JPS5853831B2 - Color solid-state imaging device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a color solid-state imaging device that obtains a color television signal using one two-dimensional solid-state imaging device.

The human eye has a high resolving power for achromatic colors, so when trying to obtain a color television signal with a single image sensor, it is necessary to use achromatic light or colored light corresponding to the luminous sensitivity curve. In order to obtain high resolution for green light that is close to a curve, the number of pixels that are sensitive to these color lights that require high resolution is changed to be sensitive to two other types of color light (for example, red and blue or yellow and cyan). Regarding a color image sensor having more than the number of each pixel, Japanese Patent Application Laid-Open No. 51-11
It is known as 2228 (color image sensor).

Generally, in order to obtain a color television signal with a single image sensor, the image of the subject is transferred to the photocathode through a stripe filter that separates the light into red, green, and blue light, which correspond to the three primary colors. The most common method is to form an image and convert it into an electrical signal.

(eg US Pat. No. 2,634,328). However, with this method, sufficient image quality cannot be obtained unless the width of each stripe filter is made fine and a high-resolution image sensor is used.

Simply thinking, this requires three times the resolution of a monochrome camera.

In particular, such a method is not sufficient when high resolution cannot be obtained, such as with a solid-state image sensor.

For example, when using an image pickup tube as a countermeasure,
- By superimposing two types of stripe filters in an intersecting manner as in the 8699 (frequency-separated single image tube color television camera), - a certain regularity can be achieved with a spatial phase difference of 1800 for each scanning line. A method is used in which electrical signals corresponding to each of the three source colors are obtained from a time-delayed signal corresponding to a scanning line and a non-delayed signal by arranging the color filters in such a manner that the color filters are arranged in a manner similar to that of the scanning line.

If this method is adopted, the band of the luminance signal can be obtained up to the characteristics of the image pickup tube, and if the image resolution of the image pickup device is equivalent to that of a monochrome camera, there will not be much of a problem.

However, in this type of color filter, the shape of the single smallest filter element is trapezoidal, and if the image pickup tube has a certain thickness for the scanning beam, the beam will also be irradiated to areas of other colors, thus avoiding color mixture. I can't.

The above-mentioned Japanese Patent Laid-Open No. 51112228 applied the concept of such a cross-shaped color filter to a solid-state image sensor.

The mosaic shape of the color filter is due to the rectangular shape of the pixel of the solid-state image sensor, and the mosaic shape is necessary to prevent color mixing.

As shown in Japanese Unexamined Patent Publication No. 112228/1983, color filters are arranged in a mosaic pattern in which pixels sensitive to colored light, which require high resolution, are arranged alternately in each row in the row direction. In the column direction, they are arranged alternately every scanning line,
Pixels that are sensitive to the other two types of colored light that do not require high resolution are placed between the pixels that are sensitive to the colored light that requires high resolution.In the same row, pixels that are sensitive to the same color light are placed. The method of arranging pixels sensitive to different colored lights in each column in the column direction, and the above-mentioned characteristics regarding the pixels sensitive to colored lights that require high resolution as shown in FIG. It is similar to 112228/1972, but the other two types of pixels that are sensitive to colored light that do not require high resolution and are placed in the gap are generally arranged alternately in the row direction. Conceivable.

In the former case, a signal processing method is used in which a signal corresponding to colored light that does not require high resolution, which is missing in each row in the column direction, is compensated for with a signal from the previous row.

In this method, there is a spatial shift in the column direction between the signal corresponding to color light that requires high resolution and the other two types of signals corresponding to color light that do not require high resolution, and Even between two types of color light that do not require color, there will be a spatial shift in the column direction by one row, and when the signal obtained from this method is displayed as an image on a monitor, etc., the color will be shifted vertically (column direction) at the outline. This will result in misalignment, which will be very unsightly.

In so-called home VTRs, a method is employed in which the color signal is delayed by a certain number of scan lines and combined with a non-delayed signal to eliminate crosstalk between recording tracks.

Therefore, when a color television signal obtained using the color filter of JP-A-51-112228 is recorded on the above-mentioned home VTR, the reproduced image will not only have the color shift that occurs in the imaging device, but also the color shift that occurs in the home VTR. are superimposed, promoting image quality deterioration.

The same applicant has previously proposed a color solid-state imaging device that eliminates these conventional drawbacks and reduces aliasing distortion due to spatial sampling using the image array of solid-state imaging (Japanese Patent Application No. 53129244).

