JP2000196961A - Image pickup device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an image pickup device of a high quality picture adapted to human eyes by allowing a nonlinear resistant element to nonlinearly converting the electric charge of optical signals stored in photoelectric conversion elements in the storing period of photoelectric conversion elements. SOLUTION: In the image pickup device having the photoelectric conversion elements 1, 2 and the nonlinear resistant element 3 connected to the elements 1, 2, the element 3 nonlinearly converts the electric charge of the optical signals stored in the elements 1, 2 in the storing period of the elements 1, 2. In this image pickup device, reset of each photoelectric conversion cell, reading of noise and signals and control of photoelectric conversion are executed by a vertical register and control of a noise removing and memory part formed of a noise removed signal storage constitution part is executed by a horizontal shift register. Then, a signal is amplified by an amplifier to be converted into a digital signal through an AGC and an A/D converter.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an imaging device,
In particular, the present invention relates to an imaging device having a non-linear photoelectric conversion characteristic. INDUSTRIAL APPLICABILITY The present invention is suitably used for an imaging apparatus in which amplifying means (amplifier) is shared between pixels.

[0002]

2. Description of the Related Art Since a conventional solid-state imaging device is mainly formed on a single-crystal silicon substrate, the photoelectric conversion element is p-type.
A reverse-biased diode (photodiode) having a -n junction or a potential well (photogate) formed by a surface electrode is used. The photoelectric conversion characteristics of these photoelectric conversion elements are such that the lower part detects several tens of electrons as the photocarriers and the upper part detects tens of thousands of electrons. The photoelectric conversion characteristics are basically linear because a pair of photocarriers (electrons and holes) are generated for one photon, and the accumulated charges are charge-voltage converted by the capacitance of the photoelectric conversion element. Characteristics. The dynamic range of the photoelectric conversion characteristic is about four digits, the S / N ratio of the photoelectric conversion characteristic is large on the high frequency side (≦ 10,000),
It is small (≧ 1) on the low frequency side.

[0003] However, the human eye is made to attach importance to contrast, and it is difficult to distinguish between the two with an illuminance difference of 1.5% or less. Therefore, in a system in which a human looks directly at the output characteristics of the imaging device, the high-frequency characteristics of the photoelectric conversion characteristics are useless, and the low-frequency characteristics are unsatisfactory.

[0004] For these reasons, logarithmic compression type imaging devices have recently been receiving attention. Logarithmic compression methods are roughly divided into those using the forward current of a diode, MOSF
Some use the sub-threshold characteristic of ET.
Particularly, the latter is well studied because it has good compatibility with the photoelectric conversion element.

[0005] Examples of imaging devices utilizing the sub-threshold characteristics of MOSFETs include the journal Vol.
52, No.2 pp214 (1998), ISSCC98 FA11.5: A256 × 256
CMOS Imaging Array With Wide Dynamic Range Pixels
and Column-Parallel Digital Output,
Japanese Patent Application Laid-Open No. 167848, Japanese Patent Application Laid-Open No. 5-347515, and the like.

In a CMOS sensor, as shown in FIG. 7, an MO for resetting a photodiode 101 or a source follower amplifier 102 as amplifying means is used.
Because the SFET 104 is connected, the logarithmic compression is
This is feasible without adding new devices. That is, the MOSF corresponding to the reset MOSFET
As long as the ET 104 is used in an area exhibiting the subthreshold characteristic, the photocurrent is log-compressed and output as pointed out in the above-mentioned Journal of the Institute of Image and Television Engineers.

Also, an MO exhibiting a subthreshold characteristic
The SFET 104 does not necessarily need to be active, and the load (load) M having the gate and the drain short-circuited as described in the above-mentioned Japanese Patent Application Laid-Open No. 5-347515.
It may be an OSFET. However, reset MO
Since it cannot be used as an SFET, reset M
It is required to newly provide an OSFET in parallel with the load MOSFET.

Further, as described above, logarithmic compression is also possible by applying a voltage φ _RES to the gate of the reset MOSFET 104 at a certain timing and opening and closing the gate intermittently.

