JPH04126445A - Image sensor - Google Patents

Abstract

PURPOSE:To eliminate a fixed pattern noise by allowing a read switching element to be thrown to the position of read operation before and after a reset switching element in each photodetector section is reset so as to surely correct dispersion in dark state output. CONSTITUTION:A read signal Y-1 is kept to an H level even after a reset signal R-1 goes to an L level till a clock signal goes to an H level succeedingly, then a signal read MOSFET 14-1 keeps ON state. Then a signal current at non-exposure with a value in response to an anode level Va-1 at non-exposure flows to a signal read line 16. Since a video signal current in response to the exposure and the signal current at non-exposure are outputted from the signal read line 16, a correction video signal in which the effect of the dispersion in the characteristic such as an amplification factor gm of an amplifier MOSFET 13-1 and a threshold voltage VT is obtained by obtaining a difference of both the currents.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to an image sensor used as a core device in an image input device that reads an image of a document, and particularly in such an image sensor, high S/N, high speed,
This invention relates to an improvement that enables low afterimage and highly sensitive reading operations.

Conventional technology Conventionally, as an image sensor that can achieve high resolution without deteriorating S/N, an increase-width type MOSFET has been developed that is equipped with a MOSFET for each photodiode corresponding to a pixel.
An S image sensor is used.

More specifically, this type of image sensor is constructed by connecting the gate of a MOSFET to the anode or cathode of each photodiode, and the drain current is controlled based on the gate potential modulation of the MOSFET depending on the amount of carriers generated by incident light. By increasing or decreasing it, the S/N can be improved.

By the way, in the above amplification type image sensor, if the characteristics such as the amplification factor gm and gate threshold voltage VT in the MOS FET of each pixel vary, the exposure amount-output characteristics of each pixel will vary. In other words, if the amplification factor gm of each MOSFET varies, the sensitivity to the exposure amount will vary, so in the case of a positive image sensor, the video signal level will vary mainly at the white level with a large amount of light, while the negative In the case of a type image sensor, the video signal level varies when the amount of light is low and the black level is low. Furthermore, if the gate threshold voltage VT varies, the video signal level will vary when the amount of light is opposite.

For this reason, even if each pixel is exposed to light in a certain manner, so-called fixed pattern noise occurs, in which the video signal level for each pixel varies in a fixed pattern.
This may lead to deterioration of S/N.

Taking the above positive image sensor as an example, variations in the video signal level at the white level may not have much effect on the quality of the video signal because the video signal level itself is high, but at the black level In this case, the variation has a large difference in level relative to the video signal level, so it has a particularly large effect on the quality of the video signal.

Therefore, the present applicant applied for patent application No. 2-7386 related to the earlier application.
In this issue, we propose an image sensor that can reduce variations in output, especially during dark times.

This image sensor will be explained below based on the circuit diagram for one pixel shown in FIG. 7 and the timing chart shown in FIG. 8.

In FIG. 7, 1 is a photodiode, 2 is a constant voltage source line, 3 is an amplification MO5FET for amplifying the signal from the photodiode, 4 is a signal readout MO3FET for reading out the amplified signal, and 5 is the amplification MO for
A reset MOS FET is used to set the gate potential of the 3FET3 to the reset potential Vrs, 6 is a signal read line, and 7 is a reset voltage source line for setting the reset potential Vrs.

Gate terminal 4a of MO3FET4 for signal readout and gate terminal 5a of MO3FET5 for reset! Ha,
A readout signal Yn and a nosenote signal Rn which go to "H" level at the timings shown in FIG. 8 are inputted, respectively.

Here, as the signal readout MO5FET 4 and the reset MOS FET 5, an O3F between N channels is used.
An example in which ET is used is shown.

When the photodiode 1 is exposed to light when the reset signal Rn is at L'' level, charges corresponding to the amount of exposure are integrated and accumulated on the anode side.

