JPH0310562A - Signal correction device - Google Patents

Abstract

PURPOSE:To extract an accurate picture signal from which noise is eliminated by providing a correction means applying a voltage corresponding to noise detected by a noise detection means to a signal input line as a reset correction voltage. CONSTITUTION:A switching element S1 is turned on by the operation of a shift register or the like constituting a drive circuit and a reset noise output by switching noise is fetched in a sample-and-hold circuit of a noise detection means 3 and sampled and held synchronously with the clock pulse (CK). Moreover, the signal is synchronized with a pulse, an inverting signal inverted by an inverter of a correction means 6 is processed into unity gain to obtain a correction voltage and the voltage is fed to a point X of a signal input line 10. The operation is implemented for each picture element to eliminate the reset noise output as to all picture elements. Since the level is a reset level corrected specific to the picture element from the next readout, the succeeding output (especially dark output) from the signal output line 2 is suppressed uniform and low.

Description

Detailed Description of the Invention (Field of Industrial Application) The present invention relates to a signal correction device in a circuit that reads a signal from a signal input line, resets the signal input line to a constant potential using a reset switch, and reads the signal again. For example, when reading the charge accumulated in each photodiode from each signal input line connected to each of the plurality of photodiodes constituting the image sensor, the variation in the dark output potential that occurs for each photodiode must be considered. It can be used as a driving device for an image sensor that can correct and obtain a constant dark output.

(Prior art) A contact type image sensor, for example, as shown in FIG.
It is composed of a long light receiving element array 60 and a plurality of IC chips 70 that drive the light receiving element array 60. The light receiving element array 60 includes a plurality of individual electrodes 61 . Photoconductive layer 6
2. Composed of a common electrode 63 stacked in sequence, the photoconductive layer 6
2 between the individual electrode 61 and the common electrode 63 forms a photoelectric conversion element. The individual electrode 61 consists of a rectangular pixel portion 61a and an extraction electrode portion 61b drawn out from the pixel portion 61a. Since the width of this contact type image sensor needs to be approximately the same as the width of the document, a large number of pixel portions 61a are arranged in a line. Every 64 or 128 extraction electrode portions 61b drawn out from the pixel portion 61a are connected to one IC chip 70 via a bonding wire 80.

FIG. 7 shows an equivalent circuit corresponding to one IC chip 70 of this contact type image sensor. To explain the operation of one photoelectric conversion element constituting the light receiving element array, light from a light source (not shown) forms an image on the photoelectric conversion element, and light from an oscillator (not shown) forms an image on the photoelectric conversion element. Individual electrode 6 within the accumulation period according to the accumulation period signal obtained based on the clock signal.
It is temporarily stored in capacitor C, which has a wiring capacitance of 1.

On the other hand, control signals (■A, C
Q of shift register SR based on K).

A pulse signal is outputted, and both this pulse signal and the INH signal become "H" level, and the switch element S1 is turned on based on the AND condition of the AND gate. When the switch element S1 is turned on, the photoelectric conversion signal accumulated in the capacitor C is extracted to the output line Tout via the voltage follower type amplifier A. Further, the switch element S2 discharges the residual charge of the capacitor C to reset the charge. After the potential of the output line Tout is changed in accordance with the photoelectric conversion signal, the switch element S3 is turned on to reset the output line Tout in preparation for extraction of the photoelectric conversion signal from the next photoelectric conversion element. Return to potential VR.

Therefore, if the shift register SR starts operating in response to a control signal and the switch elements S1 are sequentially turned on in response to a signal from the shift register SR, the output from the photoelectric conversion elements (64 or 128) connected to one IC chip 70 is Electrical signals can be extracted in time series to the output line Tout. When the extraction of signals from 64 or 128 photoelectric conversion elements is completed, the pulse signal output from the IB terminal of the shift register SR is manually input to the IA terminal of the second IC chip, and input to the second IC chip. INH
The signal, the IA multiplied signal, and the CK multiplied signal are extracted from the photoelectric conversion element in the second IC chip in the same manner as described above. By sequentially performing the above operations on each IC chip, it is possible to extract the electrical signals accumulated in all the photoelectric conversion elements.

In the above operation, when the switch element S2 is turned off after resetting the charges, switching noise is generated and superimposed on the potential at point P.