This uses the mosaic filter shown in Figure 1A to generate a video signal (hereinafter referred to as the original signal) obtained from the solid-state image sensor.
Then, a color signal corresponding to each color of the color filter is separated from an IH delayed signal obtained by delaying this video signal by a time equivalent to one horizontal scanning period (hereinafter referred to as IH), and gain adjustment and gamma adjustment are performed for each color signal. Although signal processing such as correction is performed to produce a color television signal using a color encoder, there are two types of colored light that do not require high resolution, for example, Fig. 1B.
To separate the color signals corresponding to R and B, that is, the red signal and the blue signal shown in , first, a signal in which the red signal and the blue signal are continuously and alternately present is created, and this signal is separated by a time corresponding to one pixel. By mixing the delayed signal, aliasing distortion caused by spatial sampling by the pixel array of the solid-state image sensor is reduced, and then the red signal is gated through a gate circuit that gates only the red signal, and only the blue signal is gated. The green signal is separated through a gate circuit.

In this device, two methods are proposed to obtain a signal in which a red signal and a green signal are successively and alternately present.

That is, the first method consists of a first gate circuit that gates only the signals corresponding to the red signal and the green signal from the original signal, and a second gate circuit that gates only the signals corresponding to the red signal and the green signal from the IH delayed signal. In this method, the output signal of the first gate circuit and the output signal of the second gate circuit are mixed to obtain a signal in which a red signal and a blue signal are continuously and alternately present.

The second method has a gate circuit that gates only the red and green signals from the original signal, and mixes the output signal of this gate circuit with a signal obtained by IH delaying this signal to produce a continuous red and blue signal. ≠ to obtain alternating signals.

In the color solid-state imaging device using the first method, four gate circuits are required to gate each pixel in order to obtain a red signal and a green signal. In the device, there are three gate circuits, one less than in the first method, but an IH delay line is additionally required.

For example, if the number of pixels in the horizontal direction of the +f solid-state image sensor is approximately 400 in the output signal from the solid-state image sensor, a signal for each pixel will exist on the -horizontal scanning line approximately every 130 ns.

Therefore, the above-mentioned gate circuit needs to operate with a gate pulse having a pulse width of about 130 ns.

This is approximately 7.8MHz when converted into a frequency,
In order to construct a gate circuit that operates with such a high frequency gate pulse, the circuit becomes complicated and it is necessary to use expensive parts.

Furthermore, since the gating period is within the effective scanning period, switching noise caused by the gating mixes into other video signals through the power supply line and ground line and appears as synchronous noise on the playback screen, significantly degrading the image quality. Therefore, the fewer the number of gate circuits gated within the effective scanning period, the less deterioration of image quality due to noise will occur.

SUMMARY OF THE INVENTION An object of the present invention is to alleviate these drawbacks and provide an inexpensive color solid-state imaging device with good image quality.

According to the present invention, photoelectric conversion elements having the first spectral characteristic are arranged every other pixel in the row direction, alternately arranged every predetermined scan line in the column direction, and the photoelectric conversion elements having the first spectral characteristic are arranged alternately every other pixel in the column direction. A two-dimensional solid-state image sensor in which photoelectric conversion elements having a second spectral characteristic and photoelectric conversion elements having a third spectral characteristic are arranged alternately, and an output signal from the two-dimensional solid-state image sensor is arranged in a constant scan line. Using a delay line that delays the corresponding time and a pulse created from a clock pulse for driving the two-dimensional solid-state image sensor, only the electrical signal corresponding to the photoelectric conversion element having the first spectral characteristic is transmitted to the two-dimensional solid-state image sensor. a circuit that separates the output signal from the image sensor and the output signal from the delay line; and first and second circuits that switch between the output signal from the two-dimensional solid-state image sensor and the output signal from the delay line for every fixed scan line. The output signals of the first and second switch circuits are different for each fixed scanning line, and the output signal of the first switching circuit is different for each fixed scanning line. When the output signal of the element (the output signal of the delay line) is the output signal of the second switch circuit, the output signal of the second switch circuit becomes the output signal of the delay line (the output signal of the two-dimensional solid-state image sensor). and a gate pulse generation circuit that generates a gate pulse for driving the second switch circuit; a delay line that delays the output signal of the first switch circuit by a time corresponding to one pixel; a mixing circuit that mixes a delayed signal delayed by a time corresponding to the output signal of the second switch circuit; and an electrical signal corresponding to the photoelectric conversion element having second and third spectral characteristics from the output signal of the mixing circuit. A color solid-state imaging device is obtained, characterized in that it is equipped with a circuit that separates the .

The present invention will be described below with reference to the mosaic filter 1' of FIG. 1B.
An embodiment of the present invention using FIG. 2, FIG. 3, and FIG. 4 will be described below.

Figure 2 is a block diagram of a color solid-state imaging device, Figure 3,
FIG. 4 is a waveform diagram of each part.

The output signal of the color solid-state image sensor 8 is applied to an IH delay line 10 that delays this signal by IH, and a delay line 9 that delays a minute time input signal.

