JPS5945780A - Solid-state image pickup device - Google Patents

Abstract

PURPOSE:To obtain an excellent picture quality without a line variable density phenomenon, by changing the amplitude of a video signal in the direction where no line variable density is generated at every one or plural horizontal lines' intervals without changing the drak level of a video signal of each horizontal line. CONSTITUTION:If no gain control signal is applied to the base of transistors (TR)s 30, 31, the base potential of TRs 29-32 is the same. Then, the gain control signal is applied to the base of the TRs 30, 31 and the level of the gain control signal is adjusted, then a signal having waveforms e, f is obtained at the collector of the TRs 30, 31. Further, the video signal where the line variable density is corrected is obtained from the output terminal 38 of a differential amplifier 37. If the signal amplitude is changed over plural horizontal lines, a gain control amplifier and a clamp circuit are extended and the number of input terminals of an analog switch is extended to change the control pulse so that the desired switching is attained. Thus, the excellent picture without line variable is density is obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to a television camera using a solid-state image sensor.

Conventional structure and its problems Solid-state image sensors include charge-coupled (CCD) image sensors,
There are several types of image pickup devices, such as X-Y address type (MOS type) image pickup devices, and each type has advantages and disadvantages.

As is well known, a solid-state image sensor is composed of a photoelectric conversion section that converts an optical signal from an object image into an electrical signal, and a scanning section that reads out the optical signal charge obtained by the photoelectric conversion section. Generally, a photodiode is used as the photoelectric conversion element, and a combination of a CCD or a shift register and a MOS switch is used as the scanning section.

FIG. 1 shows a well-known equinodal circuit of a MOS type image sensor. 1
is a photodiode as a photoelectric conversion element, 2 is a MOS-FET as a vertical readout switch, 3 is a vertical signal line, and 4 is a MOS-FET as a horizontal readout switch.
, 5 is a vertical scanning circuit, 6 is a vertical address line for controlling the vertical readout switch, 7 is a horizontal scanning circuit, 8 is a horizontal signal line, and 9 is a preamplifier.

The operation of this solid-state image sensor is performed by the vertical scanning circuit 6. J: A scanning pulse is applied to the gate of MOS-FET 2, which is a vertical readout switch, via the vertical address line 6.
-FET2 all ONσ transfers the optical signal charge obtained by the photodiode to the vertical signal line 3. After that, the scanning pulse of the horizontal scanning circuit 7 is applied to the gate of MOS-FET4, which is a horizontal readout switch, and the MOS-FET
4 are sequentially connected to the preamplifier 9 via the horizontal signal line 8.
The optical signal charge is guided to the preamplifier output terminal and taken out as a video signal. Here, a photodiode 1 as a photoelectric conversion element and a MOS-FET 2 as a vertical readout switch are manufactured as an integrated structure.

FIG. 2 shows a basic structural plan view of one element of the light receiving section in this device. In FIG. 2, 1o is a photodiode, 11 is a drain, 12 is a gate, 13 is a vertical address line, and 14 is a vertical signal line. these 10
, 11.12 constitute one MOS-FET, and the photodiode 1o serves as the source. Also,
The drain 11 is coupled to a vertical signal line. Generally, the vertical address line 13 is made of polysilicon, and the vertical address line 14 is made of polysilicon.
is an aluminum wire.

A sectional view taken along line XX in FIG. 2 is shown in FIG. 3. In FIG. 3, the silicon substrate is of P type, and a photodiode (source) 10. Impurities are diffused into each drain 11 so that it becomes N type. The gate 12 is placed on the oxide film 16, the vertical signal line 14 is placed on the oxide films i5 and 16, and an oxide film 17 is formed to protect the entire surface of the image sensor.

When a MOS type image sensor with this configuration is used to image the entire subject where the center part a is bright as shown in Figure 4, and the video signal obtained by the image is reproduced on a television receiver, the image as shown in Figure 6 is obtained. Bright area in the center - all outside the top of the screen,
A pseudo-slightly bright portion (a) and (c) are generated in the lower part. This pseudo signal is a smear.

Next, we will discuss the cause of this smear.

As shown in Figure 6, smear occurs equally above (area b) and below (area C) of the bright subject image (area a), but this is due to the optical signal generated by the bright subject image (area a). This occurs because the charge is not captured only in the photodiode in region a, but a portion of the optical signal charge is directly captured in the drain.