Further, examples of an image pickup apparatus which is effective when the number of pixels is increased and which shares an amplifying means between pixels include, for example, JP-A-63-100879 and JP-A-9-46596.

FIG. 8 shows an example in which amplifying means (amplifier) is shared by four pixels. As shown in the drawing, the unit cell includes four photodiodes 111-1 to 111-4, a source follower amplifier 112 shared by the photodiodes 111-1 to 111-4, and four photodiodes 111 connected to the gate of the source follower amplifier 112. Transfer gates 115-1 to 115-4 and a reset MOSF provided between them so as not to mix signal charges from -1 to 111-4.
The ET 114 includes a MOSFET 113 for selecting a unit cell.

[0011]

However, in general, the logarithmic compression type sensor tends to have poor image quality as compared with a normal linear sensor.

As described above, the S / N on the low frequency side is not sufficient even with a linear sensor, but the purpose of the conventional logarithmic compression type sensor is to expand the dynamic range, and therefore, a wide range of light amount is required. Since the photoelectric conversion characteristics are set, the S / N ratio on the high frequency side tends to be sacrificed. In addition, in the case of a linear sensor, the gain and the like are adjusted as much as possible, and the system is configured to be used in the high frequency side. Therefore, the S / N ratio in the high frequency side is different from that of the logarithmic compression type sensor in linearity. The difference from the sensor was obvious.

[0013] Further, since the imaging apparatus is generally used at room temperature, it is affected by thermal noise. In order to operate at this normal temperature, the logic level is at least kTln2 / q
(V) It is said that the degree is necessary.

Although this is on the order of several tens of mV, it is a sufficiently large value even in the above-mentioned linear sensor. That is,
The bias applied to the photoelectric conversion element is about several volts,
Therefore, the noise level is several V / 10,000 and the number of commas is m
V is required. However, in reality, the magnitude of the random noise derived from the thermal noise is on the order of several tens of mV as described above, and it can be seen that the influence of the random noise on the S / N ratio is extremely large.

As can be seen from FIG. 8, in a system in which an amplifier is shared between pixels, the reset MOSFET 114
Is provided on the amplifier 112 side, so that the signals of the photodiodes 111-1 to 111-4 can be logarithmically compressed during the accumulation period in which the transfer gates 115-1 to 115-4 are closed. Can not.

Therefore, the signal is transmitted when the transfer gate 115 is opened. The signal current flowing during the transfer is a current for charging the gate capacitance of the amplifier 112, and is not necessarily a steady current. This indicates that the assumption shown in equation (1) in the above-mentioned document does not hold, and for this reason, the principle of logarithmic compression does not hold.

Further, at the time of transfer, the reset MOSFET 11
Logarithmic compression is also possible by intermittently opening the gate of gate 4, but the transfer time is as short as about 1 μsec and the charging current is non-stationary, so the timing becomes very critical. It is virtually useless.

[0018]

The image pickup apparatus of the present invention is an image pickup apparatus having a photoelectric conversion element and a non-linear resistance element connected to the photoelectric conversion element in a pixel, wherein the non-linear resistance element is provided. The device is an imaging device that nonlinearly converts an optical signal charge accumulated in the photoelectric conversion element during an accumulation period of the photoelectric conversion element.

Further, the image pickup apparatus of the present invention comprises at least a plurality of photoelectric conversion elements, a plurality of non-linear resistance elements respectively connected to the plurality of photoelectric conversion elements, and an amplifying signal from the plurality of photoelectric conversion elements. A unit cell is configured with the amplifying unit, and the amplifying unit is an imaging device shared by the plurality of photoelectric conversion elements, wherein the non-linear resistance element is
An imaging apparatus for nonlinearly converting optical signal charges accumulated in the photoelectric conversion element during an accumulation period of the photoelectric conversion element.

The photoelectric conversion characteristic of the image pickup apparatus of the present invention is not linear but non-linear, and is close to that of a logarithmic compression type sensor, in consideration of the characteristics of the human eye. Since the sensitivity can be adjusted by other means, such as an electronic shutter (exposure time), a wide dynamic range is not necessarily required.

