JP3514663B2 - Semiconductor imaging device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a semiconductor image pickup device that outputs a voltage according to the amount of received light, and is used in an image pickup apparatus used in various environments such as an outdoor surveillance camera and a vehicle-mounted camera. The present invention relates to an image sensor.

[0002]

2. Description of the Related Art In the human visual system, each photoreceptor cell outputs the difference between the amount of light received by itself and the average of the amount of light received by each peripheral photoreceptor cell as a signal for the received light quantity. Have a function. With this function, FIG.
As shown in, each photoreceptor cell has a constant width of the light receiving sensitivity range, but shifts the light receiving sensitivity range according to the average of the light intensity incident on the periphery thereof. That is, even when subjects with extremely different brightness are present in the same visual field, the photoreceptor cells that see a bright subject shift their light-receiving sensitivity range to the bright side, and the photoreceptor cells that see a dark subject shift to the dark side. Due to such a function of shifting the light receiving sensitivity range, the human visual system as a whole realizes a light receiving sensitivity range wider than the light receiving sensitivity of individual photoreceptor cells.

Further, since each photoreceptor cell outputs a difference from the output of photoreceptor cells in the vicinity thereof, the human visual system strongly reacts to the portion where the brightness change exists. For this reason, the human visual system automatically extracts the outline of the subject, and recognizes it by giving a high contrast to the object being viewed.

On the other hand, when a solid-state image sensor such as a CCD or CMOS imager is used to photograph an object in which a subject having extremely different brightness exists in the same field of view, the spread of the brightness distribution of the object is sufficiently wide by human vision. Even within a recognizable range, it is often found that a bright subject causes halation or a dark subject is crushed in black in an image.

A solid-state image pickup device such as a CCD or a CMOS imager has a structure in which a plurality of functional units (pixels) for converting light incident on the device into an electric signal according to light intensity and exposure time are arranged. . Generally, each pixel has the same light receiving sensitivity range, and the light receiving sensitivity range is fixed. The light receiving sensitivity range of the element is also equal to the light receiving sensitivity range of each pixel. Therefore, when an object having a brightness spread exceeding the light receiving sensitivity range of each pixel is to be photographed, the above-described phenomenon such as halation of a bright subject or blackout of a dark subject occurs. Further, in the image captured by the solid-state image sensor, the contour of the object is unclear as compared with the case where a human looks at the same object. For this reason, in many cases, the edge enhancement processing for clarifying the contrast is required before displaying the image on a cathode ray tube or the like.

The conventional solid-state image pickup device is far inferior to the human visual system in the light receiving sensitivity range and the contrast detecting function. Considering that the conventional solid-state image sensor is used as a substitute for the human vision such as a surveillance camera, or when the image is recorded and then viewed by a human, the conventional solid-state image sensor receives light. It is clear that the performance is insufficient in terms of sensitivity range and contrast detection function.

Therefore, in order to realize a wide light receiving sensitivity range, studies have been made so far to widen the width of the light receiving sensitivity range of each pixel.

As an example, there is a method of expanding the dynamic range by superimposing images taken at different exposure times. For example, there is a method disclosed in "Wide Dynamic Range Imaging Technology" by Hirohito Kobuchi (Proceedings of the 7th Image Input Technology Symposium, pp.85-92). With this method, when shooting a subject with extremely different brightness in the same field of view, you can use a long exposure time to shoot a dark subject and a short exposure time to shoot a bright subject. By superimposing it on the captured image, both bright and dark subjects are displayed on the image without halation or blackout.

[0009] Alternatively, there is a study to expand the light receiving sensitivity width of each pixel itself by utilizing the nonlinear input / output characteristics in the subthreshold region of the transistor. For example, S.
G. SG Chamberlain and J. P.
Y. "A Novel Wide Dynamic Range Silicon Photodetector" by Lee (JPY)
and Linear Imaging Array) ”(IEEE Trans on ED, v
ol.ED-31, No.2, pp.175-182, Feb., 1984).

