apparatus of the present invention. In FIG. 1, the output of the image sensor has four channels (four outputs), the color filters of the pixels arranged in a matrix of the image sensor are arranged in a checkered pattern, and the G (green) filter is half of the checkered pattern. R (red) filter, B (blue)
The filters are arranged in half on the other half of the checkered pattern.

In the case of high-definition readout (system 1), each pixel signal is read out independently and all pixels are read out. That is, the output A outputs pixel signals G11, G13, G15,... By the readout circuit 11, and the output B outputs pixel signals G22, G24, G26,.
Are output from the output C, and the readout circuit 12 outputs pixel signals B21, B23, B25,..., And the output D outputs pixel signals R12, R14, R16,. Then, a color signal is formed from two lines (for example, the V1 line and the V2 line).

In the case of low-resolution reading (system 2)
In addition, addition and thinning-out reading are performed. In system 2, color signals are formed from four lines. G (green) signal is 4
Obliquely add from the center two lines in the line, and R (red)
The signal and the B (blue) signal perform vertical addition and horizontal addition from four lines. Interlace scanning is performed between fields by changing the combination of four lines. That is, system 2
In the even field, four lines (for example, V
1, V2, V3, and V4 lines), and a color signal is formed from the output A by the readout circuit 11 to the pixel signals G31 + G22,
G33 + G24,... Are output, and from the output C, pixel signals B21 + B41 + B23 + B43, B25 + B
45 + B27 + B47,... Are output, and from the output D, the pixel signals R12 + R32 + R14 + R34, R16
+ R36 + R18 + R38,... Are output. 4 lines in odd field (for example, V3, V4, V5, V6 lines)
, A color signal is formed, and the output A
Output the pixel signals G51 + G42, G53 + G44,... From the output C by the readout circuit 12
+ B61 + B43 + B63,... Are output, and from the output D, pixel signals R32 + R52 + R34 + R54,
... is output. Note that the line interval of the pixel on which the G filter (to be the first color) to be added is arranged is the second interval.
It is smaller than the line spacing of the pixels where the R and B filters for the color and the third color are arranged. With such an arrangement, the G signal becomes a signal having higher resolution than other colors, and a high-frequency luminance signal can be generated.

With the above configuration, the G signal is vertical 5
A resolution of 00 lines is obtained.

Next, a configuration example of the pixel portion will be described. The pixel portion is configured by arranging pixels called CMOS sensors in a matrix.

FIG. 2 is a circuit diagram showing a CMOS sensor and a readout circuit. Since CMOS sensors have variations in each pixel amplifier and reset noise in the gate section,
In order to remove the noise, a signal memory CT1 is provided at the output unit.
And a noise memory CT2 to remove noise by subtraction processing.

In FIG. 2, a broken line area indicates one pixel of a CMOS sensor, PD is a photodiode, MTX is a transfer transistor, MRES is a reset transistor,
MSEL is an amplification transistor serving as a pixel amplifier, and MSEL is a selection transistor for selecting a pixel. After the reset transistors MRES and MRV are turned on to reset the pixel portion and the vertical output line, a noise signal is accumulated in the noise memory CT2 via the pixel amplifier, the selection transistor MSEL, and the transistor MCT2. Further, the transfer transistor MTX is turned on, and the signal photoelectrically converted from the photodiode PD is transferred to the gate of the amplification transistor MSEL serving as a pixel amplifier, and the signal is transmitted via the pixel amplifier, the selection transistor MSEL, and the transistor MCT1. A signal including a noise signal component is stored in the memory CT1. Then, the signal including the noise signal component accumulated in the signal memory CT1 and the noise signal accumulated in the noise memory CT2 are output to a horizontal output line, subjected to a subtraction process, and the variation of the pixel amplifier and the reset of the gate unit. A signal from which noise components such as noise have been removed is obtained. φSEL, φTX, φRES, φRV,
φTS and φTN are an amplification transistor MSEL, a transfer transistor MTX, and a reset transistor MRES, respectively.
MRV is a control signal for controlling the transistors MCT1 and MCT2. The transistor ML is a load of the pixel amplifier MSF. .phi.L may be driven in common with .phi.SEL, or may be always set to the H level as a resistor.

