JP2005192123A - Solid-state imaging device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a solid-state imaging device which improves the total gain and reduces any noise of circuits. <P>SOLUTION: The solid-state imaging device has a plurality of pixel units 1 with a photoelectric conversion device and an amplifier for outputting an electric charge generated by the photoelectric conversion device as a voltage, and a noise canceling part 12 for reducing noise caused by differences in characteristics of the pixel units. The noise canceling part 12 has a first capacitor C31 for charging a reference output signal from the pixel unit 1, a second capacitor C33 for charging an output signal from the pixel unit 1 when the time of exposure has passed, and a third capacitor C35 for detecting a signal difference between the positive electrode of the first capacitor C31 and the positive electrode of the second capacitor C33. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a solid-state imaging device that can remove noise components from charge signals accumulated by photoelectric conversion and thermal detection, and in particular, can achieve high sensitivity, low noise, low power consumption, and low cost.

2. Description of the Related Art Conventionally, a MOS solid-state imaging device includes a noise cancellation circuit in order to remove fixed pattern noise caused by variations in transistor threshold voltage Vt.
FIG. 11 is a diagram illustrating a circuit example of a pixel unit, a noise canceling unit, and a signal output unit in a conventional MOS type solid-state imaging device. The pixel unit 1 in FIG. 1 is only for one pixel among a plurality of pixel units arranged on a matrix, and the noise cancellation circuit 22 is a component among a plurality of noise cancellation circuits 22 provided for each column (vertical signal line). The signal output unit 23 shows only one of the plurality of selection transistors 10 provided for each column (vertical signal line).

In the pixel unit 1, the charge signal generated by photoelectric conversion by the photoelectric conversion element 2 is transferred to the floating diffusion 3 through the transfer transistor 4. The charge signal transferred to the floating diffusion 3 is amplified by the source follower 5 and detected as a voltage, and is output as a voltage signal to the vertical signal line through the vertical address transistor 6. The noise cancellation circuit 22 is provided to remove the noise component of the fixed pattern noise. This fixed pattern noise occurs due to variations in the threshold voltage Vt of the source follower 5 and appears as vertical stripe noise on the image.
The signal output unit 23 outputs the pixel signal after noise removal to the outside through the horizontal signal line.

  FIG. 12 is a timing chart showing the timing of the noise removal operation in the solid-state imaging device. As shown in the figure, the floating diffusion 3 is reset to the power supply voltage by turning on the reset transistor 7 at time t0. At this time, the signal output to the vertical signal line by the source follower 5 is used as a reset signal, and is charged by being connected to the negative electrode of the clamp capacitor C1 via the sample hold transistor 8 at time t2. The positive electrode of the clamp capacitor C1 is previously clamped to a constant voltage (CLDC) via the clamp transistor 9 in advance at time t1.

Next, at time t3, the transfer transistor 4 is turned on, and a charge signal is transferred from the photoelectric conversion element 2 to the floating diffusion 3. At this time, the signal output to the vertical signal line by the source follower 5 is used as the accumulated signal Vsig. The transistor 8 is connected to the negative electrode of the clamp capacitor C1 for charging.
Then, the potential of the negative electrode of the clamp capacitor C1 changes to the negative potential side. In conjunction with this, the potential of the positive electrode of the clamp capacitor C1 in the floating state also changes to the negative potential side. Then, the potential Vsig1 expressed by the following formula (1) is charged to the sample capacitor C2 by the capacity distribution between the clamp capacitor C1 and the sample capacitor C2.