[0010]

However, regardless of the method, the method of enlarging the dynamic range of each pixel and the light receiving sensitivity width has a problem that the contrast of the entire screen becomes weak when displaying a photographed image. Occur. Therefore, a method has been proposed in which an image is divided into several areas and a large number of gradations of the display system are given to the brightness range concentrated in each area. For example, there is "Wide dynamic range image synthesis processing technology" by Atsushi Morimura, Takeo Azuma, and Kenya Uomori (Journal of the Institute of Image Information and Television Engineers vol.51, No.2, pp.228-232). However, this method has a problem that complicated post-processing of images is required.

A solid-state image pickup device, such as the human retina, which has a function of shifting the light-receiving sensitivity range of each pixel according to the output of peripheral pixels has already been described in C.I. Prototype by C. Mead ("Adaptive Retin (Adaptive Retin
a) ”(Analog VLSI Implementations of Neural Syste
ms, 1989, pp.239-246))). One pixel circuit which constitutes this element is shown in FIG. As shown in this figure, one pixel circuit includes a photodetector PD and an operational amplifier OP, and is connected to an adjacent pixel via a resistor R.
The pixel circuit configured in this way averages the output with the surrounding pixels via the resistance network of the resistor R, and differentiates the difference between the average of the peripheral pixel output and its own output. The operation is performed using the amplifier OP. In this device, each pixel shifts the light receiving sensitivity range like a photoreceptor cell of the human visual system, thereby realizing a wide light receiving sensitivity range as the entire device. Since the light receiving sensitivity range of each pixel is not widened, there is no problem that the contrast of the entire screen is lowered due to the widening of the light receiving sensitivity range in the display system. However, in this device, an operational amplifier O
Since P is required, there is a practical problem that the circuit scale becomes large and the power consumption also becomes large.

In the future, it is expected that the demand for semiconductor image pickup devices used in an image pickup device having a wide light receiving sensitivity range comparable to that of the human retina and a high contrast detection function and a high environmental adaptability will increase more and more.

The present invention has been made to solve the above problems, and an object of the present invention is to realize a solid-state image pickup device which can be highly integrated, has low power consumption, and has a wide light receiving sensitivity range and a high contrast detecting function. To provide.

[0014]

In order to solve the above problems, the semiconductor image pickup device of the present invention has the following configuration. A first semiconductor image sensor according to the present invention is a semiconductor image sensor in which a plurality of pixel circuits are connected, and the pixel circuit includes a first photodetector element, a second photodetector element, and a first resistance element. However, the first photodetection element and the second photodetection element are the first
Are connected in series via the resistance element. Each pixel circuit is connected to an adjacent pixel circuit via a second resistor at a connection node between one of the two photo detection elements and the first resistance element.

A second semiconductor image pickup device according to the present invention is such that, in the first semiconductor image pickup device, both ends of the first resistance element of the pixel circuit are set to different capacitance values.

According to a third semiconductor image pickup device of the present invention, in the first semiconductor image pickup device, at least one of the first resistance element and the second resistance element in the pixel circuit is constituted by a MOS transistor, The resistance value of the resistance element is set according to the gate voltage of the MOS transistor.

According to a fourth semiconductor image sensor of the present invention, in the third semiconductor image sensor, the gate voltage for setting the resistance value of the resistance element is at least two for each of the first and second resistance elements. It is commonly applied to the above pixel circuits.

A fifth semiconductor image pickup device according to the present invention is characterized in that, in the first semiconductor image pickup device, a reset voltage for initializing a state of a pixel circuit is applied to at least one end of the first resistance element. The semiconductor image sensor according to claim 2.

According to the sixth semiconductor image pickup device of the present invention, in the fifth semiconductor image pickup device, the reset voltage is commonly applied to at least two or more pixel circuits.

[0020]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of a semiconductor image pickup device according to the present invention will be described in detail below with reference to the accompanying drawings.

Embodiment 1. FIG. 1 shows a circuit configuration of one pixel which constitutes the semiconductor image sensor of the first embodiment. As shown in FIG. 1, one pixel circuit includes two photodetectors PD1 and PD2, a resistor Ri, an amplification transistor Tr3, a readout transistor Tr4, and switches SW1 and SW2. Photo detector PD1,
The PD 2 is a photo-detecting element including a photodiode and the like, and generates a current according to the intensity of the received light. The two photo detectors PD1 and PD2 have the same light receiving sensitivity characteristic.