FIG. 3 is a block diagram showing a configuration of a signal readout circuit of the image pickup apparatus of the present invention. The noise elimination configuration described with reference to FIG. 2 is omitted here for simplification of description.

On the upper side of the pixel portion, two lines of G memories MG1, MG2,... Are provided, and on the lower side, two lines and two horizontal lines of B and R memories MB11, MB12, MB31, MB32,.
.., MR21, MR22, MR41, MR42,. Pixel signals are read from the pixel unit to the upper and lower memories. The upper memory is a horizontal scanning circuit (H-SR) 21
Is controlled by the horizontal scanning circuit (H · S).
R) 22. Reading of signals from the pixel unit is controlled by a vertical scanning circuit (V · SR) 23.

The addition of signals is performed as follows. In the upper memory (memory for G), the adjacent two
Control is performed so that signals for pixels are added on a memory or on a horizontal signal line. For example, the signal transferred to memory MG1 and the signal transferred to memory MG2 are added, and the signal transferred to memory MG3 and the signal transferred to memory MG4 are added.

Lower memory (memory for B and memory for R)
Is controlled so that signals of four pixels are added by the addition pulse φadd. For example, each of the memories MB11, MB
B12, the signals transferred to MB31 and MB32 are added,
The signals transferred to the memories MR21, MR22, MR41, MR42 are added together.

FIG. 4 is a circuit configuration diagram showing a more detailed configuration of the signal readout circuit, FIG. 5 is a timing diagram for reading all pixel signals, and FIG. 6 is a timing diagram for addition and thinning-out readout.

First, the case of reading all pixel signals will be described with reference to FIGS.

As shown in FIG. 5, in a period t0, the readout circuit is reset by setting the control signals φTG1, φTG2, φTB1, φTB2, φTR1, φTR2, φRV to the H level.

Next, in a period t1, the V1 line is selected.
When the control signals φTG1, φTR1, and φTR2 are set to H level, the pixel signal G11 is transferred to the upper memory MG1, and the pixel signal R12 is transferred to the lower memories MR21 and MR22. In the lower memory, the two memories MR21 and MR22 are commonly used to increase the read gain from the memory to the horizontal output line.

Next, in the period t2, the V2 line is similarly selected, and when the control signals φTG2, φTB1, and φTB2 are set to the H level, the pixel signal G22 is sent to the upper memory MG2 and the pixel signal B21
Is transferred to the lower memories MB11 and MB12. In the lower memory, the two memories MB11 and MB12 are commonly used to increase the read gain from the memory to the horizontal output line.

Next, a control signal φH11 is output from the horizontal scanning circuit 21.
, And φH21, φH12 and φH22, φH13 and φH23,... Are sequentially output simultaneously, and signals are transferred from each upper memory to two horizontal output lines and output from outputs A and B. Control signal φ
While H11 and φH21, φH12 and φH22, φH13 and φH23,... Are output, φHC is at the H level, and the horizontal output line is reset. Although not shown, similarly, from the horizontal scanning circuit 22, control signals φH11 ′ and φH21 ′, φH12 ′ and φH22 ′ having the same phase as the control signals φH11 to φH13, φH21 to φH23,
are simultaneously output sequentially, signals are transferred from each lower memory to two horizontal output lines, and output from outputs C and D. As a result, signals in units of 2 × 2 pixels are output from outputs A, B, C, and D. Thereafter, similarly, the V3 line and the V4 line are selected, and the signal is read out.

FIG. 4 shows addition and thinning-out reading.
This will be described with reference to FIG. Here, the case of the even field (Feven) will be described. However, in the case of the odd field (Fodd), addition and thinning-out reading can be performed by the same operation.

As shown in FIG. 6, in a period t0, the control signals φTG1, φTG2, φTB1, φTB2, φTR1, and φTR2 are set to the H level to reset the read circuit.

Next, during the period t1, φTR1 goes high, and the pixel signals R12, R14,.
Are transferred to the lower memories MR21, MR41,..., And the G signal is not transferred to the memories.