In other words, the vertical scanning of the image sensor is performed in the area or area C.
This is because the optical signal charge accumulated in the vertical signal line is read out every time a horizontal scan is performed even when the vertical signal line is scanned. ).

Here, the main reason why optical signal charges are directly captured in the drain will be explained using FIG. 3.

FIG. 3 shows three types of light incident paths that cause smear generation.

In the thread path (2), the light with a long chord included in the incident light penetrates deep into the semiconductor, passes through the photodiode 10, and generates all optical signal charges in the P-type substrate. A portion of this charge is absorbed by the drain 11.

In the thread path (2), since the gate electrode 12 is made of polysilicon, the incident light easily passes through the gate electrode and generates optical signal charges in the P-type substrate. A part of this charge is on the drain 11
absorbed into.

In yarn path (2), multiple reflections of incident light and obliquely incident light are directly incident on the drain 11, and optical signal charges are generated in the drain 11. This charge becomes a trap trapped in the drain 11.

Also, although not due to direct incident light, a part of the optical signal charge generated by the photodiode is laterally diffused through the P-type substrate and absorbed by the drain 11 as a thread (2). As a result, smear signal charges are accumulated in the drain 11.

Therefore, as shown in FIG. 6, false images due to the smear signal occur equally above and below the bright image.

Further, since the vertical address line 13 shown in FIG. 2 is made of polysilicon, incident light easily passes through the vertical address line and generates optical signal charges in the P-type substrate.

A part of this charge is absorbed by the drain 11 as described above,
A part of it is absorbed by the photodiode 10. In Figure 2, for example, the optical signal charge generated by light incident on point d of the vertical address line 13 is transferred to the photodiode 10! L
.

A portion is absorbed by each of 10b, 10C, and 10d. This results in optical crosstalk isometrically for each photodiode, causing a total decrease in the modulation degree in the vertical (column) and horizontal (row) directions. In addition, a part of the optical signal charge is also absorbed by the drain 11, which may cause smearing. Current general solid-state image sensors have approximately 400 pixels in the horizontal direction and 500 pixels in the surface direction, and because they use a % inch-sized optical system, the dimensions of one element (pixel) in the light receiving section are 24 μm in the horizontal direction.
It is approximately 14 μm in the vertical direction. Furthermore, since the solid-state image sensor is manufactured using a semiconductor IC process, for example, the width of the aluminum line of the vertical signal line 14, the polysilicon of the vertical address line 13, and the polysilicon of the gate electrode 12 is 3 μm.
The above is necessary. Roughly speaking, if the light receiving section is actually designed according to IC design rules, the proportion of the photodiode in one element of the light receiving section will be about 30%. The SR ratio is a factor that determines the image quality of solid-state image sensors.
Regarding the R ratio, the larger the area of the photodiode, that is, the larger the ratio of the photodiode to one element of the light receiving section, the better the SN ratio can be obtained.

FIG. 6 shows a light receiving section designed according to actual IC design rules for the above purpose.

In FIG. 6, 106.1Of is a photodiode as a photoelectric conversion element, and 11 is a drain.

12a and 12b are gates, 131L and 13b are vertical address lines, and 14 is a vertical signal line. Here, photodiode 2 is used to increase the area of the photodiode.
One drain is arranged for each. A photodiode 11 is arranged for the tab and photodiodes 10el and 10f. In this way, the ratio of the photodiode to the light receiving section can be set to 36% to 4096%.

However, the reduction in the modulation degree in the vertical and horizontal directions and the occurrence of smear are the same as in the case of FIG. 2.

In order to prevent the above-mentioned smear generation and decrease in the modulation degree in the vertical and horizontal directions, in the light receiving section, it is necessary to prevent incident light from irradiating areas other than the photoelectric conversion element, that is, the photodiode.
It is sufficient to mask everything except the photodiode part and block light.

FIG. 7 shows an aluminum mask 18 for the purpose of blocking light.
FIG. 3 is a diagram showing a state in which areas other than the photodiode portion are masked by . Note that the vertical signal line 14 made of aluminum wire shown in FIG. 6 needs to be doubly shielded from light by an aluminum mask 18, and it is easier to manufacture if the shape of the aluminum mask 18 is as shown in FIG. This shape is used because it is easy. With the configuration of the light-receiving section as shown in FIG. 7, the main causes of a decrease in the degree of modulation in the vertical and horizontal directions and the occurrence of smear are eliminated, so that an image sensor with good image quality can be obtained.