This potential is +10 to 200 mV, which is much higher than the dark output (-1 to 2 mV) when the photoelectric conversion element is not irradiated with light.
, and the dark output potential of the image sensor output from T out was increased.

Furthermore, since the dark output potential differs from pixel to pixel (±50 to 100 mV), the output voltage varies from pixel to pixel, making it difficult to achieve a high S/N ratio.

Therefore, in order to suppress variations in the dark output voltage when driving the image sensor, the following configuration has been proposed.

That is, a dummy image sensor is arranged separately from the image sensor that reads the image of the original, and the dummy image sensor is prevented from being irradiated with light. Then, as shown in FIG. 8, every time the N line signal (COM) of the document is read by the start pulse signal IA, the dark output signal (DCOM) is read from the dummy image sensor, and the difference between COM and DCOM is read. By obtaining a dynamic output, the signal of one line of the original from which dark output noise has been removed is read.

(Problem to be Solved by the Invention) However, according to the above method, it is necessary to separately provide a dummy image sensor, and the structure and specifications of this dummy image sensor are completely different from those of the image sensor that reads the image of the original. There is a drawback that the sensor manufacturing process is difficult because it has to be made in the same manner.

In addition, in order to extract the signals from the image sensor and the dummy image sensor that read the image of the original with one drive circuit, the dark output signal (DCOM) is required every time the N-line signal (COM) of the original is read. Since reading is repeated, each requires storage time, which doubles the scanning time, which is extremely disadvantageous for high-speed reading. In contrast, the dark output signal (D
COM) and calculate the true signal by subtracting the memorized dark output signal from the signal read on each line, but this is expensive as it requires a circuit to store the signal. It has the disadvantage of becoming

The present invention has been made in view of the above-mentioned circumstances, and is designed to eliminate the switching noise of the reset switch in a circuit that reads a signal from a signal input line, resets the signal input line to a constant potential using a reset switch, and then reads the signal again. It is an object of the present invention to provide a signal correction device that can remove

(Means for Solving the Problems) In order to solve the problems of the conventional example, a signal correction device according to the present invention includes a signal input line and a reset means for supplying a reset voltage to the signal input line via a switching element. and a noise detection means for detecting noise from the signal input line, and a correction means for supplying a voltage corresponding to the noise detected by the noise detection means to the signal input line as a reset correction voltage. It is a feature.

(Function) According to the present invention, after the signal input line is set to a reset voltage by the closing operation of the switching element, noise generated in the signal input line by the opening operation of the switching element is detected by the noise detection means, and the noise is dealt with. By supplying the reset correction voltage to the signal input line, noise generated by the opening operation of the switching element can be removed and the signal input line can be maintained at the reset voltage.

(Example) An example of the present invention will be described with reference to the drawings.

An example in which the signal correction device according to the present invention is added to an image sensor driving device will be described with reference to the block diagram of FIG. 2.

The signal extraction line 2 of the image sensor 1 is connected to the noise detection means 3
and a branch switch 5 that selectively selects the signal output line 4. This noise detection means 3 is connected to a correction means 6 that outputs a voltage corresponding to the detected noise as a reset correction voltage. The reset potential line 7 of the image sensor 1 is connected to a branch switch 8 that selectively selects the reset potential VR and the correction voltage. The noise detection means 3 is formed by a sample and hold circuit, and the correction means 6
is formed by a unity gain inverter circuit.

The drive pulse circuit 9 includes a start pulse (IA) that operates the drive L]I of the image sensor 1, a clock pulse (CK) that operates the sample hold circuit of the noise detection means 3, and a pulse φ91 that operates the branch switch 5.8. This is a drive pulse generation circuit that generates pulses φ7, etc., and is composed of logic ICs such as flip-flops and counters.

The branch switches 5 and 8 are general-purpose CMOS analog switches (for example, Toshiba TC4066BP).

Harris H1-201H3, etc.) are used. The branch switch 5.8 receives a pulse φ3 which is a signal at the same timing.
．． φ, and the signal extraction line 2 is connected to the branch switch 5.
When the signal output line 4 and the signal output line 4 are connected, the reset potential line 7 and the reset potential VR are connected at the branch switch 8.