This delay line 9 is a delay line for matching the phase of the output signal of the IH delay line 10 (hereinafter referred to as IH delay signal 28) and the output signal of the color solid-state image sensor 8.

Therefore, it is not necessary to insert the delay line 9 if the IH delay signal 28 is exactly IH delayed with respect to the input signal of the IH delay line 10, or even if it is not exactly IH delayed, it is within the tolerance range.

Also, when the delay time of the IH delay line 10 is shorter than the IH and is not within the tolerance range, the delay line 9 is inserted immediately before or after the IH delay line 10, instead of at the position shown in the block diagram of FIG. There is a need.

In the present invention, by inserting the delay line 9 at the position shown in the block diagram of FIG. 2, the following description will be made assuming that the phase of the IH delay signal 28 and the output signal of the delay line 9 (hereinafter referred to as the original signal 27) match. .

Assuming that the video information obtained from the array in a certain row (referred to as the Nth row) in FIG. The image information obtained from the array in the previous row of N rows (N-1st row) is obtained by the N-1st scan, and word signals 27 as shown in FIGS. 3A and 4A are obtained. It will be done.

Here, the signals on the left half of FIGS. 3A-E and 4A-L represent the signals on the N-1st scanning line, and the signals on the right half represent the signals on the Nth scanning line. , these signals are repeated in all scanning lines.

IH delayed signal 28 is shown in FIGS. 3B and 4B.

That is, the video information of the N-2th line (N-1st line) is obtained on the N-1st (N-th) scanning line.

In these figures 3 and 4, the red signal of the original signal 27 is shown as R1, the green signal is shown as G, the blue signal is shown as B, and IH
The red signal of the delayed signal 28 is R', the green signal is missing, and the green signal is B
’.

The original signal 27 and the IH delayed signal 28 are sent to the green signal separation circuit 11
and is supplied to the switch circuit 12.

In the green signal separation circuit 11, the original signal 27 of FIG. 3A is gated with the gate pulse of FIG. 3C, and the IH signal of FIG. 3B is
A continuous green signal (G signal) as shown in Fig. 3 E is obtained by mixing the delayed signal 28 with the gate pulse shown in Fig. 3.
get.

This green signal is converted into a continuous video signal through a low frequency converter 22 whose cutoff frequency is the Nyquist frequency (1/2 of the horizontal clock frequency that drives the color solid-state image pickup plate 8), and the phase with the red signal and blue signal is The signal is sent to a color encoder 25 through a delay line 23 for adjusting the signal, and a process amplifier 24 that performs video signal processing such as gamma correction as a camera device.

The original signal 21 and the IH delayed signal 28 supplied to the switch circuit 12 are converted into two types of signals by two switch circuits, in which the original signal 27 and the IH delayed signal 28 are exchanged for each IH, that is, FIG. 4E and FIG. 4 Figure F.

The switch circuit 12 includes a first switch circuit 29 consisting of an input terminal a, an input terminal S, and an output terminal C, and a second switch circuit 30 consisting of an input terminal d, an input terminal e, and an output terminal f. It consists of two switch circuits.

These switch circuits switch for each IH, so
Since the frequency of the gate pulse that drives these switch circuits is very low at approximately 7.8 MHz, an analog switch using a low frequency switching element such as a MOS transistor may be used.

In addition, the input terminal a (input terminal b) of the first switch circuit 29
) is the input terminal e (input terminal d) of the second switch circuit 30
), and the IH delay signal 28 (original signal 27)
is applied.

FIG. 4 C and D created using the gate pulse generation circuit 13
The first switch circuit 29 and the second switch circuit 30 are driven by the gate pulse shown in .

That is, during the period corresponding to the N-1st, N+1st, N+3rd, . . . , scanning lines due to the gate pulse in FIG. a
and output terminal C are short-circuited, and input terminal d and output terminal f of the second switch circuit 30 are short-circuited.

Similarly, by the gate pulse shown in the fourth diagram, the Nth, Nth
During the period corresponding to the +2nd, N+4th, . and the output terminal f are short-circuited.

That is, in the video signals obtained at the output terminal C of the first switch circuit 29 and the output terminal f of the second switch circuit 30 of the switch circuit 12, the original signal 27 and the IH delayed signal 28 are exchanged for each IH. The video signal obtained at the output terminal C is the signal shown in FIG. 4F, and the video signal obtained at the output terminal f is the signal shown in FIG. 4E.

At this time, since switching for each IH is performed within the horizontal blanking period, switching noise due to switching does not appear on the playback screen.

The video signal obtained at the output terminal C is supplied to a one-pixel delay line 14 which delays the signal by a time corresponding to one pixel, resulting in a one-pixel delayed signal shown in FIG.