Then, when the readout signal Yn becomes "H" level, the readout MO3FET4 is turned on, and during the video signal readout period tin until the reset signal Rn becomes "H" level, the photodiode determined by 7 is determined by the accumulated charge. A video signal corresponding to the anode potential of 1, that is, the gate potential Van of MO3FET 3 for amplification is transmitted to the readout line 6.
obtained from.

Also, during the reset output read period t, both the read signal Yn and the reset signal Rn are at "H" level.
2n, the reset MO5FET5 is also turned on, so the anode potential Van of the photodiode l becomes equal to the reset potential Vrs, and the reset output signal or
It is obtained from readout line 6.

Therefore, by determining the difference between the video signal and the reset output signal using a delay circuit and a differential amplifier circuit (not shown), a corrected video signal in which the dark output variation is corrected can be obtained.

That is, the video signal level of all pixels is such that the anode potential Van of the photodiode 1 is the reset potential Vrs.
Since the output signal at the time of reset is obtained as a reference when

Problems to be Solved by the Invention By the way, the gate terminal 5a of the reset MO5FET5
A parasitic capacitance C exists between the photodiode 1 and the anode terminal of the photodiode 1.

Therefore, the reset signal Rn changes from "H" level to "L" level.
Due to the fluctuation of capacitive coupling when changing the level, the anode potential of photodiode F1 becomes the reset potential Vr.
The potential is slightly lower than s. Since this lowered potential becomes the anode potential during non-exposure, the reset output signal includes an offset with respect to the video signal during non-exposure.

This offset amount is mainly due to the reset MO3FET.
The variation in the parasitic capacitance C is generally small, but the variation in the amplification factor gm tends to be relatively large, so the amount of offset during non-exposure varies from pixel to pixel. It becomes difficult. Therefore, even if the difference between the video signal and the reset output signal is determined as described above, the dark output variation may not be completely corrected and the S/N may decrease.

In view of the above-mentioned points, the present invention aims to provide an image sensor that can reliably correct variations in dark output and significantly improve S/N.

Means for Solving the Problems In order to achieve the above object, the present invention includes a photodiode, an amplification element for amplifying a signal from the photodiode, a switching element for reading out a video signal for reading out the amplified signal, and a switching element for reading out a video signal for reading out the amplified signal. A plurality of photodetecting units each include a reset switching element that resets the output level of the diode to a predetermined level, and a readout switching element before and after the reset switching element in each photodetection unit performs a reset operation. The device is characterized in that it is provided with a photodetector control means for performing a read operation.

Operation According to the above configuration, the photodetection unit control means causes the readout switching element to perform the readout operation before and after causing the reset switching element in each photodetection unit to perform the reset operation.

That is, a level signal corresponding to the exposure amount and an accurate non-exposure level signal are read out from the photodetector.

First Embodiment A first embodiment of the present invention will be described with reference to FIGS. 1 to 4.

Figure 1 is a circuit diagram showing the configuration of the image sensor, Figure 2 is a circuit diagram showing the configuration of the correction circuit for the video signal read out from the image sensor, and Figures 3 and 4 are the operations of each part of the image sensor. FIG.

In FIG. 1, 10-1 to 10-n are each 1
In the photodetection section of the pixel, 11-1 to 11-n are photodiodes, 12 is a chamber voltage source line, and 13-1 to 13-n are amplification MOS Fs that amplify signals from the photodiodes.
ET, 14-1 to 14-n are signal readout MOSFETs for reading out the amplified signals, and 15-1 to l.
5-n is the amplification MO3FETI 3-1 to 13-n
16 is a signal readout line, 17 is a reset voltage source line for setting the above-mentioned reset potential Vrs, 21 is a shift register, 22 is a start signal input terminal, 23 is a carry Signal output terminal, 24 is a clock signal line, 25 is a reset original signal generation circuit, 261 to 26
-n is an A N D circuit.

In FIG. 2, 30 is a video signal correction circuit, 31 is a current-voltage converter, 32 and 35 are analog switches, 33 is a sample and hold capacitor, 34 is a buffer, 36 is a differential amplifier, and 37 is a corrected video signal output line. It is.