The photodetector PD1 and the photodetector PD2 are connected in series via a resistor Ri so as to be reverse-biased between a power supply for supplying a power supply voltage Vdd and a ground GND for supplying a reference potential. The gate of the amplifying transistor Tr3 is connected to a node 2 connecting the resistor Ri and the photodetector PD2, and the reading transistor Tr4 is connected in series to the amplifying transistor Tr3. A resistor Rn is connected to a node 1 which connects the photodetector PD1 and the resistor Ri. A switch SW1 and a switch SW2 are connected to the node 1 and the node 2, respectively. When these switches SW1 and SW2 are turned on, a predetermined reset voltage Vpdc is applied to the node 1 and the node 2 so that they are initialized. It has become.

FIG. 2 is a diagram for explaining the connection between pixels in the image element. In the image sensor as shown in the figure,
A plurality of pixel circuits 11 are connected in a matrix, and one pixel circuit 11 is connected to a node 1 of an adjacent pixel circuit via a resistor Rn at a node 1.

FIG. 3 is a circuit diagram showing the pixel circuit shown in FIG. 1 in more detail. As shown in this figure, in the present embodiment, the switches SW1 and SW2 are composed of PMOS transistors Tr1 and Tr2. The resistors Ri and Rn are realized by diffused resistors or high resistance polysilicon.
Further, in the pixel circuit 11, a signal line 21 for extracting an output signal connected to one end of the readout transistor Tr4, a control line 23 for giving a reset voltage Vpdc, and a control line 25 for controlling the transistors Tr1 and Tr2, A control line 27 for controlling the read transistor Tr4 is wired.

The operation of the pixel circuit thus configured will be described. For convenience of description, the resistance values of the resistors Ri and Rn are represented as Ri and Rn, and the voltage of each node is ground GN.
The potential of D is used as a reference.

In the pixel circuit 11 shown in FIG. 1, when light is incident on the reverse-biased photodetectors PD1 and PD2, a current is generated according to the amount of incident light (or the intensity of incident light), which causes a reverse current to the resistor Ri. The directional current Id flows, and a potential difference occurs between the nodes 1 and 2, that is, the resistor Rn. Further, since the pixel-to-pixel connection via the resistor Rn causes the charge to move between the nodes 1 of the peripheral pixels via the resistor Rn, the potential of the node 1 is averaged between the neighboring pixels. As a result, the potential of the node 2 becomes a potential representing the difference between the amount of light received by the pixel and the average value of the amount of light received by the pixels adjacent to the pixel. The potential change of the node 2 is amplified by the transistor T for amplification.
It is amplified by r3 and taken out as an output signal via the read transistor Tr4. The above operation will be described in more detail below.

The pixel circuit is reset at a predetermined timing before outputting the image signal (the timing of this reset operation will be described later). Specifically, the resistance Ri
Nodes 1 and 2 at both ends of the switch are switches SW1 and SW
When 2 is turned on for a predetermined time, a predetermined initial voltage Vpdc is applied as a reset voltage to both nodes 1 and 2, and the voltages of node 1 and node 2 are reset to voltage Vpdc. The reset voltage Vpdc has a value between the power supply voltage and 0V. After that, the photo detectors PD1 and P
When D2 receives light, the potentials of the nodes 1 and 2, that is, the voltages V1 and V2 of the nodes 1 and 2 change according to the amount of received light.

In explaining the changes in the voltages V1 and V2 at the nodes 1 and 2 of the pixel circuit shown in FIG.
In the circuit shown in FIG. 1, an amplifying transistor Tr3 and a resistor R
Changes in the node voltages V1 and V2 when the connection of n is not taken into consideration will be described. That is, changes in the voltages V1 and V2 at the nodes 1 and 2 in the circuit shown in FIG. 4 will be described.

FIG. 5 is a diagram showing changes in the voltage V1 at the node 1 and the voltage V2 at the node 2 with respect to the incident light amount D in the pixel circuit shown in FIG. Note that FIG. 5 shows a state after the time t from the end of the reset has sufficiently passed. The broken lines show the states of the voltages V1 and V2 when the time t is almost zero.

