JPH10164324A - Image reading method - Google Patents

Abstract

(57) [Problem] To provide an image reading method using an inexpensive image sensor and capable of high-speed reading. A light image signal is obtained by photoelectrically converting incident light information from a document 51 to an image sensor 53 while a light source 57 is turned on, and scanning and reading accumulated charges accumulated in the image sensor 53 by the photoelectric conversion. Light cycle, and the original 51 with the light source 57 turned off.
The image sensor 53 photoelectrically converts the incident light information from the image sensor 53 into the image sensor 53, and the stored charge accumulated in the image sensor 53 by the photoelectric conversion is scanned while being read and output as a dark image signal. The image information of the document is read based on the difference between the bright image signal and the dark image signal. At this time, the light source 57 is turned on only during a blanking period after the end of reading of the stored charges of the image sensor 53 in the dark cycle.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image reading method for reading the image information of a document to be read based on the difference between a bright image signal obtained when a light source illuminating the document to be read is turned on and a dark image signal obtained when the light source is turned off. In particular, the present invention relates to an image reading method capable of improving a document reading speed by devising a lighting method.

[0002]

2. Description of the Related Art Recently, image reading apparatuses have been widely used as reading apparatuses such as facsimile machines, copiers, scanners and handy scanners. A device in which a display device and an image reading device are integrated as described above has also been proposed.

In this type of image reading apparatus, in order to eliminate the influence of external light and input highly accurate image information, the light source for illuminating the document to be read is repeatedly turned on / off for each line. It has been proposed to read a bright image signal when the light source is turned on and a dark image signal when the light source is turned off for the same reading line of a document, and take the difference between the two.

FIG. 7 schematically shows an image reading apparatus attached to the sensor driving means. In FIG. 7, an image reading device 20 attached to a sensor driving unit 23 includes a long image sensor 21A, a self-fax lens (an erecting equal-magnification rod lens array) 21B, and a line light source 22 that can be turned on / off. Consists of

[0005] Then, the reflected light from the input surface 26 on which the original to be read or the like is positioned is scanned by an image sensor 21B having a linear array of light receiving elements, and the scanning control means 25 controls the sensor driving means 23 to produce an image. By detecting the reading device 20 while scanning in the vertical direction, information on the input surface 26 is sequentially read line by line on the input surface 26. At this time, the information is read twice for each line of the input surface 26, when the light source is turned on and when the light source is turned off.

The information read by the image sensor 21B is processed by the reading control means 24. In this case, the difference between the bright image signal when the line light source 22 is turned on and the dark image signal when the line light source 22 is turned off is output from the background light removing unit 24a, and the difference signal is an image based only on the illumination of the light source. Signal. Here, as the image sensor 21B, C
BAS created by CD or BiCMOS process
High-speed reading is possible by using IS (Base-Stored Image Sensor; Journal of the Institute of Television Engineers of Japan, Vol. 47, PP. 1177). These devices convert optical information into electric charges using a photodiode or phototransistor, and then collectively send the electric charges to a transfer stage, so that the image information is optically input to the photodiode or phototransistor of the image sensor during the transfer of the electric charge. However, since the charge transfer time to the transfer stage can be shortened, the accumulation time for each pixel does not differ. Note that the accumulation time refers to a time when a certain time interval (signal charge of a bright signal and a dark signal) is integrated or accumulated.

[0007] However, since these elements are expensive, most monochrome image sensors have a one-dimensional array of photodiodes and phototransistors, and the accumulated charge photoelectrically converted based on image information is transferred to a common signal line by a shift register. An image sensor that sequentially outputs data is used. Such an image sensor is created by a CMOS process or a bipolar process.

An example of scanning a photodiode with a CMOS shift register is disclosed in Japanese Patent Application Laid-Open No. 7-245533, and an example of scanning a bipolar phototransistor with a thyristor shift register is disclosed in US Pat. No. 4,845,56.
Japanese Patent Application Laid-Open No. 1-19 discloses an example disclosed in the specification of Japanese Patent Application Laid-Open No. 7-197, for scanning a bipolar photodiode with a CMOS shift register.
No. 8183.

Since these latter sensor chips can be easily manufactured at a low cost, they are advantageous for a contact type image sensor requiring a long sensor, and the above-mentioned portable type or an image obtained by integrating a display device and an input device. By using it for a reading device, a reduction in price can be expected. FIG. 8 is a timing chart when an image is read using the above-mentioned inexpensive image sensor. In FIG. 8, (a) shows a start signal S for instructing to start reading one line of image information.
t, (b) shows an image signal from the first bit to the n-th bit output from the image sensor in response to the start signal St, and (c) shows the on / off state of the light source.