The above MO3FET13-1 for amplification, MO5FET14-1 for signal readout, and M for resetting, , and
O3FET15-1... finally O3F between N channels
ET is used, and MOSFET14-1 for signal readout
..., and reset MO3FET15-1...
are gate terminals 14a-1...15a-1, respectively.
When the read signal Y-1... input to... or the reset signal R-1... goes to "H" level, it becomes ON state and performs the scooping operation or reset operation. It looks like this.

As shown in FIG. 3, in the shift register 21, after the start signal goes to "H" level, the clock signal goes to "H" level.
Each time the read signal Y-1... is set to the "H" level, the read signal Y-1 of the final stage photodetector 10-n is set to the "H" level. After reaching the level, a carry signal of H'' level is output. This carry signal is used as a start signal for the next-stage image sensor when a plurality of image sensors are arranged in series to form a long image sensor.

The reset original signal generation circuit 25 has a clock signal of “H”.
Each time the reset signal RO reaches the "H" level, the reset original signal RO is outputted at the "H" level for a time t12, delayed by a time tll from the rising timing. More specifically, this reset original signal generation circuit 25 generates a signal with a pulse width t when, for example, the clock becomes "H" level.
It is composed of a one-shot multi-pipe break that generates 11 pulses, and a one-shot multi-by-break that generates a pulse with a pulse width t12 when the above pulse falls.

The AND circuits 26-1... pass the reset original signal RO only when the read signal Y-1... is at a "H" level, and output the reset signal R-1.
It is designed to be output as...

The video signal correction circuit 30 is configured so that the reset original signal RO is
The video signal is corrected by differentially amplifying the signals read before and after reaching the H'' level by the differential amplifier 36.

In the above configuration, each of the photodiodes 11-1 in the photodetector 10-1 is moved to the anode side by exposure after a reset operation is performed as described later. The load is accumulated and becomes an anode potential Va-1 . . . corresponding to the amount of exposure.

When the clock signal goes to "H" level while the start signal is at "H" level, the read signal Y-1 of the shift register 21 goes to "H" level. Therefore, the MO3FET for signal readout of the photodetector 10-1
14-IJ (becomes ONa' state, signal readout line 1
6 is the anode potential Va of the photodiode 11-1.
A video signal current flows with a magnitude corresponding to -1, that is, a magnitude corresponding to the exposure amount.

A current-voltage converter 31 of the video signal correction circuit 30 converts the video signal current into a voltage, and this voltage is passed to a sample-hold capacitor via an analog switch 32 that is turned on under the control of a control device (not shown). It is accumulated in 33.

Time t after read signal Y-1 becomes “H” level
When the ll video signal readout period has elapsed, the reset original signal RO output from the reset original signal generation circuit 25 becomes H'' level.Here, the readout signal Y~1...
Of these, only the read signal Y-1 is at the "H" level, so only the reset signal R1 output from the AND circuit 26-1 is at the "H" level, and the reset signal Y-1 is at the "H" level.
The O3FET 15-1 is turned on, and the anode potential Va-1 of the photodiode 11-1 becomes the reset potential Vrs, as shown in FIG.

Further, when the reset original signal RO becomes "L" level after the reset time of time t12 has elapsed, the reset signal R-1
It also goes to the "L" level, the reset MO3FET 152 turns off, and the 7 node potential Va-1 of the photodiode 11-1 changes due to the change in capacitive coupling caused by the change in the reset signal R-1. is the reset potential V
The potential is slightly lower than rs. This lowered potential becomes the anode potential during non-exposure.

In addition, the read signal Y-1 is reset when the reset signal R-1 is "
Even after the clock signal goes to the "L" level, the "H" level is maintained until the next time the clock signal goes to the "H" level, so the signal reading MO5FETI4-1 maintains the ON state.Therefore, the above-mentioned non-exposure The non-exposure signal current of the magnitude corresponding to the anode potential Va-1 at the time is the signal readout line 1.
It flows to 6.