As a result of this series of operations, the accumulated signal Vsig is superimposed on the basis of the reset signal, so that the fluctuation potential of the clamp capacitor C1 is equivalent to the noise component canceled between the charge signal of the pixel and the reset signal. It has become.
Vsig1 = Vsig × C1 / (C1 + C2) (1)
Thereafter, the accumulated signal of the pixel appearing on the horizontal signal line by the horizontal selection transistor 10 at time t4 becomes a potential Vsig2 expressed by the following equation (2). Here, the capacitance C3 is a parasitic capacitance of the horizontal signal line.
Vsig2 = Vsig1 × C2 / (C2 + C3) (2)
As described above, the conventional method of changing the potential of the clamp capacitor C1 by superimposing the accumulation signal Vsig on the basis of the reset signal can be realized with a relatively simple circuit.
JP 2000-4399 A

However, the noise removal circuit in the conventional solid-state imaging device has the following problems. That is, when the pixel accumulation signal Vsig passes through the noise component removal circuit, a capacity distribution between the clamp capacitor C1 and the sample capacitor C2 occurs, and the signal is reduced to a signal represented by Vsig1 in the following equation. is there.
Vsig1 = Vsig × C1 / (C1 + C2)
In this equation, if the clamp capacitor C1 is increased and the sample capacitor C2 is decreased, the decrease in Vsig1 can be suppressed. However, a configuration in which the noise component removal circuit is connected to the horizontal signal line via the horizontal selection transistor 10 in the subsequent stage. Thus, the influence of the horizontal signal line capacitance C3 cannot be ignored.

That is, at this time, when the capacity distribution between the sample capacitor C2 and the horizontal signal line capacitor C3 occurs and is read out to the horizontal signal line, the accumulated charge Vsig of the pixel is further expressed by Vsig2 in the following equation. Decrease in signal.
Vsig2 = Vsig1 × C2 / (C2 + C3)
= (Vsig × C1 / (C1 + C2)) × C2 / (C2 + C3)
In the end, even if the design is made in consideration of the balance among the capacitors C1, C2, and C3, it is difficult to increase the signal amount, and the total gain of the circuit becomes a considerably small value, which leads to SN degradation.

  Another problem is that when clamping to a constant voltage (CLDC) through the clamp transistor 9 in advance, depending on the potential setting, the noise component is affected by the threshold voltage of the horizontal selection transistor 10 from the vertical signal line. There is a possibility that the accumulated signal from which “” is deleted cannot be normally read out to the horizontal signal line and the fixed pattern noise in the vertical direction is generated. As a countermeasure, it is necessary to optimize the setting of the clamp potential in consideration of the variation in transistor characteristics.

  The present invention has been made in view of the above-mentioned problems, and a solid-state imaging device in which the influence of the reduction due to the capacity distribution with the horizontal signal line capacity is reduced, the total gain of the circuit is increased, and the SN is improved. The purpose is to provide.

  In order to achieve the above object, in a solid-state imaging device according to the present invention, pixels each including a photoelectric conversion element and an amplifier that outputs a voltage generated by the element in a pixel are arranged in a two-dimensional manner. An image sensor in which pixel signals that are simultaneously read out are output for each column, and has a noise canceling unit as a noise component removing circuit. FIG. 1 is an explanatory diagram showing the operating principle of the noise canceling unit. The noise cancellation unit 12 includes a first capacitor C31 that charges an output signal serving as a reference from the pixel unit, a second capacitor C33 that charges an output signal when the exposure time from the pixel unit has elapsed, and the first capacitor C31. And a third capacitor C35 for detecting a signal difference between the positive electrode of the first capacitor and the positive electrode of the second capacitor. Here, the reference output signal is an output signal when the accumulated charge is reset in the pixel portion or an output signal from the light-shielded pixel portion after the exposure time has elapsed.

With this configuration, it is configured to detect the difference signal between the output signal due to the accumulated charge after the elapse of a certain exposure time and the output signal that becomes the reference signal, and the noise component is deleted by taking the difference signal, Thus, the influence of the decrease due to the capacity distribution with the horizontal signal line capacity is further reduced, the total gain of the circuit is increased, and the SN is improved.
The magnitude of the accumulated signal VsigS of the pixel from which the noise component has been removed depends on the magnitudes of the first to third capacitors C31, C33, and C35, and has a relationship represented by the following equation (3). Here, VsigA represents a signal based on accumulated charges after a certain exposure time has elapsed, and VsigB represents a signal when accumulated charges are reset. The first capacitor C31 and the second capacitor C33 have the same size, and the third capacitor C35 is assumed to be α times as large as the first capacitor C31 or the second capacitor C33.