As shown in FIG. 5, the value of the voltage V1 changes according to the amount D of light incident on the pixel, and the reset voltage V1
The value becomes higher by the voltage value obtained from pdc by the product of the current Id and Ri / 2. Since the current Id is proportional to the incident light amount D, the difference between the voltage V1 and the reset voltage Vpdc is the light amount D.
Increases in proportion to. Similarly, the voltage V2 also changes its value according to the incident light amount D, and the reset voltage Vpdc changes to the current Id.
And Ri / 2, the voltage value obtained is lower. That is, the voltage V2 decreases at a constant rate in proportion to the light amount D.

Further, the longer the elapsed time t from the time of reset, the larger the absolute value of the slope of the voltages V1 and V2.
That is, immediately after the completion of the application of the reset voltage Vpdc (the elapsed time t is substantially 0), both the voltage V1 and the voltage V2 are substantially equal to the reset voltage Vpdc, and as the time t elapses, the voltage from the reset voltage Vpdc is changed according to the time constant of the circuit. The difference is increasing. Therefore, when reading the output signal from the pixel circuit, the reading should be performed after the time t has sufficiently passed.

Next, changes in the node voltages V1 and V2 in consideration of the connection of the amplifying transistor Tr3 to the node 2 will be described with reference to FIG.

A parasitic capacitance Cm exists at the gate of the amplifying transistor Tr3. Therefore, as shown in FIG. 6, the gate capacitance Cm of the transistor Tr3 is equal to the node 2 connected to the node 2. Further, since the two photodetectors PD1 and PD2 also have parasitic capacitance Cpd,
The capacity Cpd is connected to the node 1, and the capacity Cm is connected to the node 2 in addition to the capacity Cpd. Therefore, the capacitance values connected to the node 1 and the node 2 are different, and the node 2 has a larger capacitance than the node 1.

FIG. 7 shows the voltage V1 at the node 1 and the voltage V2 at the node 2 when the time t from the reset time in the pixel circuit shown in FIG. 6 has sufficiently elapsed. Also in this case, FIG.
Similar to the case shown in (1), the difference between the voltage V1 and the voltage V2 becomes a value according to the product of the current Id and the resistance value Ri. However, unlike the case shown in FIG. 5, the voltage V2 with respect to the light amount D is
The absolute value of the slope of the voltage V1 is small, and the absolute value of the slope of the voltage V1 is large. This is because the node 2 has a larger capacity (Cpd + C) than the capacity (Cpd) connected to the node 1.
m) is connected. Also in this circuit configuration, when the elapsed time t is almost 0, both the voltage V1 and the voltage V2 are substantially equal to the reset voltage Vpdc as shown by the broken line, but the voltage constants are different because the time constants of the node 1 and the node 2 are different. The absolute values of the slopes of V1 and voltage V2 are different. That is, since the node 2 having a larger connected capacity than the node 1 changes more slowly than the node 1, the light amount D
The absolute value of the slope of the voltage V2 with respect to is small.

With reference to the above points, in the pixel circuit shown in FIG. 1, the voltage V1 of the nodes 1 and 2 with respect to the amount of incident light,
The change in V2 will be described.

The pixel circuit shown in FIG. 1 is obtained by further connecting the pixel circuit shown in FIG. 6 to an adjacent pixel at a node 1 via a resistor Rn. In this way, by connecting the nodes 1 between the adjacent pixels via the resistor Rn, the electric charges move between the adjacent pixels via the resistor Rn, so that the potential of the node 1 is equal to the potential of the node 1 of the adjacent pixels. It becomes the value. The range of neighboring pixels in which the potential is averaged can be adjusted by the resistance value of the resistor Rn, and the averaged range can be narrowed by increasing the resistance value Rn.
The range can be widened by decreasing. Also,
The sensitivity of the pixel circuit can be adjusted by changing the value of the resistor Ri.

Now, consider a case where the amount of incident light on pixels within one averaged range is distributed in the range of the amount of light a to the amount of light b. When there is no pixel-to-pixel connection by the resistor Rn, the node voltages V1 and V2 with respect to the incident light amount D change as shown by solid lines V1 'and V2' in FIG. That is, when there is no pixel-to-pixel connection, when the light amount a is received, the voltage V of the node 1 becomes V
1 becomes the voltage value V1a, and when the light amount b is received, the voltage V1 becomes the voltage value V1b.