In the period t2, φTG2 and φTB1 become H level, and the pixel signals G22 and G2, which are the G signals of the V2 line.
Are transferred to the upper memories MG2 and MG4, and the B signals B21, B23,... Are transferred to the lower memories MB11 and MB31.

In the period t3, φTG1 and φTR2 become H level, and the pixel signals G31 and G3, which are the G signals of the V3 line.
Are transferred to the upper memories MG1, MG3, and the R signals R32, R34,... Are transferred to the lower memories MR22, MR42.

In the period t4, φTB2 becomes H level,
The pixel signals B41, B43,... Which are the B signals of the V4 line are transferred to the lower memories MB12 and MB32, and the G signals are not transferred to the memories.

During period t5, φadd is applied, and the G signal is added between adjacent upper memories. That is, G
31 + G22, G33 + G24,... Are added. In the lower memory, a signal in the vertical direction is added by the memory by φadd. That is, the addition process of B21 + B41, B23 + B43,... And the addition process of R12 + R32, R14 + R34,. Further, the signal (B21 + B41) and the signal (B
23 + B43), signal (R12 + R32) and signal (R14 + R34)
Are added at the horizontal output line, and the signal (B21 + B41 + B23 +
B43), and a signal (R12 + R32 + R14 + R34) is obtained.

In the timing chart of FIG. 6, a signal is output to the output A of the horizontal shift pulse by application of φH1n, and φH2n is L
The ow state is maintained, and no signal is output to the output B. ΦH1n 'and φH2n' are applied to the lower memory signal in phase.
Signals added in the vertical and horizontal directions (signals added by the lower memory and the horizontal output line) are output to outputs C and D.

FIG. 7 is a diagram showing a pixel use area of each color when obtaining an addition signal of R, G, and B pixels when interlace driving is performed. In the even field and the odd field, the R pixel and the B pixel are partially used in common, but the G pixel is not commonly used. The signal from the G pixel is obtained by adding two adjacent pixels in the oblique direction, thereby improving the horizontal resolution and the vertical resolution.

In the embodiment described above, color signals are formed from four lines in addition and thinning-out reading. However, the present invention is not limited to this. For example, color signals are formed from eight lines. be able to. FIG. 8 is a diagram showing a pixel use mode when eight pixel rows are used. This embodiment is obtained by adding pixel signals in more lines to the embodiment of FIG. In the figure, for the R pixel, nine R pixels surrounded by circles in the figure are added from the V3, V5, and V7 lines, and for the G pixel,
Six G pixels surrounded by □ in the figure are added from lines 4 and V6, and nine B pixels surrounded by △ in the figure are added from B2, V4 and V6 lines for B pixels. Of the eight lines, the B and R pixels are selectively added from among the five lines, and
Pixels are selectively added from three lines.

In the embodiments described above, the color filters using G, B, and R colors have been described. However, other than this, the color filters for luminance signals requiring high resolution may be used. What is necessary is to make the line interval of the addition of the pixels in which the filters are arranged narrower than the line interval of the addition of the pixels in which the color filters of the other colors are arranged.

FIG. 9 shows a schematic diagram of a system using the above-mentioned imaging apparatus. As shown in the figure, the image light incident through the optical system 71 forms an image on the CMOS sensor 72. CM
The optical information is converted into an electric signal by the pixel array arranged on the OS sensor 72. The electric signal is subjected to signal conversion processing by a signal processing circuit 73 by a predetermined method and output. The signal processed signal is
The information is recorded or transferred by an information recording device by a recording system and a communication system 74. The recorded or transferred signal is reproduced by the reproduction system 77. CMOS sensor 7
2. The signal processing circuit 73 is controlled by a timing control circuit 75, and the optical system 71, the timing control circuit 75, the recording / communication system 74, and the reproduction system 77 are controlled by a system control circuit 76. The timing control circuit 75 can select either independent reading or addition / thinning-out reading.

The horizontal and vertical drive pulses are different between the above-described high pixel readout (all pixel readout) and low pixel readout (addition / thinning-out readout). Therefore, it is necessary to change the drive timing of the sensor, the resolution processing of the signal processing circuit, and the number of recording pixels of the recording system for each reading mode. These controls are performed by the system control circuit 76 in accordance with each read mode. In addition, in the reading mode, the sensitivity differs depending on the addition. For example, the signal amount is doubled in addition reading in comparison with high pixel reading. As it is, the dynamic range is 1/2
Therefore, an appropriate signal is obtained by controlling the aperture (not shown) to be smaller by half the aperture. As a result, at low illuminance, photographing can be performed up to half the brightness. The signal processing circuit and the recording system may be separately provided for high definition and for moving images.