VsigS = α (VsigA−VsigB) / (1 + 2α) (3)
According to this equation (3), the value of VsigS increases as α increases, that is, as the third capacitance C35 increases.
For example, when the third capacitor C35 is equal to the first capacitor C31 or the second capacitor C33,
VsigS = (VsigA−VsigB) / 3
When the third capacitor C35 is twice the first capacitor C31 or the second capacitor C33,
VsigS = (VsigA−VsigB) × 2/5
become.

  The noise cancellation unit 12 is configured to be connected to the horizontal signal output line 15 via a horizontal selection switch 13 in the signal output unit 14 at the subsequent stage. The capacity distribution between the third capacitor C35 for signal difference detection and the horizontal signal line capacitor C40 occurs, but the larger the third capacitor C35, the higher the output to the horizontal signal line as expressed by the following equation (4). The magnitude of the accumulated signal VsigS2 of the pixels to be increased.

VsigS2 = VsigS × C35 / (C35 + C40) (4)
= (Α (VsigA−VsigB) / (1 + 2α)) × C35 / (C35 + C40)
According to this formula (4), the larger the third capacitance C33, the larger the value of VsigS2, so that it is possible to easily design the balance of the respective capacitances required in the conventional noise cancellation unit 12, and Since the total gain is also increased, SN can be improved.

  In the noise canceling unit 12 described above, it is only necessary to read the potential of the positive electrode of the third capacitor C35 for signal difference detection to the horizontal signal line. It is not necessary to optimize the setting of the clamp potential for countermeasures against the fixed pattern noise in the vertical direction generated due to the influence of the threshold voltage of the selection transistor, and the design is facilitated.

  The solid-state imaging device of the present invention has the above-described noise component removal circuit, and can increase the gain of a pixel signal from which noise is removed simply by designing the difference detection capacitor to be larger than other capacitors. In addition, since it is only necessary to read the potential of the positive electrode of the differential detection capacitor to the horizontal signal line, a clamp for countermeasures against the fixed pattern noise in the vertical direction generated from the vertical signal line due to the influence of the threshold voltage of the horizontal selection transistor. It is not necessary to optimize the potential, and noise components can be easily removed.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
(Embodiment 1)
FIG. 2 is a block diagram showing a schematic configuration of the solid-state imaging device according to Embodiment 1 of the present invention. The solid-state imaging device includes a plurality of pixel units 1 arranged in a matrix, a plurality of noise cancellation units 12 provided for each column, and a signal output unit 14.

  Each pixel unit 1 has the same configuration as the pixel unit 1 shown in FIG. 11, and outputs a reset signal as a reference output signal and a charge signal after the exposure time has elapsed on a vertical signal line. The pixel units 1 for one row are read all at once, and the read reset signal and charge signal are output to the noise canceling unit 12. Each noise canceling unit 12 is a noise component removing circuit including first to third capacitors as in FIG. The signal output unit 14 includes a horizontal selection switch 13 provided for each column and sequentially turned on one by one, and outputs the output signal after noise removal from the noise canceling unit 12 to the outside through the horizontal output signal line 15. .

  FIG. 3 is a circuit example showing the noise canceling unit 12 and the horizontal selection switch 13 in the signal output unit 14. As shown in the figure, the noise canceling unit 12 charges the output signal of the pixel unit 1 when the accumulated charge is reset, and transfers the signal from the pixel unit 1 to the first capacitor C31. 1 transfer switch T32, a second capacitor C33 for charging the accumulated charge after a certain exposure time has elapsed, a second transfer switch T34 for transferring the signal from the pixel unit 1 to the second capacitor C33, A third capacitor C35 for signal difference detection in which the negative electrode is connected to the positive electrode of the first capacitor C31 and the positive electrode is connected to the positive electrode of the second capacitor C33, and the charges charged in the first and second capacitors The difference detection switch T36 for transferring the signal, the first reset switch T37 for holding the positive side of the third capacitor C35 for signal difference detection at the reset potential, and the negative side of the third capacitance C35. Hold at reset potential A second reset switch T38 to, and a horizontal selection switch T39 between the positive electrode and the horizontal signal line of the third capacitor C35 for the signal difference detection.