However, in the case where there is a pixel-to-pixel connection by the resistor Rn, the node 1
Voltage V1 is averaged between nodes 1 of neighboring pixels,
This average value V1c (this value is determined by the distribution of the light quantity) regardless of the incident light quantity. That is, the voltage V1 of the node 1 when receiving the light quantity a is the difference ΔVa (= V1c−V1) from the average value of the voltage V1a when there is no pixel-to-pixel connection.
The voltage V1 of the node 1 when shifted upward by a) and receiving the light amount b is the voltage V1b when there is no pixel-to-pixel connection.
From the average value is shifted downward by ΔVb (= V1b−V1c).

As a result, the voltage V2 at node 2 is
The voltage shifts according to the change in the voltage V1. That is, the voltage V2 of the node 2 is
When it receives the light amount a, it shifts upward by ΔVa, and when it receives the light amount b, it shifts downward by ΔVb. When the pixel receives a light amount equal to the average value m of the received light amount of the neighboring pixels, the voltage V1 becomes the same value as when there is no inter-pixel connection, and the voltage V2 also has the same value as when there is no inter-pixel connection. Becomes

That is, in the case where there is pixel-to-pixel connection, the voltage of the node 2 in one pixel is a value representing the difference between the average value of the amount of light received by a pixel in the vicinity thereof and the amount of light received by that pixel.

As described above, by connecting the node 1 of each pixel to each other through the resistor Rn, the voltage V2 of the node 2 connected to the gate of the amplifying transistor Tr3 is the average value of the amount of light received in the vicinity of the pixel. As a reference, a signal indicating the difference between the amount of light received by the pixel and the average value of neighboring pixels is shown. Here, the reference voltage value is equal to the voltage value V2 of the node 2 when there is no pixel-to-pixel connection, and the difference between this voltage V2 and the reset voltage Vpdc increases as the light amount increases (see FIG. 8). ). This means that the voltage V
It is shown that 2 depends on the amount of incident light and that it does not change uniformly depending on the difference from the average value. It is preferable that the component dependent on the amount of received light due to this shift be small, and the shift between the voltage V2 and the reset voltage Vpdc can be reduced by increasing the capacitance value connected to the node 2. In the present embodiment, by utilizing the gate capacitance of the amplifying transistor Tr3, the capacitance value connected to the node 2 is increased as compared with the node 1 and the light amount dependent component of the output signal is reduced. Furthermore, the capacitance value connected to the node 2 may be increased by separately connecting a capacitor having a predetermined capacitance value to the node 2.

FIG. 9 is a diagram showing an example of changes in the voltage V2 at the node 2 with respect to the incident light quantity D in each light quantity region. In FIG. 9, for example, the solid line P shows the change in the voltage V2 of the node 2 when the light receiving amount range of the neighboring pixels is from the light amount a to the light amount b. At this time, the voltage V2 indicates a difference from the average light amount with the average light amount m as a reference. Similarly, the solid line P'indicates a change in the voltage V2 of the node 2 when the received light amount range of the neighboring pixels is from the light amount a'to the light amount b '. As described above, the pixel circuit according to the present embodiment can shift the light receiving sensitivity range according to the amount of ambient light as in the case of the human retina.

FIG. 10 shows the voltage V2 of the node 2 shown in FIG.
FIG. 3 is a diagram showing an output signal obtained by reading out by the amplifying transistor Tr3 and performing inverting amplification. As shown in this figure, since a large amplitude can be assigned to the limited range of the light amount a to the light amount b, a high contrast detection function is realized. By changing the amplification factor of the amplifying transistor Tr3, the change (slope) of the signal with respect to the change of the light amount D can be changed, and the contrast can be adjusted. However, the limit of the amplification factor and the light receiving sensitivity detection area is limited by the slope of the voltage V2 with respect to the light amount D when there is no inter-pixel connection.

FIG. 11 is a timing chart of control signals for the above pixel circuit. 11A shows a reset signal for resetting the node 1 and the node 2, and FIG. 11B shows a read control signal that gives a timing for reading the output voltage. As shown in FIG. 11A, the reset signal is set to L during the time T0 to T0 '.
ow (0V). As a result, the transistors Tr1 and T
r2 is turned on, and nodes 1 and 2 are connected via the control line 23.
The reset voltage Vpdc is applied to. After that, the voltage V1 of the node 1 is averaged between the adjacent pixels through the resistor Rn from the time T0 ′ to the time T1 in accordance with the light receiving operation of the photodetectors PD1 and PD2. Next, when the read control signal is set to High (power supply voltage Vdd) at time T1, the voltage V2 of the node 2 is converted into a current by the transistor Tr3, and is taken out through the read transistor Tr4 and the signal line 21. The reset operation and the read operation are repeatedly performed by performing the above operation for each pixel circuit in a predetermined cycle. in this way,
By resetting the pixel circuit before the reading operation, the reference when outputting the voltage according to the light amount can be always kept constant, and stable light amount-voltage conversion can be performed.