However, in the semiconductor IC7' process that actually manufactures image sensors, the alignment error of the IC mask is 0.6 to 1 μm.
There are more than m. In other words, in the light receiving section configured as shown in FIG.
If the horizontal scanning output signal of n(H) and the horizontal scanning output signal of n-4-1(H) are arranged shifted toward the lower side of the direction, the amplitudes of the horizontal scanning output signal of n-4-1(H) will be different. When the subject shown in Fig. 4 is imaged, the horizontal scanning blade signals of n(H) and n-1-1 (Hl) are shown in Fig. 8. Such an image sensor output signal is supplied to a television receiver. When played back, a so-called line shading phenomenon occurs, in which shading of the video signal occurs every other horizontal scanning line on the television screen. Needless to say, the line shading phenomenon does not occur in the absence of incident light.

OBJECTS OF THE INVENTION The present invention provides good image quality by eliminating the line shading phenomenon that occurs when the output signal amplitude of a solid-state image pickup device changes periodically every other horizontal line.

Structure of the Invention In order to achieve the above object, the amplitude of the video signal can be adjusted to
The change is made in a direction in which line shading does not occur every horizontal line or every plural horizontal lines.

DESCRIPTION OF EMBODIMENTS The first example of line density correction of a solid-state imaging device according to the present invention will be described below.
An example will be described using FIG. 9.

19 is the solid-state image sensor explained in FIG. 7; 20 is a pulse generator for driving and signal processing the solid-state image sensor; 2
Reference numeral 1 is a transistor, 22 is a resistor, and 21 and 22 constitute an emitter follower, which receives the output signal of the solid-state image sensor 19. 23 is a capacitor, 24 is a transistor, 25 is a power supply, and 23 to 2
5 constitutes an engineered clamp circuit. transistor 2
A clamp circuit (pulse force ball is supplied to the base of 4 from a pulse generator 2o. The dark level of the output signal of the solid-state image sensor 19 is clamped to the potential of a power supply 26. 26a and 26b are current sources, 27.2B, 29°3
0.31.32 are transistors, 33, 34.36 are resistors, and 36 is a power supply. 26 to 36 constitute a gain control amplifier.

37 is a differential amplifier whose input terminal is a transistor 30.
Each of them is connected to 31 collectors.

38 is a video signal output terminal. Transistor 30.
A control signal σ is supplied to the base of the pulse generator 31 from the pulse generator 2o.

・Next, I will explain the operation. The image sensor output signal will be explained as having the form shown in FIG. In Figure 10, (+
L) is the base waveform of the transistor 27. The base potentials of transistors 27 and 28 are the same. Now, if the gain control signal is not supplied to the bases of transistors 30 and 31, transistors 29, 30, and 31
．． Since the base potentials of transistor 32 are the same, transistor 3
Figure 10(b), (C1
A waveform appears.

Figure 10 (
If a gain control signal with a waveform of dl is supplied and the level of the gain control signal is adjusted, the collectors of the transistors 30 and 31 will have the voltages shown in FIG. 10 (el, (f).
) appears, and from the output terminal 38 of the differential amplifier 37, it is possible to obtain a video signal fql in FIG. 10 whose line density has been corrected.

Next, a second embodiment according to the present invention will be explained using FIG. 11 and FIG. 12.

In FIG. 11, the solid-state imaging device explained in FIG. 7, 20 is a pulse generator for driving and signal processing the solid-state imaging device, 21 is a transistor, and 22 is a resistor.
21 and 22 constitute an emitter follower, and serve as a buffer for receiving the output signal of the solid-state image sensor 19. 391L.

39b is a general gain control amplifier, 402L.

40b is a clamp circuit. Clamp circuit 402L
40b is supplied with a clamp pulse from the pulse generator 20, and after the output signal from the solid-state image sensor 19 is suitably amplified by gain control amplifiers 392L and 39b, the dark level of the image sensor output signal is output from the clamp circuit. They are clamped to the same potential by 4ofL and 40b. Reference numeral 41 denotes an analog switch, and the output signals of the clamp circuits 401L and 40b are alternately selected by control pulses from the full pulse generator 20. 42 is an output terminal of the analog switch.