  FIG. 4 is a time chart showing the driving timing of the noise canceling unit 12. At time t0, all the switches were turned off, and at time t1, only the first and second reset switches T37, T38 were turned on, and both electrodes of the third capacitor C35 for signal difference detection were held at the reset potential. Turn off later. The output signal when the accumulated charge of the pixel is reset at time t2 turns on only the first transfer switch T32, charges the first capacitor C31, and turns it off. At time t3, the accumulated charge after the elapse of a certain exposure time of the pixel is turned on only by the second transfer switch T34, charged to the second capacitor C33, and then turned off. At time t4, only the difference detection switch T36 is turned on, and the charges charged in the first and second capacitors C31 and C33 are transferred to the third capacitor C35. As a result, the differential voltage between the output signal when the accumulated charge of the pixel portion is reset and the accumulated signal after the fixed exposure time of the pixel portion appears at both electrodes of the third capacitor C35 for signal difference detection. The difference signal VsigS at this time is a signal based on the accumulated charge after the elapse of a certain exposure time, VsigA, a signal when the accumulated charge is reset is VsigB, and the first capacitor C31 and the second capacitor C33 have the same magnitude. If the third capacitor C35 is α times as large as the capacitor C31, the size is expressed by the following equation.

VsigS = α (VsigA−VsigB) / (1 + 2α)
As is apparent from this equation, VsigS increases as the third capacitor C35 for signal difference detection increases. After the difference detection operation is completed, only the difference detection switch T36 is turned off. At time t5, only the second reset switch T38 is turned on again, and the negative electrode of the third capacitor C35 is held at the reset potential and then turned off, so that the positive electrode potential of the third capacitor C35 becomes the reset potential. As a reference, the difference signal VsigS is charged. In other words, the accumulated signal after the lapse of the fixed exposure time from which the noise component of the pixel is deleted can be read out to the horizontal signal line only by turning on the horizontal selection switch T39 at time t6. It is no longer necessary to optimize the setting of the clamp potential to prevent fixed pattern noise in the vertical direction that is generated by the influence of the threshold voltage of the horizontal selection transistor from the vertical signal line that was required in the conventional noise component removal circuit. It was. Further, at this time, capacity distribution with the horizontal signal line capacitor C40 occurs, but the gain can be increased by increasing the third capacitor C35. When this is expressed by an expression, the pixel accumulation signal VsigS2 output to the horizontal signal line has a magnitude expressed by the following expression.

VsigS2 = VsigS × C35 / (C35 + C40)
= Α (VsigA−VsigB) / (1 + 2α) × C35 / (C35 + C40)
According to the above equation, the larger the capacitance C35, the larger the value of VsigS2, so that it is possible to easily design the balance of the respective capacitances required in the conventional noise component removal circuit and to increase the total gain of the circuit. Therefore, SN improvement can be realized. Further, in the case of the above-described noise component removal circuit, it is only necessary to read out the potential of the positive electrode of the signal difference detection capacitor C33 to the horizontal signal line, and the horizontal selection from the vertical signal line required in the conventional noise component removal circuit. It is not necessary to optimize the setting of the clamp potential for taking countermeasures against the fixed pattern noise in the vertical direction that is generated under the influence of the threshold voltage of the transistor, and the design is facilitated.