The reset signal may be applied to only one of the node 1 and the node 2. Further, it is preferable that the reset signal is commonly given to a plurality of pixel circuits so that each pixel operates uniformly among the pixels to be averaged. It is preferable to apply a constant value reset signal to at least two adjacent pixel circuits. This can improve the accuracy of the output signal obtained from the semiconductor image sensor.

As described above, the image sensor of this embodiment is
A pixel circuit having a series circuit including two diodes and one resistor is provided, and each pixel circuit is connected to an adjacent pixel circuit via the resistor. As described above, since the pixel circuit can be configured without using a differential amplifier as in the conventional case, the image sensor can be realized with a simple configuration, and the light receiving sensitivity range of each pixel is the same as that of the human retina. Since the shift can be performed according to the brightness, a wide light receiving sensitivity range and high contrast detection can be realized.

Embodiment 2. FIG. 12 shows the configuration of a pixel circuit that constitutes the semiconductor image sensor of the second embodiment. In this embodiment, the resistors Rn and Ri of the pixel circuit according to the first embodiment are MOS transistors. That is, as shown in FIG. 12, a resistor Rn that couples the pixels is used.
Is a PMOS transistor Trn, and the resistance Ri inserted between the photodetectors PD1 and PD2 is PM.
It is composed of an OS transistor Tri. Note that the pixel circuit of this embodiment operates similarly to the pixel circuit of Embodiment 1.

The resistance value realized by the MOS transistors Trn and Tri can be adjusted by controlling the respective gate voltages. Specifically, the resistance value of the transistor Trn can be adjusted by the coupling bias voltage Vn, and the resistance value of the transistor Tri can be adjusted by the correction bias voltage Vi. Bias voltages Vn and Vi are given by control lines 29 and 31, respectively, and their values are fixed during circuit operation.

The bias voltages Vn and Vi for adjusting the resistance values of the transistors Trn and Tri are set to be the same between neighboring pixels to be averaged. For example, it is preferable to apply common bias voltages Vn and Vi to at least two of the adjacent pixels. As a result, the resistance value of each pixel becomes the same value in the averaged pixel range, and the characteristics of each pixel become uniform.

As described above, by forming the resistors Rn and Ri with transistors, it can be realized easily and in a small scale as compared with the case where the resistance element is formed on the semiconductor substrate. In addition, the sensitivity and averaging range can be changed by changing the gate potential of the transistor, which reduces the burden of setting the resistance value at the time of design, and can be adjusted according to the imaging target even when mounted on the imaging device. Easy to do.

[0051]

According to the first semiconductor image pickup device of the present invention, one pixel circuit is composed of a series circuit including two photosensors and one resistor, and each pixel circuit is connected via the resistor. . This makes it possible to realize an image pickup device having a mechanism in which each pixel shifts the light receiving sensitivity range in accordance with the amount of light incident on peripheral pixels, such as a human retina, with a simple circuit configuration. Further, since a resistor is used between the photosensors, it is possible to easily change the performance such as the sensitivity of the semiconductor image sensor by changing the resistance value.

According to the second semiconductor image sensor of the present invention,
By setting different capacitance values at both ends of the resistance element, it is possible to suppress the dependency of the reference value on the incident light amount when determining the difference between the average light receiving amount of the peripheral pixels and the light receiving amount of that pixel. It is possible to output a voltage representing the difference between the average amount of light received by the peripheral pixels and the amount of light received by the pixels, without being affected by. Thus, it is possible to realize a mechanism for shifting the light receiving sensitivity range according to the amount of incident light.

According to the third semiconductor image sensor of the present invention,
Since the resistance element is composed of MOS transistors, a semiconductor image pickup element can be manufactured without increasing the circuit scale, and the resistance value can be easily changed by controlling the gate voltage of the MOS transistor, so that sensitivity adjustment is easy. Also, the load at the time of designing the resistance value is reduced.

