JPH0477172A - Image sensor - Google Patents

Abstract

PURPOSE:To realize high sensitivity and high speed reading by leading a stored optical signal of a photodiode to a base of a transistor(TR) via an access switch, amplifying the charge, outputting the charge from a picture signal output line, conducting a reset switch succeedingly and resetting an anode electrode of the photodiode and the base of the TR to a voltage of a reset power supply. CONSTITUTION:Parallel output signals Y1, Y2, Y3, Y4, Y5 of a scanning circuit and a picture output signal Sig are shown in figure together with a scanning clock signal CK and a start signal ST. For example, when the signal Y1 is active and an access FET 2a is conductive, the storage optical signal charge of a photodiode 1a is transited to a transistor(TR) and it becomes a base current and a picture signal is outputted from a picture signal output line after amplification. The photodiode 1a and a base of the TR 4a are set to a voltage of a power supply in a timing when both succeeding Y1, Y2 are active, then the photodiode enters the storage time of the optical signal. The operation is sequentially transited to 1b, 1c, 1d similarly.

Description

[Detailed description of the invention] Industrial application field Image processing Although the need for high-performance image sensors is increasing with the advancement of image communication equipment, the present invention makes it possible to read document information at high speed and with high quality. Regarding image sensors, image sensors using conventional technology integrated circuits use photodiodes or phototransistors as light detection elements.When directly outputting photocurrent using the direct method, the sensitivity of a photodiode is approximately 5. nA/lx/mm - Approximately 100 nA/mm" for a phototransistor. Phototransistors have high sensitivity because the detection element itself has an amplification function. Also, as the resolution of image sensors increases, the light receiving area increases. In order to increase the sensitivity from the frame drive method, the direct method lacks sensitivity.In order to increase the sensitivity from the frame drive method, we use an accumulation method that accumulates photocurrent for a time corresponding to the output interval of the image signal and outputs it intensively at the timing of the signal output. The driving method is adopted by various image sensors.4 For example, in a CCD image sensor, the optical signal charge accumulated in the photodiode at the same time as the reading timing is transferred to the CCD.
Even if the image signal is obtained by guiding it to the potential well of
In an OS image sensor, a photodiode or phototransistor is charged with a certain amount of charge, and then discharged by the skin photocurrent until the next reading timing. After charging a certain amount of charge, it is discharged by a blocking current until the next reading timing, and at the next recharging timing, it is amplified by the transistor function. The storage method used is the highest. % An equivalent circuit of a conventional image sensor using a phototransistor as a light detection element is shown in FIG. Even if the accumulated signal charge in the phototransistor 9 is guided to the image output line by making the access switch 2 conductive, and the emitter potential is reset to zero by subsequently making the reset switch 3 conductive, this method has a simple circuit. The problem is that the time constant for recharging is long, the afterimage is large, and high-speed reading is difficult.A Problem that the invention aims to solve When using a phototransistor as a light sensing element and operating in its charge accumulation mode, As shown in Figure 2, the umbilical base electrode is floating and charges the collector-base junction capacitance.In this case, a base-emitter junction is inserted in series with the charging circuit, which creates a nonlinear resistance and reduces exposure. In high-speed scanning, the afterimage rate increases rapidly in the area where the charging time cannot be taken for a long time. A means to solve the problem is to construct an image sensor from each pixel consisting of at least a photodiode, an access switch, a reset switch, and a transistor, and a scanning circuit section.The cathode of the photodiode is used as a common terminal between pixels, and the anode is accessed for each pixel. Connect it to one terminal of the switch, connect the other terminal of the access switch and one terminal of the reset switch in common, and connect it to the base terminal of the transistor at the same time.The other terminal of the reset switch is connected in common between pixels. The collector of the transistor is connected to the positive power supply, and the emitter terminal is commonly connected between pixels to form an image signal output line.The access and reset switches are field effect transistors (
The anode of the photodiode is set to a constant reset voltage by simultaneously making the access switch and the reset switch conductive. After that, the access switch and reset switch are made non-conductive during the storage time, and the charge accumulated when the photodiode is reset is discharged by the photodiode's photo-electric conversion current, but as a result, the voltage across the photodiode terminals decreases. . The amount of decrease in the voltage between the terminals or the amount of decrease in charge is a force proportional to the amount of irradiated light (by making the access switch conductive, the image signal is received at the base of the transistor and output after amplification to the emitter terminal, that is, the image signal output line). The base area of the transistor is smaller than that of a phototransistor, and the reset time is shorter than that of a phototransistor.Therefore, there is little afterimage even during high-speed reading, and high-sensitivity reading is possiblea. An embodiment will be described with reference to the drawings. Fig. 1 is an equivalent circuit diagram of an image sensor according to the present invention. Each pixel is a photodiode 1 (la, 1b to 1d).
, access FET 2 (2a, 2b to 2d), reset FET 3 (3a, 3b to 3d), and transistor 4 (4a, 4b to 4d). 8 is a scanning circuit consisting of a shift register, and a start signal (
ST), a clock signal (CK) is input to generate a pulse that operates the access and reset FETs. The emitter electrodes of each transistor are commonly connected to form the image signal output line 6, and the source electrodes of the reset FETs are commonly connected to form the input terminal 7 of the reset power supply.Next! 2. Operation of the image sensor according to the present invention will be explained. FIG. 3 is an operation timing chart of the image sensor according to the present invention, in which parallel output signals Y1, Y'1 .
Y3. Y4. Even if Y5 and the image output signal Sig are shown in the diagram, even if 1iYl becomes active and the access FET becomes conductive, the accumulated optical signal charge of the photodiode 1a moves to the transistor, this becomes the base current, and after amplification, the image signal is output from the image signal output line. Even if Yl and Y2 are output, the bases of photodiode 1a and transistor 4a are set to the voltage value Vrs of the reset power supply at the next timing when both Yl and Y2 become active. After that, the photodiode is activated during the optical signal accumulation time. This operation is performed sequentially. Similarly, move to 1b and IC11d. In the circuit of Figure 1 and the timing chart of Figure 3,
The pixel reset and the second pixel access have the same force (
Although the scanning circuit becomes complicated, it is also possible to separate the access timing and reset timing for each pixel.