FIG. 5 is a circuit example in which all the constituent elements of the noise canceling unit 12 shown in FIG. 4 use N-type MOS transistors. The manufacturing process of the P-type MOS transistors can be omitted, and the TAT (Turn Around Time) is shortened. This leads to cost reduction.
FIG. 6 shows a case where the reset potential of the difference detection circuit (third capacitor C35) is changed to GND instead of DC potential in the noise canceling unit 12 shown in FIG. When the potential of the positive electrode of the difference detection circuit is based on GND, it is possible to keep the fixed pattern noise in the vertical direction from being easily generated due to variations in the threshold voltage of the horizontal selection transistor.

  The noise canceling unit 12 shown in FIG. 4 may be configured with a P-type MOS transistor or a C-type MOS, unlike FIG.

(Embodiment 2)
In the first embodiment, a solid-state imaging device suitable for normal visible light has been described. In the present embodiment, a solid-state imaging device that is suitable for noise removal even when imaging a minute signal such as an image sensor that detects a minute signal, an infrared sensor, or a thermal sensor (far infrared sensor) will be described.

  FIG. 7 is a block diagram showing a schematic configuration of the solid-state imaging device according to Embodiment 2 of the present invention. Compared with the solid-state imaging device shown in FIG. 2, the solid-state imaging device of FIG. 2 includes a pixel unit 1 a instead of the pixel unit 1, and newly adds frame memories 63 and 64, an operational amplifier 65, and a mechanical shutter to the outside. It differs from the point to prepare. The description of the same points will be omitted, and different points will be mainly described.

FIG. 8 is a diagram showing one pixel unit 1 a, one noise canceling unit 12, one horizontal selection switch 13 in the signal output unit 14, frame memories 63 and 64, and an operational amplifier 65.
The pixel unit 1a detects the external signal such as light and heat by having the first pixel unit 61 having a detection function for the external signal such as light and heat and the light shielding film on the photoelectric conversion element. A second pixel portion 62 having no function. The first pixel portion 61 and the second pixel portion 62 output output signals when the mechanical shutter is closed and when the exposure time elapses after the mechanical shutter is opened. The first pixel unit 61 is connected to the second capacitor C33, and the second pixel unit 62 is connected to the first capacitor C31.

The frame memory 63 holds the output signal of all the pixel units 1a when the mechanical shutter is closed. The frame memory 64 holds the output signals of all the pixel portions 1a when the exposure time has elapsed after opening the mechanical shutter.
The operational amplifier 65 is a differential amplifier that inputs and amplifies two corresponding pixel signals of the frame memories 63 and 64.

  For noise removal, the mechanical shutter is initially closed to block the external signal, and the signal from the first pixel unit 61 having a detection function for the external signal at that time is sent to the second capacitor C33. The second pixel unit 62 that is charged and does not have a detection function for an external signal is charged in the first capacitor C31. Then, the signal difference is detected by the third capacitor C35, read to the horizontal signal line, and temporarily stored in the first frame memory 63. Next, the mechanical shutter is opened, an external signal is input, the signal from the first pixel unit 61 having a detection function for the external signal is charged to the second capacitor C33, and the external signal is detected. The second pixel portion 62 having no function is charged in the first capacitor C31. Then, the signal difference is detected by the third capacitor C35, read out to the horizontal signal line, and temporarily stored in the frame memory 64. Then, the data fetched from the frame memories 63 and 64 at the opposite addresses is inputted to the external operational amplifier 65 to differentially amplify the data. Through this series of operations, the output signals of the second pixel unit 62 having no detection function and the first pixel unit 61 having the detection function are compared at the same time under the same ambient conditions with respect to the external signal serving as a reference. The reset level and the signal accumulation level of the pixel unit 61 can be taken into the frame memories 63 and 64, respectively. Therefore, in general, the characteristics of the dark current greatly differ depending on the ambient temperature of the photoelectric conversion element. However, the solid-state imaging device of this embodiment can accurately store the pixel accumulation signal regardless of the ambient conditions such as temperature. It becomes possible to extract.