The image quality of the logarithmic compression sensor is improved by reducing its dynamic range. In other words, at room temperature, the value of kTln2 / q is 17.9 mV, which can be determined as a logic level in the case of an illuminance difference of 1.5% = 0.129 dB. Is 2.8 V / 20 dB.

As a result, the image pickup apparatus of the present invention can operate at a normal voltage of several volts.
With an applied voltage of, the dynamic range is one or two digits, but a subtle color tone difference can be expressed instead. In addition, the noise tolerance and logarithmic compression which are sufficiently large make it easy to take countermeasures against various noises, which also contributes to the improvement of the color tone.

In a normal image, the luminance difference of the object of interest is not as wide as the wide dynamic range of ~ 8 digits perceived by the human eye.

Further, the low-frequency side (black level) of the present invention is buried in noise similarly to a normal linear sensor. Considering that the reflectance of a normal image monitor is about 5%. It doesn't matter. Although the high-frequency side (white level) is more easily crushed than a linear sensor, it does not pose a problem if the gain control is adjusted to the object of interest. Rather, light resistance that does not cause side effects such as blooming and smear is required. In the present invention, the measures can be taken efficiently.

The S-characteristic of the silver halide film can be realized by using the present invention by logarithmically compressing only the high frequency side of the photoelectric conversion characteristic.

In a system in which an amplifier is shared between a plurality of photoelectric conversion elements, logarithmic compression is realized by connecting non-linear resistors such as load MOSFETs to the photoelectric conversion elements.

The principle will be described below using equations.
Now, for simplicity, consider a system of one photodiode and one non-linear resistor as shown in FIG.

The capacitor 1 and the constant current power supply 2 represent a photodiode by an equivalent circuit. Now capacity 1 for simplicity
Is a constant value C (F). ip represents a photocurrent (A) generated by light reception. 3 is a load MOSFET
A non-linear resistor consisting of

[0029]

(Equation 3) V _S: source voltage V _T of the load MOSFET: the threshold voltage of the load MOSFET is true.

The boundary condition Vp (t =
0) = V _DD solves for t >> 0

[0031]

(Equation 4) It can be seen that logarithmic compression is performed.

Here, V ₀ and i ₀ are

[0033]

V ₀ = nkT / q (5)

[0034]

I ₀ = id ₀ exp (−V _T / V ₀ ) (6)

It should be noted that the present invention performs logarithmic compression not in the constant current mode described in the above-mentioned document but in the accumulation mode.

Further, the present invention is by changing the value of the threshold voltage V _T of the load MOSFET, it is also possible to arbitrarily set the point at which the logarithmic compression begins.

Solving equation (3) under boundary conditions gives:

[0038]

(Equation 7) Becomes

[0039] V _T dependence of formula (7) performs the specifically described embodiments.

[0040]

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

FIG. 2 shows an equivalent circuit diagram of a unit cell of a CMOS sensor having a common amplifier for every four pixels, which is one embodiment of the image pickup apparatus of the present invention.

The unit cell is composed of four pixels each having a size of 8 μm square, and has a size of 16 μm square. The cell is formed by a 0.4 μm rule CMOS process.

In FIG. 2, photodiodes 21-1 to 21-4 have a capacitance of about 10 fF. 22
Denotes a source follower amplifier as an amplifying means, and the photodiode 21- is connected via transfer gates 25-1 to 25-4.
1 to 21-4. Reference numeral 23 denotes a selection MOSFET for selecting each row of the source follower amplifier 22. Signal charge information in the gate electrode of the source follower amplifier 22 is read out from the vertical signal line 26 in the form of a current through the selection MOSFET 23. Reference numeral 24 denotes a reset MOSFET for resetting the gate electrode.

The photodiodes 21-1 to 21-4 have L = 0.4 μm, W = 1.6 μm, and V _T = 1.0, respectively.
V, V ₀ = 47.78 mV, id ₀ = 6.526 μA The power supply voltage V is set by the load MOSFETs 29-1 to 29-4.
_DD = 3.5V.