This non-exposure signal current is converted into a voltage by the current-voltage converter 31 of the video signal correction circuit 30, and the converted voltage is applied to the negative input terminal of the differential amplifier 36.

On the other hand, the voltage accumulated in the sample and hold capacitor 33 is transferred to the buffer 34 and the analog switch 35.
The corrected video signal is applied to the positive input terminal of the differential amplifier 36 via the differential amplifier 36, and the corrected video signal is outputted to the corrected video signal output line 37.

In other words, since the signal readout line 16 outputs a video signal current according to the exposure amount and a non-exposure signal current, by finding the difference between the two, the amplifying MO3FE
A corrected video signal is obtained in which the effects of variations in characteristics such as the amplification factor gm and threshold voltage VT of T13-1 are removed.

Furthermore, the reset signal R-1 is at the "L" level both when the video signal is read out and when the non-exposure signal is read out. Therefore, the effects of variations in capacitive coupling due to parasitic capacitance between the drain and gate in the reset MO3FET 15-1 are the same in either case. Therefore, the video signal corrected by the differential amplification has a high S/N and fixed pattern noise is reliably prevented. For example, when the exposure amount of photodiode l1-1 is Become exactly 0 level.

Next, when the clock signal becomes "H" level again, the read signal Y-1 of the shift register 21 becomes "L" level, while the read signal Y-2 becomes "H" level.

Therefore, in the same manner, the video signal current and the non-exposure signal current are outputted from the photodetecting sections 10-2, and the corrected video signal is outputted from the video signal correction circuit 30.

Note that in the first embodiment, the read signal Y-1
. . has explained an example in which the reset signal R-1 . . . maintains the "H" level state even while it is at the "H" level. Since the current is not used for correcting the video signal, it does not necessarily have to be at the "H" level.

In addition, MOSFET t3-1 for amplification, MO5FET 14-1 for signal readout, and M for reset
O5FET15-1...N-channel MOS
FET is used, and amplification MOS FET13-
Although an example has been described in which the anode terminal of the photodiode 11 is connected to the gate terminal 13a-1 of the photodiode 11, the invention is not limited to this.

That is, for example, in the above case, and when the amplification MO
If the SFET is a P-channel type and the cathode terminal of the photodiode is connected to the gate terminal, a positive video signal is obtained in which the output current increases as the amount of exposure increases, while the MO3FET for amplification is an N-channel type. If the cathode terminal of the photodiode is connected to the gate terminal in the type, or if the MO3FET for amplification is a P° channel type and the anode terminal of the photodiode is connected to the gate terminal, the higher the exposure A negative video signal with a smaller output current is obtained, but in either case, appropriate video signal correction is performed in the same way.

Furthermore, the signal level of the AND circuit 26-1... etc. is as follows.
Although we have shown an example of positive logic, it is not limited to this, and MO3FET14-1 for signal readout and MO3F for reset
Logic corresponding to the channel type of the ET15-1... may be used.

Second Embodiment A second embodiment of the present invention will be described below. In the second embodiment, components having the same functions as those in the first embodiment are given the same reference numerals and explanations thereof will be omitted.

FIG. 5 is a circuit diagram showing the configuration of the image sensor. In Fig. 5, 41-1... is a reset brush lMOS
FET, 42-1... is reset brush 2MOS F
It is ET.

The gate terminals 41a-1... of the reset brushes 1M05FET41-1... are each readout MOS
It is connected to the gate terminal 14a-1... of the FET 14-1-..., and the read signal Y-1... is input thereto. Also, reset brush 2MO5F ET
42-1... Gate terminal 42 a -1...-
The reset original signal RO is input to both of them.

That is, only when the read signal Y-1... and the reset original signal RO both become "H" level,
A reset operation is performed and each photodiode 1
The potential of the anode terminal in 1-1... becomes the reset potential Vrs. Therefore, the first
As in the embodiment, since the video signal current corresponding to the exposure amount and the accurate non-exposure signal current are output, the video signal can be reliably corrected.