FIG. 9 is a circuit example showing the pixel unit 1a and the noise canceling unit 12 shown in FIG. 8 in more detail. FIG. 10 is a time chart showing the drive timing of FIG.
All switches are turned off at time t0, the first and second reset switches T37 and T38 are turned on at time t1, and both poles of the third capacitor C35 are set to a reset potential and then turned off. . At time t2, the mechanical shutter is closed and the second pixel unit 62 that does not have a detection function for the external signal and the first pixel unit 61 that has a detection function for the external signal are accumulated. The charge is set to the reset level by the third reset switch T45 and the fourth reset switch. Thereafter, at time t3, the first transfer switch T32 and the second transfer switch T34 are turned on, and the signal of the second pixel portion 62 that does not have a detection function for the external signal is transmitted to the first transfer switch T32. The capacitor C31 is turned off after the signal of the first pixel portion 61 having a detection function for the external signal is charged in the second capacitor C33. After that, at time t4, the difference detection switch T36 is turned on, the signal difference is detected by the third capacitor C35 for difference detection, and then turned off. At time t5, the horizontal selection switch T39 is turned on to read out to the horizontal signal line, and the third transfer switch T41 is turned on, temporarily stored in the first frame memory 63 and turned off. The frame data stored in the frame memory 63 includes a characteristic difference between the first pixel unit 61 and the second pixel unit 63 when the mechanical shutter is closed (such as a difference in the amount of dark current generated in the reference state). become.

  Next, the mechanical shutter is opened at time t6, an external signal is input, and after a certain exposure time has elapsed until time t7, the first and second reset switches T37, T38 are turned on at time t8, Both poles of the third capacitor C35 are set to a reset potential and then turned off. At time t9, the first transfer switch T32 and the second transfer switch T34 are turned on, and the signal of the second pixel portion 62 that does not have a detection function for the external signal is sent to the first capacitor C31. In addition, the second capacitor C33 is charged with a signal of the first pixel portion 61 having a detection function for the external signal, and then turned off. After that, at time t10, the difference detection switch T36 is turned on to detect the signal difference with the third capacitor C35 for difference detection, and then turned off. At time t11, the horizontal selection switch T39 is turned on to turn on the horizontal signal. The fourth transfer switch T42 is turned on, temporarily stored in the second frame memory 64, and turned off. The frame data stored in the frame memory 64 is a characteristic difference between the first pixel portion 61 and the second pixel portion 62 when the mechanical shutter is opened and the exposure time has elapsed (difference in the amount of dark current generated in the reference state). Etc.).

  After that, at time t12, the fifth transfer switch is turned on, and the data fetched from the frame memories 63 and 64 to the respective addresses opposite to each other is input to the operational amplifier 65 to be differentially amplified and output. Yes. The differential input of the operational amplifier 65 cancels out the characteristic difference at the time of exposure in the first pixel unit 61 and the second pixel unit 62, and further increases the signal output by amplification without affecting the minute signal. The SN ratio can be improved.

  As described above, according to the solid-state imaging device of the present embodiment, even when only a minute signal can be obtained from the photoelectric conversion element, the differential signal detection of the first pixel unit 61 and the second pixel unit 62 can be performed. A signal output with improved S / N ratio can be obtained by differential amplification. For example, even a sensor that detects minute signals, such as an infrared sensor or a thermal sensor that detects far-infrared rays, can remove noise and obtain a high gain, and can generate dark current that depends on ambient conditions. The influence can be removed.

In the second embodiment, the pixel portion 1a includes the second pixel portion 62 that does not have a detection function by having a light-shielding film. However, the pixel portion 1a may be configured to be used as a plurality of pixel portions 1a. Good. For example, a configuration in which one second pixel portion 62 is provided for each column may be employed.
In Embodiment 2, the pixel signal is output twice when the mechanical shutter is closed and when the mechanical shutter is opened. Instead of closing the mechanical shutter, the pixel unit is reset, Instead of opening, the reset of the pixel portion may be canceled.