The load MOSFETs 29-1 to 29-4 are t
During the accumulation period of = 30 msec, it has the function of allowing electrons, which are accumulated charges in the photodiode, to escape to the power supply _VDD .

First, after the potentials of the gate electrode of the source follower amplifier 22 and the photodiodes 21-1 to 21-4 are reset to V _DD by the reset MOSFET 24, the transfer gates 25-1 to 25-4 are closed to store the data. Begin. The light incident on the photodiodes 21-1 to 21-4 generates electrons as signal charges, and the photodiode voltage Vp gradually decreases.

At the same time, the voltage applied between the load MOSFETs 29 increases, the channel conductance of the load MOSFET 29 increases, and a current proportional thereto flows. Then, a logarithmic compression action acts on the photodiode voltage Vp according to the principle described using equation (4).

After the accumulation time t = 30 msec, the signal charges accumulated in the photodiodes 21-1 to 21-4 are transferred to the gates of the source follower amplifier 22 through the transfer gates 25-1 to 25-4. It is.

FIG. 3 shows the photoelectric conversion characteristics calculated by the equation (7) in this embodiment.

In this embodiment, since V _T = 1.0 V, it can be seen that logarithmic compression characteristics are exhibited when the photocurrent ip is 10 ⁻¹⁴ A or more. In this embodiment, the dynamic range is very wide, and five digits or more are secured.

As another embodiment of the present invention, an example will be given in which the linear characteristic is shown halfway and the logarithmic compression type characteristic is shown halfway. The circuit configuration shown in FIG.
The V _T of FET29 is a case of a 4.0V.

FIG. 4 shows the photoelectric conversion characteristics similarly calculated by the equation (7). According to FIG. 4, it can be seen that the linear characteristic is exhibited up to about 10 ⁻¹² A, and thereafter, the mode shifts to the logarithmic compression type.

According to this embodiment, the photoelectric conversion characteristics are close to the characteristics of the S-curve seen in the silver halide film.

The voltage V _{DD at which the} photodiode saturates is
In a region close to 3.5 V, electrons serving as signal charges in the photodiode are released to the power supply V _DD , and therefore, they can be expected to have a function as an overflow drain for preventing the blooming described above.

FIG. 5 is a schematic sectional view of the unit cell of this embodiment.

In a p-well 52 having a depth of 2 μm formed on a silicon CZN (100) substrate 51, an n ⁺ diffusion layer 54 having a depth of 0.45 μm, which is one semiconductor region of a photodiode serving as a photoelectric conversion portion. Are formed.

[0057] n ⁺ diffusion layer 54 through the gate electrode 56 of the gate length 0.4μm formed of silicide other n ⁺
It is in contact with the diffusion layer 55. From the diffusion layer 55, the metal electrode 5
7 is drawn out and connected to the gate electrode 56 and a power supply V _{DD (} not shown).

N ⁺ Diffusion Layer 5 Constituting Photodiode
4, a reverse bias voltage is applied to the p well 52 via a transfer gate (not shown), and the pn junction of the n ⁺ diffusion layer 54 and the p well 52 is reverse biased. As a result, a carrier-free depletion layer spreads across the pn junction, and when light incident through the light-shielding film 58 shines thereon, a carrier pair is generated. The generated electrons are transported to the n ⁺ diffusion layer 54 and the holes are transported outward through the p well 52.

The accumulated electrons try to escape through the place with the lowest potential barrier, but the part with the lowest barrier is not a normal transfer MOSFET channel but a load MOS.
This is a channel below the gate electrode 56 corresponding to the FET.
The channel below the gate electrode 56 is in a state showing a sub-threshold characteristic, and thus easily carries electrons, which are stored charges, to the power supply _VDD .

In this embodiment, unlike a sensor having a normal linear characteristic, the electric charge accumulated through the transfer MOSFET does not leak to the source follower amplifier side, and the blooming resistance and light resistance are improved.

An image pickup apparatus according to still another embodiment of the present invention will be described below.