[0044]

As described above, according to the present invention, it is possible to record and display by switching between a high-definition image and a lower-resolution image by selecting all-pixel independent reading and addition and thinning-out reading for pixel signal reading. Can be.

Among the pixel signals of a plurality of lines, a signal of the first color (for example, a G signal) is added in an adjacent line, and signals of other colors are thinned out. Was obtained. Since the signal of the second color and the signal of the third color (for example, the R signal and the B signal) were added in the vertical direction and the horizontal direction, the moire in the vertical and horizontal directions could be greatly reduced. Also, S / N is improved by adding signals,
In the low-pixel-number readout, the driving frequency could be as low as that of the NTSC image sensor, so that low power consumption was achieved.

[Brief description of the drawings]

FIG. 1 is a schematic explanatory view showing a pixel signal reading method by a color imaging device of the present invention.

FIG. 2 is a circuit diagram showing a CMOS sensor and a readout circuit.

FIG. 3 is a block diagram illustrating a configuration of a signal readout circuit of the imaging device of the present invention.

FIG. 4 is a circuit configuration diagram showing a more detailed configuration of the signal readout circuit.

FIG. 5 is a timing chart of reading all pixel signals.

FIG. 6 is a timing chart of addition and thinning-out reading.

FIG. 7 shows R, G, and R values when interlace driving is performed.
It is a figure which shows the pixel use area | region of each color when obtaining the addition signal of B pixel.

FIG. 8 is a diagram illustrating a pixel use mode when eight pixel rows are used.

FIG. 9 is a schematic diagram showing a system according to the present invention.

[Explanation of symbols]

 11, 12 readout circuit 21, 22 horizontal scanning circuit 23 vertical scanning circuit

 ────────────────────────────────────────────────── ─── Continuing on the front page (72) Inventor Masato Shinohara 3-30-2 Shimomaruko, Ota-ku, Tokyo F-term in Canon Inc. (reference) 5C024 AA01 CA00 DA01 FA01 GA31 GA48 GA51 JA04 5C065 BB48 CC01 DD01 DD15 EE05 EE06 GG49

Claims (8)

[Claims] 1. An image pickup device having a plurality of pixels arranged in a matrix and adding a signal from a pixel provided with a plurality of color filters and a pixel provided with a color filter of the same color to the plurality of pixels. A color image pickup apparatus comprising: an adder that performs a color filter operation, wherein a line interval between pixels on which the first color filter is arranged is smaller than a line interval between pixels on which the second color filter and the third color filter are arranged. Color imaging device. 2. The color imaging apparatus according to claim 1, wherein the pixels provided with the first color filters are added between adjacent lines, and the pixels provided with the second and third color filters are provided. Is a color image pickup device wherein addition is performed between lines arranged every other line. 3. The color image pickup device according to claim 1, wherein the first color filter is a G (green) color filter, the second and third color filters are R (red) color filters, and A color image pickup device comprising a B (blue) color filter. 4. The color image pickup apparatus according to claim 1, wherein said adding means includes:
And a means for adding the signals of the pixels in the oblique direction provided with the color filters and a means for adding the signals of the pixels in the horizontal and vertical directions provided with the second and third color filters. A color imaging device characterized by being performed. 5. The color imaging device according to claim 4, wherein the line to be added is changed for each field. 6. A color image pickup apparatus according to claim 1, wherein: means for independently reading a signal from each pixel; and means for adding and reading a signal from each pixel for each color. A color image pickup apparatus having a switching unit for switching between color images. 7. The color imaging device according to claim 3, wherein a higher-frequency luminance signal is formed than a signal from a pixel provided with a G color filter. 8. A color imaging device according to claim 1, an optical system for imaging light onto said color imaging device, and a signal for processing an output signal from said color imaging device. An imaging system comprising a processing circuit.