  Since the noise removal circuit of the present invention can detect a difference with a simple circuit configuration, it is useful when incorporated in a semiconductor integrated circuit where a reduction in circuit area is desired. For example, a camera incorporated in a digital camera or a mobile phone It is useful for solid-state imaging devices such as image sensors and infrared sensors that need to detect minute signals.

It is explanatory drawing which shows the principle of operation of the noise cancellation part of this invention. It is a block diagram which shows schematic structure of the solid-state imaging device in Embodiment 1 of this invention. 3 is a circuit example showing a noise canceling unit 12 and a horizontal selection switch 13 in a signal output unit 14. It is a time chart which shows a drive timing. This is a circuit example in which all the constituent elements of the noise canceling unit 12 use N-type MOS transistors. The case where the reset potential of the difference detection circuit (third capacitor C35) is changed to GND instead of DC potential is shown. It is a block diagram which shows schematic structure of the solid-state imaging device in Embodiment 2 of this invention. It is a figure which shows one of the noise cancellation parts 12, and an external peripheral circuit. It is a circuit example which shows the pixel part 1a and the noise cancellation part 12 in detail. It is a time chart which shows a drive timing. It is a figure which shows the circuit example of the pixel part in a conventional MOS type solid-state imaging device, a noise cancellation part, and a signal output part. It is a timing chart which shows the timing of the noise removal operation | movement in the conventional solid-state imaging device.

Explanation of symbols

12 Noise Canceling Unit 13 Horizontal Selection Switch 14 Signal Output Unit 15 Horizontal Output Signal Line 15 Horizontal Signal Output Line C31 First Capacitor C33 Second Capacitor C35 Third Capacitor T32 First Transfer Switch T34 Second Transfer Switch T36 Difference detection switch T37 First reset switch T38 Second reset switch T39 Horizontal selection switch C40 Horizontal signal line capacitance T41 Third transfer switch T42 Fourth transfer switch T45 Third reset switch 61 First pixel unit 62 Second pixel portion 63 Frame memory 64 Frame memory 65 Operational amplifier

Claims (10)