The unit cell of the pixel portion of this embodiment has the same configuration as that shown in FIG. 2, but the manufacturing process conditions and device parameters of the CMOS sensor are different. Ie, p-well concentration is high concentration as 10 ¹⁸ cm ^-3, and the gate oxide film thickness of the load MOSFET29 is non-linear resistor is 1.5μm and thick.

In the equation (5), n is called an n-factor.

[0064]

(Equation 8) It is expressed as Here, C _DEP is the capacity of the depletion layer,
_G is the gate capacitance.

The capacity C _DEP is 1.42 × 10 per unit area.
^-7 F / cm ² , whereas the gate capacitance _CG is 2.
The n-factor is as large as 61.2 because it is about two orders of magnitude smaller, 36 × 10 ⁻⁹ F / cm ² .

This large n-factor makes V ₀ a large value (1.58 V) as can be seen from equation (5). This large value of V ₀ increases the slope of the photoelectric conversion characteristic of the logarithmically compressed portion of equation (4).

Now, if the increase in the optical output voltage V _DD -Vp when the photocurrent ip increases by one digit is S,

[0068]

(Equation 9) Is represented by

Now, since V ₀ = 1.58 V, the value of S is 3.63 V / 20 dB.

According to this embodiment, as described above, a 1.5% contrast ratio that can be detected by human eyes can be sufficiently detected without being buried in thermal noise. This gives the n factor

[0071]

(Equation 10) Has been achieved by choosing.

FIG. 6 shows still another embodiment of the present invention.
An equivalent circuit diagram of a unit cell of a CMOS sensor having a common amplifier for every four pixels is shown. The principle of operation will be described below.

[0073] The positive voltage is applied to the phi _RES wiring connected to the gate electrode of each MOSFET69-1～69-4 First, to conduct the MOSFET69-1～69-4,
By connecting each of the photodiodes 61-1 to 61-4 to the _VDD power supply, the charges in the photodiodes 61-1 to 61-4 are reset and removed.

Next, by opening the transfer gates 65-1 to 65-4, the source follower amplifier 62 is connected to the photodiodes 61-1 to 61-4, and indirectly connected to the _VDD power source, so that the source Follower amplifier 6
2 is reset.

Next, the transfer gates 65-1 to 65-4 are closed, and the φ _RES wiring is kept at the same potential as the _VDD power supply.
Each MOSFET 69-1 to 69-4 is a load MOSFET
To work as At that time, the MOSFET 69-1
～69-4, as shown Subthreshold, the V _TH is is set to be high (> V _DD). This starts the accumulation of the optical signal.

After a certain period of time has elapsed, the gate of the transfer gate of any pixel is opened to transfer the accumulated optical signal charges to the gate electrode of the source follower amplifier 62. At this time, the optical signal charge is logarithmically compressed by the function shown in the equation (7) as described above. Then select MO by φ _SEL
The SFET 63 is opened to transmit a signal in the form of a current to the vertical signal line 66.
Read from Next, the MOSFET 63 is closed, a positive voltage is applied to the MOSFET 69 of any pixel, and the gates of the photodiode 61 and the source follower amplifier 62 are reset to V _DD . Then, transfer gates 65-1 to 65-6
5-4 closed, after the gate potential of MOSFET69-1~69-4 was to load MOS mode to V _DD, reads in the same manner as the signals of other pixels.

In the present embodiment, the reset MOSFET shown in FIG.
The connection between the gate and the drain of the MOSFET corresponding to the ET becomes unnecessary, and the horizontal wiring between the cells is six (two upper and lower φ _RESs can be made one), which is more advantageous in layout. Becomes

The non-linear resistor according to the present invention has no load MOS.
The present invention is not limited to the FET, but may be a device utilizing a known forward characteristic of a diode. However, since the n-factor of a diode is generally small, it is desirable to use a material other than silicon in order to satisfy Expression (10).