In the present invention, the photodiode and transistor are separate elements, and each is completely reset to Vrs, so afterimages due to the history of read images can be removed.
When high-speed reading is required, the response speed of the transistor is particularly important.In this case, the voltage of the image signal output line is kept constant and the image signal is output as a current signal, and Reset power supply voltage V
By setting rS 0.1 to 0.5 V higher, the response speed of the transistor becomes faster and high-speed reading becomes possible.

Figure 4 shows the access switch with P channel FET and L
The planar (a) and cross-sectional (b) structures of a pixel are shown when the source of the FET is shared with the diffusion layer of the photodiode and the drain is shared with the base diffusion layer of the transistor.

By doing this, the required area can be reduced and a compact design is possible. Effects of the Invention The present invention has made it possible to significantly reduce image sensor afterimages, resulting in high sensitivity and high speed. It is possible to read
Its industrial effects are similar to that of dogs.

[Brief explanation of the drawing]

FIG. 1 is an image sensor circuit according to the present invention. FIG. 2 is a conventional image sensor circuit 1. FIG. 3 is an operation timing chart of an image sensor according to the present invention. FIGS. 4(a) and (b) are respectively Figures 1a to 1d are a plan view and a cross-sectional view of the pixel section when the access switch is configured with a P-channel FET, the source of the FET is shared with the diffusion layer of the photodiode, and the drain is shared with the base of the transistor. Transis Ku 2a~2d...
・Access switch, 3a to 3d...Reset switch, 4a to 4d...Transistor connection 5...Positive power line, 6...Image signal output line, 7...Reset power line, 8...Scanning Pictures i9a, 9b~
9d... Phototransistor. Name of agent: Patent attorney Shigetaka Awano

Claims (3)

[Claims] (1) Consisting of at least each pixel consisting of a photodiode, an access switch, a reset switch, and a transistor, and a scanning circuit section, the accumulated optical signal of the photodiode is guided to the base electrode of the transistor via the access switch, and after amplification, the image signal is An image sensor characterized in that an anode electrode of a photodiode and a base electrode of a transistor are reset to a voltage value of a reset power supply by outputting from an output line and subsequently turning on a reset switch. (2) The voltage of the image signal output line is kept constant, the image signal is output as a current signal, and the voltage of the reset power supply is set 0.1 to 0.5 V higher than the voltage of the image signal output line. The image sensor according to claim (1). (3) The access switch according to claim (1), characterized in that the access switch is composed of a P-channel FET, the source of the FET is shared with the anode diffusion layer of the photodiode, and the drain is shared with the base diffusion layer of the transistor. image sensor.