A solid-state imaging device comprising: a plurality of pixel units each having a photoelectric conversion element and an amplifier that outputs a voltage generated by the photoelectric conversion element; and a noise cancellation circuit that reduces noise caused by a characteristic difference between the pixel units. And
The noise cancellation circuit is
A first capacitive element that charges a reference output signal from the pixel portion;
A second capacitive element that charges an output signal when the exposure time from the pixel portion has elapsed;
A solid-state imaging device comprising: a third capacitive element for detecting a signal difference between a positive electrode of the first capacitive element and a positive electrode of the second capacitive element. The solid-state imaging device according to claim 1, wherein the reference output signal is an output signal when the pixel unit is reset. The solid-state imaging device according to claim 2,
A first transfer switch for transferring a signal of the pixel portion to a first capacitor element that charges an output signal when the accumulated charge is reset;
A second transfer switch for transferring the signal of the pixel to a second capacitor element that charges the accumulated charge after a certain exposure time has elapsed;
Charges charged from the positive electrode of the first capacitive element to the negative electrode of the third capacitive element and from the positive electrode of the second capacitive element to the positive electrode of the third capacitive element, respectively, to the first and second capacitive elements. A difference detection switch for transferring
A second reset switch for holding the positive side of the third capacitive element for signal difference detection at a reset potential;
A first reset switch for holding the negative electrode side of the third capacitive element at a reset potential;
A readout switch between the positive electrode of the third capacitive element for signal difference detection and a horizontal signal line;
A solid-state imaging device comprising: The solid-state imaging device according to claim 3,
First, turn off each of the switches,
Second, by turning on only the first and second reset switches, both electrodes of the third capacitive element for signal difference detection are reset and then turned off.
Third, by turning on only the first transfer switch when the pixel unit is reset, the output signal when the pixel unit is reset is charged to the first capacitor element and then turned off.
Fourth, by turning on only the second transfer switch after the exposure time elapses, the accumulated charge after elapse of a certain exposure time of the pixel is charged to the second capacitor element and then turned off.
Fifth, by turning on only the difference detection switch, the charges charged in the first and second capacitive elements are transferred to the third capacitive element, and the accumulated signal after a certain exposure time has elapsed. Turn off after detecting the difference,
Sixth, by turning on only the second reset switch again, the negative electrode of the third capacitive element is held at the reset potential and then turned off.
Seventhly, by turning on the readout switch, the potential of the positive electrode of the third capacitive element is turned off after the accumulated signal from which the noise component of the pixel is deleted is output to the horizontal signal line. apparatus. The solid-state imaging device according to claim 1,
The plurality of pixel portions include a light-shielded pixel portion and a non-light-shielded pixel portion,
The first capacitor element charges an output signal from the light-shielded pixel portion as the reference output signal,
The solid-state imaging device, wherein the second capacitor element charges an output signal from a pixel portion that is not shielded from light. The solid-state imaging device according to claim 5, further comprising: an output in which a difference is detected in the third capacitive element based on output signals of both the light-shielded pixel portion and the non-light-shielded pixel portion before the start of exposure. A first holding unit for holding a signal;
A second holding unit for holding an output signal detected by the third capacitive element based on output signals of both the light-shielded pixel unit and the non-shielded pixel unit after the start of exposure;
A solid-state imaging device comprising: an amplification unit that differentially amplifies two output signals held by the first holding unit and the second holding unit. A first transfer switch for transferring an output signal from the light-shielded pixel portion to the first capacitor;
A second transfer switch for transferring an output signal from the pixel portion that is not shielded to the second capacitor;
The positive electrode of the first capacitive element is charged to the negative electrode of the third capacitive element, the positive electrode of the second capacitive element is charged to the positive electrode of the third capacitive element, and the first and second capacitive elements are charged respectively. A differential detection switch for transferring charge;
A second reset switch for holding the positive side of the third capacitive element for signal difference detection at a reset potential, and a first reset switch for holding the negative side of the third capacitive element at a reset potential;
A readout switch provided between the positive electrode of the third capacitive element for signal difference detection and a horizontal signal line;
A third transfer switch provided between the readout switch and the first holding unit;
A fourth transfer switch provided between the readout switch and the second holding unit;
And a fifth transfer switch for transferring the output signals held in the first and second holding units to the amplification unit. The solid-state imaging device according to claim 7,
First, turn off each of the switches,
Second, by turning on only the first and second reset switches in a state before the start of exposure, both electrodes of the third capacitor element for signal difference detection are held at a reset potential and then turned off.
Third, the output signal from the light-shielded pixel unit is turned off after charging the first capacitive element by turning on only the first transfer switch,
Fourth, the output signal from the pixel portion that is not shielded from light is turned on only the second transfer switch, the second capacitor element is charged, and then turned off.
Fifth, by turning on only the difference detection switch, the charge charged in the first and second capacitive elements is transferred to the third capacitive element, and the accumulated signal is differentially detected and then turned off. ,
Sixth, by turning on only the second reset switch again, the negative electrode of the third capacitive element is held at the reset potential and then turned off.
Seventh, by turning on the readout switch, the potential of the positive electrode of the third capacitive element is read out to the horizontal signal line as an accumulated signal after noise component deletion, and then turned off.
Eighth, turn on the third transfer switch to hold it in the first holding unit,
Ninth, in the state after the start of exposure, the output signals detected by the second holding unit are held in the second holding unit by controlling the first to seventh switch control and the fourth transfer switch,
Tenth, a solid-state imaging device, wherein the fifth transfer switch is turned on and amplified by a differential amplifier. A solid-state imaging device according to any one of claims 1 to 8,
A solid-state imaging device, wherein the noise cancellation circuit includes an NMOS transistor and an NMOS capacitor element. A solid-state imaging device according to any one of claims 1 to 8,
A solid-state imaging device, wherein a reset potential of both electrodes of the third capacitive element is set to a GND level.

Images