This mixing circuit has an output terminal f of the second switch circuit 30.
The video signal obtained in the above is also supplied and mixed with the one-pixel delayed signal described above, and a mixed signal of the original signal 27 and the IH delayed signal 28 as shown in FIG. 4H is obtained at the output of the mixing circuit 15. It will be done.

The signals shown in FIG. 4H are repeated signals of B+B', G+σ, R+R', and G+ for each scanning line, except for the signal of the first pixel.

Therefore, by gating each color signal every four pixels from this repeated signal, it is possible to separate it into a signal of only B+B' and a signal of only R+R'.

That is, the output signal of the mixing circuit 15 is supplied to the color separation circuit 16 and the color separation circuit 17, and gated by the red signal gate pulse shown in FIG. 4J and the blue signal gate pulse shown in FIG. A red signal shown in FIG. 4 is obtained as an output, and a blue signal shown in FIG. 4K is obtained as an output of the color separation circuit 17.

The color separation circuit 16 and the color separation circuit 17 may be the above-mentioned gate circuits or may be sample and hold circuits.

The output of the color separation circuit 16 is at the Nyquist frequency of the red signal (1/of the horizontal clock frequency that drives the color solid-state image pickup plate 8).
8) is sent to the encoder 25 via the process amplifier 20.

=The output of the old color separation circuit 17 is sent to the encoder 25 via the process amplifier 21 through a low-pass filter 19 whose cutoff frequency is the Nyquist frequency of the green signal.

The encoder 25 combines the sent red, green, and blue signals into a color television signal, and this signal is taken out from the output terminal 26.

As described above, according to the present invention, in order to obtain color signals that do not require high resolution, only one gate circuit for each pixel is required for each color signal, which simplifies the circuit. This makes it possible to realize an inexpensive color solid-state imaging device with good image quality.

As an embodiment of the present invention, a case has been described in which the color mosaic filter shown in FIG. 1 is made to correspond to a solid-state image sensor. It is clear that the same holds true for cases with different characteristics.

[Brief explanation of the drawing]

FIGS. 1A and 1B are diagrams for explaining a color mosaic filter that can be used in the present invention, FIG. 2 is a block diagram for realizing the color solid-state imaging device of the present invention, and FIGS. 3A to 3E and FIGS. 4A to 4L are operation waveform diagrams for explaining the block diagram of FIG. 2. 1.1': Color mosaic filter, 2D filter, 3: White filter, 4: Yellow filter, 5:
Blue filter, 6: Green filter, 7: Red filter, 8: Color solid-state image sensor, 9: Delay line, 10: IH delay line, 11: Green signal separation circuit, 12: Switch circuit, 13:
Gate pulse generation circuit, 14 pixel delay line, 15: Mixing circuit, 16° 17: Color separation circuit, 1B, 19: Low pass filter, 20, 21, 24: Process amplifier, 22: Low pass filter device, 23: delay line, 25: color encoder, 26
: Output terminal, 27: Original signal, 28: IH delay signal, 29
: first switch circuit, 30: second switch circuit.

Claims (1)

[Claims] 1 In the odd rows (or even rows), photoelectric conversion elements having the first spectral characteristic are arranged every other pixel in the row direction, and photoelectric conversion elements having the second spectral characteristic and the third spectral characteristic are arranged for the remaining pixels. Photoelectric conversion elements having spectral characteristics are arranged alternately, and in even numbered rows (or odd numbered rows), photoelectric conversion element groups having the same arrangement as the odd numbered rows are arranged with a phase difference of 900 with respect to the odd numbered rows. A two-dimensional body image sensor, a delay line that delays an output signal from the two-dimensional body image sensor by a time corresponding to a certain scanning line, and a pulse created from a clock pulse for driving the two-dimensional body image sensor. The first
a circuit that separates only an electrical signal corresponding to a photoelectric conversion element having spectral characteristics from an output signal from the two-dimensional body image sensor and an output signal of the delay line; and an output from the two-dimensional body image sensor. first and second switch circuits that switch between the signal and the output signal of the delay line for each fixed scan line, and the output signals of the first and second switch circuits are different for each fixed scan line, and When the output signal of the first switch circuit is the output signal of the two-dimensional body image sensor (the output signal of the delay line) at any fixed scan line, the output signal of the second switch circuit is the output signal of the delay line. a gate pulse generation circuit that generates a gate pulse for driving the first and second switch circuits so that the output signal (output signal of the two-dimensional body image sensor) is obtained;
a delay line that delays the output signal of the second switch circuit by a time corresponding to one pixel; a mixing circuit that mixes the delayed signal delayed by the second switch circuit by the delay line with the output signal of the second switch circuit; A color solid-state imaging device comprising a circuit that separates electrical signals corresponding to photoelectric conversion elements having second and third spectral characteristics from the output signal of the mixing circuit.