Further, when the signal extraction line 2 and the noise detection means 3 are connected at the branch switch 5, the reset potential line 7 and the correction means 6 are connected at the branch switch 8.

The sample and hold circuit of the noise detection means 3 selects the capacitance of the hold capacitor so that the acquisition time is short. Further, the unity inverter circuit of the correction means 6 uses an OP amplifier with a fast slew rate and a wide bandwidth.

FIG. 1 shows an equivalent circuit diagram of the above device, focusing on one pixel of the image sensor. In FIG. 1, the same reference numerals as in FIG. 2 indicate the same parts.

One end of a photodiode D that performs photoelectric conversion in the image sensor is connected to a signal input line 1o, and the other end of the photodiode D is connected to a bias power supply voltage VEE that serves as a common electrode for each photodiode that constitutes the image sensor. . The capacitor C1 is the capacitance of a photodiode, and the capacitor C2 is equivalent to the wiring capacitance, and carriers generated in the photodiode D are temporarily stored in this capacitor C2 as charges.

The signal input line 10 is connected to a voltage follower amplifier A, and this voltage follower amplifier A is connected to a branch switch 5.
It is connected to the signal output line 4 via. Further, the X point of the signal input line 10 is connected to the reset potential VR via the branch switch 8. Between the branch switch 5 and the branch switch 8, a sample and hold circuit as the noise detection means 3 and an inverter and a buffer circuit as the correction means 6 are connected. The sample hold circuit is Y
The voltage generated at the point is sampled and held in synchronization with the above-mentioned CK drive pulse. The inverter and buffer circuit supplies the inverted signal of the sampled and held signal to the X point of the signal input line 10 with a unity gain. The switching element S1 is for extracting the charge accumulated in the capacitor C2 via the voltage follower amplifier A, and the switching element S2 is for setting the potential at point X to the reset potential VR.

Next, the operation of this embodiment will be explained with reference to FIGS. 3 and 4.

First, in order to detect the dark output signal, the image sensor 1
to drive. That is, when the first pulse 101 of the start pulse IA is input to the drive circuit of the image sensor 1, the switching element S1 is turned on by the operation of the shift register etc. that constitute the drive circuit, and the voltage follower amplifier is connected to the signal extraction line 2. The charge due to the dark output that exists in the wiring capacitance C2 of the signal input line 10 is extracted through A.

In this state, the branch switch 5 switches the signal extraction line 2
and a signal output line 4 are connected.

Next, switching element S1 is turned off and switching element S2 is turned on. In this state, since the reset potential line 7 and the reset potential VR are connected at the branch switch 8, the residual charges of the capacitor C2 and the capacitor C1 are removed and the signal input line 10 is reset to the reset potential VR. After the reset is completed, the switching element S2 is turned off, but the switching noise at this time is caused by the signal input line 1.
remains at 0.

This operation is performed for each pixel, and dark output signals for all pixels are read.

Next, the branch switch 5 and the branch switch 8 are operated by the pulse φ2°φ from the drive pulse generation circuit 9, and the signal extraction line 2 and the noise detection means 3. The reset potential line 7 and the correction means 6 are connected, and in this state, the image sensor 1
The second pulse 102 of the start pulse IA is input to the drive circuit that drives the start pulse IA. As a result, the switching element S1 is turned on by the operation of the shift register and the like constituting the drive circuit, and the reset noise output due to the switching noise is taken into the sample and hold circuit of the noise detection means 3, and synchronized with the clock pulse (CK). Hold the sample (Figure 3). After synchronizing with pulse φ,
The inverted signal inverted by the inverter of the correction means 6 is subjected to unity gain to obtain a correction voltage, which is supplied to the X point of the signal input line 10.

This operation is performed for each pixel, and reset noise output is removed for all pixels. Therefore, from the next readout, since the pixel is uniquely corrected or set to a set potential, the subsequent output from the signal output line 2 (particularly the dark output) can be kept uniform and low.