FIG. 9 is a schematic configuration diagram of a unit cell and a signal readout circuit. The configuration of the unit cell is the same as the configuration shown in FIG. The unit cell S in FIG. 9 includes four photoelectric conversion units and one common amplifier. For example, the photoelectric conversion units a11, a12, a21, and a22 in FIG.
(Red), G1 (green), G2 (green), and B2 (blue) are provided, and R1 signal, G1 signal, B2 signal, G2
A signal is output from the common amplifier A1 via a vertical output line. The noise after reset is also output from the common amplifier A1 via the vertical output line. That is, before the signal is transferred from the photoelectric conversion unit to the input unit (gate) of the amplifying unit, the reset unit 24 is turned on by the reset signal applied to the wiring _φRES , and the input unit of the amplifying unit 22 is reset. Is sent to the vertical output line 26 via the selection means 23.

Reference numeral 100 denotes a capacitor C _S for signal storage, a capacitor C _N for noise storage, and transistors M 1 and M M for capacitance switching.
2, and a signal storage component composed of signal output transistors M3 and M4. One vertical output line is provided with a signal storage capacitor _CS and a noise storage capacitor _CN in parallel via transistors M1 and M2. The signals stored in the capacitor C _S and the noise stored in the capacitor C _N are read out to the respective horizontal output lines by simultaneously turning on the transistors M 3 and M 4 controlled by the horizontal shift register (H · SR) 13. , The noise is removed from the signal by the subtraction amplifier 10, the AGC (auto gain control) 11, the A / D
The signal is converted into a digital signal via the converter 12. The horizontal output line is reset by a transistor M6 controlled by φHC. It should be noted that a signal storage capacitor, a noise storage capacitor, a capacitance switching transistor, and a signal output transistor are provided for each of the photoelectric conversion units, and noise-removed R, G, and B signals are provided. Can also be output at the same time.

FIG. 10 shows an overall configuration diagram of the image pickup apparatus. Reset of each photoelectric conversion cell, readout of noise / signal, and control of photoelectric conversion are performed by vertical shift registers (V / SR)
15, the noise removal / signal accumulation component 10
Noise elimination / memory unit 14 composed of 0, 200, and 300
Is controlled by a horizontal shift register (H-SR) 13, the signal is amplified by an amplifier 10, and AGCs 11, A
The signal is converted into a digital signal via the / D converter 12.
The timing generator 16 has a vertical shift register (V
SR) 15, horizontal shift register (HSR) 13,
The operation of the amplifier 10, the AGC 11, and the A / D converter 12 is controlled. Reference numeral 17 denotes an image sensor unit in which photoelectric conversion cells are arranged in a matrix.

FIG. 11 is a schematic diagram of an imaging system. As shown in the figure, the image light incident through the optical system 71 and the stop 80 forms an image on the CMOS sensor 72. CMOS
The optical information is converted into an electric signal by a pixel array arranged on the sensor 72, and is output after noise removal. The output signal is subjected to signal conversion processing by a signal processing circuit 73 by a predetermined method and output.
The signal that has been subjected to the signal processing is recorded by a recording system and a communication system 74 by an information recording device, or information is transferred. Records,
Alternatively, the transferred signal is reproduced by the reproduction system 77. Aperture 80, CMOS sensor 72, signal processing circuit 7
3 is controlled by the timing control circuit 75, and the optical system 7
1. The timing control circuit 75, the recording / communication system 74, and the reproduction system 77 are controlled by the system control circuit 76.

[0083]

As described above, according to the present invention,
An imaging device suitable for the characteristics of the human eye can be easily realized.
In addition, it has a high color tone, is resistant to thermal noise, has excellent light fastness, and can easily obtain characteristics similar to the S-shaped characteristics of a silver halide film.

[Brief description of the drawings]

FIG. 1 is an equivalent circuit diagram for explaining the principle of the present invention.

FIG. 2 is an equivalent circuit diagram of a unit cell of a CMOS sensor having a common amplifier for every four pixels, which is an embodiment of the imaging apparatus of the present invention.

FIG. 3 is a characteristic diagram showing a photoelectric conversion characteristic of an example of the present invention.

FIG. 4 is a characteristic diagram showing a photoelectric conversion characteristic of an example of the present invention.

FIG. 5 is a schematic sectional view of a unit cell according to an embodiment of the present invention.