Next, the operation of the second embodiment will be explained. The image sensor output signal will be explained as having the shape shown in FIG.

In FIG. 12, (hl is the base waveform of the transistor 21, and the video signal amplitude is different for every other horizontal line.
Line shading occurs. This signal is supplied to gain control amplifiers 39a and 39b and amplified at different amplification factors. At this time, when setting the amplification factor of the gain control amplifier, for example, the output signal of 39a is n
The signal amplitude of (H), the output signal of 39b is n + 1 (Hl
Focus on the signal amplitude of , and set it so that both are equal. The output signal waveforms of the gain control amplifiers 391L and 39b are shown in FIG. Clamping. The output signals of the clamp circuits 40a and 40b are shown in FIG.
FIG. 12(f) shows the analog switch is switched by the control pulse, and a video signal is obtained from the output terminal 42. The analog switch output signal is shown in FIG. 12(q). As is clear from the waveform of FIG. 12 (ql), a signal whose line density has been corrected is obtained from the analog switch output terminal 42.

In this embodiment, a case has been described in which the signal amplitude changes every other horizontal line, but there may be cases where the signal amplitude changes over multiple horizontal lines due to the configuration of the light receiving section, and the present invention is also applicable to such cases. In other words, it goes without saying that in the second embodiment, it is sufficient to add a gain control amplifier and a clamp circuit, add an input terminal of an analog switch, and change the control pulse so as to perform the desired switching.

Furthermore, in the present invention, the cause of line shading has been explained as misalignment of the light-shielding aluminum mask of the light-receiving section. ,
The present invention is applicable to correction of all kinds of line shading caused by misalignment between the image sensor and the color separation filter, manufacturing accuracy of the color separation filter, or the like.

Effects of the Invention According to the present invention, lines that occur because the amplitude of a video signal obtained from an output signal from a solid-state image sensor changes every other horizontal line, every multiple horizontal lines, or across multiple horizontal lines. Since the shading can be corrected with a simple configuration, a good image without line shading can be obtained. Further, by applying the present invention, it is also possible to treat solid-state image sensors, which were conventionally considered to be defective products, as good products.

[Brief explanation of the drawing]

Figure 1 is a configuration diagram of the MO8 type solid-state image sensor, Figure 2 is the MO8 type solid-state image sensor.
Figure 3 is a diagram showing the light receiving part of an MO8 type solid-state image sensor, Figure 3 is a diagram showing the principle of smear generation in an MO8 type solid-state image sensor, Figure 4 is a diagram showing an example of a subject, and Figure 5 is a diagram showing the smear. Figure 6 is a diagram showing the actual light-receiving section configuration of an MO8 type solid-state image sensor, and Figure 7 is a diagram showing the phenomenon, and Figure 7 is a diagram showing the configuration of an MO8 type solid-state image sensor.
A diagram showing an O8-type solid-state image sensor subjected to a light-shielding aluminum masker, FIG. 8 is a diagram showing an output signal with line shading generated from the solid-state image sensor of FIG. 7, and FIG. 9 is a diagram showing the first embodiment according to the present invention. 10 is a diagram showing signal waveforms of each part in FIG. 9, FIG. 11 is a diagram showing a second embodiment of the present invention,
FIG. 12 is a diagram showing signal waveforms at various parts in FIG. 11. 19... Solid-state image sensor, 39a, 39b...
...Gain control amplifier 7-. 4oa, 40b
... Clamp circuit, 41 ... Analog switch. Name of agent: Patent attorney Toshio Nakao and 1 other person No. 1
Figure 2 Figure 3 Figure 4 Figure 10 Figure 1 O Figure 11 Figure 12 Mountain Time

Claims (1)

[Claims] A light receiving section consisting of a plurality of photoelectric conversion elements arranged in the horizontal (row) direction and vertical (column) direction photoelectrically converts the optical signal from the subject image, and the signal charge obtained by the photoelectric conversion is scanned or transferred. The solid-state image sensor is equipped with a solid-state image sensor that obtains an image sensor output signal, and this solid-state image sensor has a difference in sensitivity of each row of photoelectric conversion elements in the horizontal direction. A solid-state imaging device characterized in that the amplitude is changed for each horizontal line without changing the dark level of the output signal of each horizontal line.