After the above operations are completed, the document (not shown) placed on the photoelectric conversion element of the image sensor is irradiated with light from a light source (not shown), and the reflected light is irradiated to each photodiode D, and the document is Carriers are generated in proportion to the density of the image, and are stored as charges in the capacitor C2. Then, when the pulse 101 of the start pulse IA is input to the drive circuit in the same manner as described above, the switching element S1 is turned on by the operation of the shift register etc. that constitute the drive circuit, and the signal output line 4
(Tout), the N lines of charges accumulated in the capacitor C2 are extracted via the voltage follower amplifier A (FIG. 4). This C0M output includes image information of the original, and daily variations in each pixel due to dark output have been corrected.

Next, after the residual charges in the capacitors C2 and C1 are removed and the signal input line 10 is reset to the reset potential VR, the next pulse 102 of the start pulse IA is manually applied to the drive circuit, and the reset noise is output in the same manner as above. is removed to prepare for reading the next line (N+1) of the original.

That is, when a start pulse 101 is manually applied to the drive circuit that drives the image sensor 1, the switching element S1 is turned on in the same way as described above, and the reset noise output due to the switching noise is taken into the sample and hold circuit of the noise detection means 3, and the clock Sample and hold in synchronization with pulse (CK). After synchronizing with the pulse φ, the inverted signal inverted by the inverter of the correction means 6 is given a unity gain to obtain a correction voltage, which is supplied to the X point of the signal input line 10.

Therefore, as shown in FIG. 4, at point Y of the signal extraction line 2, the corrected COM signal and the noise output are extracted alternately. There is no need to provide a separate storage time for this, and the reading time can be shortened. Then, only the corrected COM signal is output to the signal output line 4 (Tout).

In this embodiment, since the dark output can be lowered and variations between pixels can be reduced, by connecting an amplifier to the signal output line 4, the sensitivity can be significantly improved.

In the above embodiment, the sample and hold circuit is constructed from a dedicated IC, but it can also be constructed from a simple circuit as shown in FIG. This circuit includes three analog switches 20,
The pulses φ and φ, which are composed of a capacitor 21 and control the analog switch 20, have a complementary relationship. To explain its operation, pulse φ is on, pulse φ
is held when OFF, and output is obtained from the OUT terminal. The capacitance of the capacitor 21 is appropriately selected depending on the time constant. The analog switch 20 needs to be fast, and preferably has a rise and fall time of 40 ns or less.

According to this circuit, the sample hold time can be easily increased because of the individual analog switches 20, and the unit cost of the circuit components is low, and the circuit can be manufactured at low cost.

According to the above embodiment, from each signal input line connected to each of the plurality of photodiodes constituting the image sensor,
When reading the charges accumulated in each of the photodiodes, it is possible to obtain a constant dark output by correcting variations in the dark output potential that occur for each photodiode.

(Effects of the Invention) As described above, according to the present invention, after reading a signal from the signal input line, after resetting the signal input line to a constant potential by closing the reset switch, when reading the signal again, the switching The noise generated in the signal input line due to the opening operation of the element is removed by supplying a reset correction voltage corresponding to the noise detected by the noise detection means to the signal input line, and the signal input line is set to the reset voltage. Therefore, it is possible to extract an accurate image signal from which noise has been removed.

[Brief explanation of the drawing]

FIG. 1 is an equivalent circuit diagram showing one pixel of an embodiment in which the present invention is applied to an image sensor drive circuit, FIG. 2 is a block diagram of an embodiment in which the present invention is applied to an image sensor drive circuit, and FIG. The figure is a timing chart for obtaining the correction voltage, Figure 4 is a timing chart for driving the image sensor, Figure 5 is a circuit diagram showing another embodiment of the sampling and hold circuit, and Figure 6 is a plan view of the contact type image sensor. An explanatory diagram, FIG. 7, is the IC in the contact type image sensor shown above.
FIG. 8 is an equivalent circuit diagram for one chip and a timing chart showing the driving method of the contact type image sensor. 2... Signal extraction line 3... Noise detection means 4... Signal output line 5... Branch switch 6... Correction means 7. ... Reset potential line 8 ... Branch switch 10 ... Signal input line Sl, S2 ... Switching element 1 Fig. 2 Fig. 4 Fig. 5 Fig. 5 Figure 6

Claims (1)

[Claims] A signal input line, a reset means for supplying a reset voltage to the signal input line via a switching element, a noise detection means for detecting noise from the signal input line, and a response to the noise detected by the noise detection means. and a correction means for supplying a voltage to the signal input line as a reset correction voltage.