FIG. 6 is an equivalent circuit diagram of a unit cell of a CMOS sensor having a common amplifier for every four pixels, which is still another embodiment of the present invention.

FIG. 7 is an equivalent circuit diagram showing a configuration of one pixel of a CMOS sensor.

FIG. 8 is an equivalent circuit diagram of a unit cell of an imaging device in which four pixels share an amplifying unit (amplifier).

FIG. 9 is a schematic configuration diagram of a unit cell and a signal readout circuit.

FIG. 10 is a schematic configuration diagram illustrating an overall configuration diagram of an imaging apparatus.

FIG. 11 is a schematic configuration diagram illustrating an imaging system.

[Explanation of symbols]

REFERENCE SIGNS LIST 1 capacitance 2 constant current source 3 load MOSFET 51 substrate 52 p-well 53 oxide film 54 n ⁺ diffusion layer 55 n ⁺ diffusion layer 56 gate electrode 57 wiring 58 light-shielding film 21, 61, 101, 111 photodiode 22, 62, 102, 112 Source follower amplifier 23, 63, 103, 113 Selection MOSFET 24, 104, 114 Reset MOSFET 25, 65, 115 Transfer gate 26, 66, 116 Output line 29, 69 Load MOSFET

Claims (12)

[Claims] 1. An image pickup apparatus having, in a pixel, a photoelectric conversion element and a non-linear resistance element connected to the photoelectric conversion element, wherein the non-linear resistance element is provided during the accumulation period of the photoelectric conversion element. An imaging device for non-linearly converting an optical signal charge accumulated in a photoelectric conversion element. 2. A unit comprising at least a plurality of photoelectric conversion elements, a plurality of non-linear resistance elements respectively connected to the plurality of photoelectric conversion elements, and an amplifying means for amplifying signals from the plurality of photoelectric conversion elements. A cell is configured, wherein the amplifying unit is an imaging device shared by the plurality of photoelectric conversion elements, wherein the non-linear resistance element is stored in the photoelectric conversion element during a storage period of the photoelectric conversion element. Imaging device for performing non-linear conversion of optical signal charges to be converted. 3. The imaging device according to claim 1, wherein at least a part of the obtained photoelectric conversion characteristic has a logarithmic compression type region, and the inclination of the region is 2.8 V / 20.
An imaging device characterized by being at least dB. 4. The imaging device according to claim 1, wherein the non-linear resistance element is a load MOSFET. 5. The imaging device according to claim 4, wherein a value of an n-factor representing a sub-threshold characteristic of the load MOSFET is: An imaging device characterized by the above. 6. The imaging apparatus according to claim 1, wherein at least a part of the obtained photoelectric conversion characteristic has a logarithmic compression type region, and the logarithmic compression type characteristic is the photoelectric conversion characteristic. An imaging apparatus characterized by being located on a high frequency side of the image. 7. The imaging apparatus according to claim 6, wherein the voltage at which the logarithmic compression type characteristic starts is determined by a threshold voltage of a load MOSFET serving as the non-linear resistance element. 8. The imaging device according to claim 4, wherein the load MOSFET functions as an overflow drain for releasing excess charges generated when strong light is irradiated. 9. The imaging device according to claim 1, wherein the imaging device is a CMOS sensor. 10. The imaging device according to claim 2, wherein
An imaging apparatus wherein the non-linear resistance element also serves as reset means for resetting the photoelectric conversion element and the control electrode of the amplification means. 11. An image pickup apparatus having a photoelectric conversion element and a non-linear resistance element connected to the photoelectric conversion element in a pixel, wherein at least a part of the obtained photoelectric conversion characteristic has a logarithmically compressed region. And the inclination of the region is 2.8 V / 20 dB
An imaging device characterized by the above. 12. An image pickup apparatus having, in a pixel, a photoelectric conversion element and a non-linear resistance element connected to the photoelectric conversion element, wherein the non-linear resistance element is a load MOSFET, and a substrate of the load MOSFET. The value of the n-factor representing the threshold characteristic is: An imaging device characterized by the above.