In this way, the reset MO3FET of the first embodiment
Since the circuit configuration can be simplified by using two MOSFETs in place of the AND circuits 26-1 and 26-1, the image sensor can be made more compact. , the degree of freedom in design also increases.

In addition, the reset brush 2M0SFET42-1...
As above, reset brush 1M0SFET41-1...
・The photodiode 11-1 is not limited to the one provided on the reset voltage source line 17 side as shown in FIG.
...It may be installed on the side.

Effects of the Invention 4. As explained above, according to the present invention, the photodetection section control means causes the readout switching element to perform the readout operation before and after the resent switching element in each photodetection section performs the reset operation. By providing a signal of a level according to the exposure amount from the photodetector
Since the accurate non-exposure level signal is read out,
This has the effect of reliably correcting dark output variations, removing fixed pattern noise, and significantly improving S/N.

[Brief explanation of the drawing]

FIG. 1 is a circuit diagram showing the configuration of an image sensor in a first embodiment of the present invention, FIG. 2 is a circuit diagram showing the configuration of a correction circuit for a video signal read out from the image sensor, and FIGS. The figure is a timing chart showing the operation of each part of the image sensor, FIG. 5 is a circuit diagram showing the configuration of the image sensor in the second embodiment of the present invention, FIG. 6 is a circuit diagram showing still another example, and FIG. 8 is a circuit diagram showing an example of a conventional image sensor, and FIG. 8 is a timing chart showing the operation of each part of the same. 10-1 to 10-n=photodetection section, 11-1 to 11-n
...Photodiode, I2...Constant voltage source line, 13-1 to 13-n...Amplification MOS FETl
4-1 to 14-n...MO3F for signal reading
ET, 15-1 to 15-n... MO3FE for reset
T, 16...Signal readout line, 17...Reset voltage source line, 21...Shift register, 22...
- Start signal input terminal, 24... Clock signal line, 25... Reset original signal generation circuit, 26-1 to 2
6-n...AND circuit, 30! ...Video signal correction circuit, 33...Sample and hold capacitor, 36...
・Differential amplifier, 37...Correction video signal output line, 4
1-1... Reset brush 1M03FET, 42-1.
...2nd M03FET for reset Agent Patent attorney Shiro Nakajima Figure 2 Figure 4 Figure 7

Claims (4)

[Claims] (1) A photodiode, an amplification element that amplifies a signal from the photodiode, a video signal readout switching element that reads out the amplified signal, and a reset switching element that resets the output level of the photodiode to a predetermined level. A plurality of photodetecting units comprising a plurality of photodetecting units, and a photodetecting unit control means for causing a readout switching element to perform a readout operation before and after causing the reset switching element in each photodetection unit to perform a reset operation are provided. An image sensor characterized by: (2) The image sensor according to claim 1, wherein the photodetector control means outputs a readout signal that sequentially becomes an active level based on a clock signal, and selectively controls each photodetector. readout control means for setting the readout state; and reset original signal generation means for outputting a reset original signal that becomes active level during a predetermined period after each readout signal becomes active level until it becomes inactive level. , an AND circuit that performs a logical product of each readout signal and a reset original signal and outputs the logical product as a reset signal to reset the photodetection section in the readout state. (3) The image sensor according to claim 1, wherein the photodetector control means outputs a readout signal that becomes active level sequentially based on a clock signal, and selectively reads out each photodetector. readout control means for setting the state, and reset original signal generation means for outputting a reset original signal that becomes active level during a predetermined period after each readout signal becomes active level and until it becomes inactive level. The reset switching element of the photodetector includes a control terminal to which the read signal is input and a control terminal to which the original reset signal is input, and both terminals are at an active level. An image sensor characterized in that the image sensor is configured to perform a reset operation when the (4) The image sensor according to claim 1, further comprising a delay means for delaying the video signal read out before the reset operation is performed, and a signal from the delay means and the image sensor before the reset operation is performed. An image sensor comprising differential means for determining a difference between a video signal read out after being read out.