According to the fourth semiconductor image sensor of the present invention,
A gate voltage is commonly applied to the peripheral pixels to make the resistance value a constant value. This enables uniform operation of each pixel in the peripheral pixels and improves the accuracy of the semiconductor image sensor.

According to the fifth semiconductor image sensor of the present invention,
Since a reset voltage can be applied to initialize each pixel circuit, a stable output can be obtained.

According to the sixth semiconductor image sensor of the present invention,
By applying the reset voltage in common to the peripheral pixels, each pixel can be uniformly operated in the peripheral pixels, and the accuracy of the semiconductor image sensor can be improved.

[Brief description of drawings]

FIG. 1 is a circuit diagram of a pixel forming a semiconductor image sensor according to a first embodiment of the present invention.

FIG. 2 is a diagram illustrating a state in which the pixel circuits according to the first embodiment are arranged in an array.

FIG. 3 is a more detailed circuit diagram of the pixel circuit according to the first embodiment.

FIG. 4 is a circuit diagram of a pixel circuit in the pixel circuit according to the first embodiment in which inter-pixel connection and a read transistor are not considered.

5 is a diagram showing a relationship between an incident light amount of the pixel circuit shown in FIG. 4 and each node voltage.

FIG. 6 is a circuit diagram in which the capacitance of each part is clearly shown in the pixel circuit according to the first embodiment.

7 is a diagram showing the relationship between the amount of incident light and the voltage of each node of the pixel circuit shown in FIG.

FIG. 8 is a diagram showing a relationship between an incident light amount and each node voltage in a pixel circuit connected between adjacent pixels.

FIG. 9 is a diagram showing an example of a relationship between a voltage at a node connected to a gate of a reading transistor and an amount of incident light in the pixel circuit of Embodiment 1.

10 is a diagram showing a relationship between an incident light amount and an output voltage when the node voltage shown in FIG. 9 is inverted and amplified and taken out as a read signal.

11 is a timing chart of control signals of the pixel circuit of Embodiment 1. FIG.

FIG. 12 is a circuit diagram of a pixel circuit included in the semiconductor image sensor according to the second embodiment.

FIG. 13 is a diagram showing how local sensitivity is automatically adjusted in the human retina.

FIG. 14 is a circuit diagram of a pixel having a function of shifting a conventional light receiving sensitivity range.

[Explanation of symbols]

1, 2 node, 11, 11a pixel circuit, Cpd,
Cm capacity, PD1, PD2 photo detector,
Rn resistance, Ri resistance, SW1, SW2 switches,
Tr1, Tr2, Tri, Trn MOS transistors, T
r3 amplification transistor, Tr4 readout transistor

Front page continued (58) Fields surveyed (Int.Cl. ⁷ , DB name) H01L 27/146 H01L 31/10 H04N 5/335

Claims (6)

(57) [Claims] 1. A semiconductor image pickup device comprising a plurality of pixel circuits connected to each other, wherein the pixel circuit has a first photo-detecting element, a second photo-detecting element and a first resistance element. A detection element and a second light detection element are connected in series via a first resistance element, and each of the pixel circuits connects one of the two light detection elements and the first resistance element. A semiconductor image pickup device characterized in that it is connected to an adjacent pixel circuit via a second resistance element at a node. 2. The semiconductor image sensor according to claim 1, wherein both ends of the first resistance element are set to different capacitance values in the pixel circuit. 3. In the pixel circuit, at least one of the first resistance element and the second resistance element is composed of a MOS transistor, and the resistance value of the resistance element depends on the gate voltage of the MOS transistor. The semiconductor image pickup device according to claim 1, wherein the semiconductor image pickup device is set according to claim 1. 4. The gate voltage for setting the resistance value is commonly applied to at least two or more pixel circuits for each of the resistance elements of the first and second resistances. The semiconductor image pickup device described. 5. The semiconductor image sensor according to claim 1, wherein a reset voltage for initializing a state of the pixel circuit is applied to at least one end of the first resistance element. 6. The semiconductor image sensor according to claim 5, wherein the reset voltage is commonly applied to at least two or more pixel circuits.