In the case of FIG. 8, the first start signal St
The light source is turned on for approximately two cycles until all the image signals from the first bit to the n-th bit have been output in response to the generation of the second start signal St from the occurrence of
During the next two cycles, the light source is turned off, and the image signals from the first bit to the n-th bit output from the image sensor in response to the second start signal St are used as the bright image signals, and the fourth start signal is output. The image signals from the first bit to the n-th bit output from the image sensor in response to the signal St are defined as dark image signals. In this manner, by turning on the light source for approximately two cycles, the light receiving element corresponding to the pixel of the first bit and the light receiving element corresponding to the pixel of the nth bit are irradiated with light for the same time, The difference in irradiation time between pixels can be eliminated.
In FIG. 8B, T ₁ indicates the irradiation time (accumulation time) to the light receiving element corresponding to the pixel of the first bit.
_n indicates the irradiation time (accumulation time) to the light receiving element corresponding to the n-th bit pixel.

[0011]

However, CCDs
Unlike the method of collectively reading out the charges that have been photoelectrically converted and stored in a photodiode, such as BAT and BASIS, the method of directly reading out the charges that have been photoelectrically converted and stored directly starts charge storage as soon as reading is completed. When the light source is switched on / off for each readout period, signals having different accumulation times are accumulated in each pixel.

For this reason, as shown in FIG. 8, it is necessary to scan the same line twice before switching the light source from on to off or from off to on, and to discard the previous image signal. In FIG. 8, in response to the start signal St of (a), the bright image signal of (b) is output with the first bit to the n-th bit for one line. At this time, the accumulation time of each bit, the first bit period T _1, the n-th bit of the final bit is a period T _n. For this reason, the illumination light source
It is necessary to illuminate between two lines. Then, it is necessary to discard the image signal of the line corresponding to the first start St in the illumination period shown in (c) and extract only the image signal of the line corresponding to the second start signal St. Also, when the light is turned off, the image signal is read out for a certain period from the third start signal St and for a certain period from the fourth start signal St, respectively.
, It is necessary to discard the image signal of the line corresponding to the above (4) and take out the image signal of the line of the fourth start signal St as the light-off signal.

As described above, in the conventional method, it is necessary to read the image signal four times for the same line.
There is a disadvantage that the reading speed is significantly reduced. In view of the above problems, the present invention provides an image reading method that can use an inexpensive image sensor and that can perform high-speed reading even in a portable image input device that needs to remove disturbance light (background light). The purpose is to:

[0014]

According to an image reading method of the present invention, information on light incident on an image sensor from a document to be read is photoelectrically converted by the image sensor while a light source for illuminating the document to be read is turned on. A light cycle in which the stored charge stored in the image sensor is read while scanning and output as a bright image signal, and light information incident on the image sensor from the original to be read is photoelectrically sensed by the image sensor with the light source turned off. The dark cycle of converting and reading the stored charge stored in the image sensor by photoelectric conversion while scanning and outputting it as a dark image signal is repeatedly executed, and read based on the difference between the bright image signal and the dark image signal. This is a method of reading the image information of the target document. To light only in the blanking period after the end of the take.

The above-mentioned image sensors include, for example, the following two types. First, as an image sensor, a plurality of photodiodes that are arranged and arranged to photoelectrically convert incident light information, and a plurality of photodiodes respectively provided in the plurality of photodiodes and each of which accumulates charges accumulated by a signal photoelectrically converted by the plurality of photodiodes, are used as impedances. A plurality of MOS field effect transistors for current amplification to be converted; a plurality of MOS field effect transistors for access provided respectively in the plurality of photodiodes for outputting a holding voltage of the plurality of MOS field effect transistors for current amplification; A plurality of reset MOS field-effect transistors provided on the respective diodes for resetting a plurality of photodiodes. The individual electrodes of the photodiodes are connected to a reset power source at fixed intervals of storage time via the reset MOS field-effect transistors. Contact, sequentially according to the scan pulse through the MOS field-effect transistor for accessing, using those configured to output a bright-field image signal and dark state image signal from the common signal line.

Second, as an image sensor, a plurality of phototransistors which are arranged and arranged to photoelectrically convert incident light information, and a plurality of phototransistors which are respectively provided in the plurality of phototransistors and which accumulate electric charges by signals photoelectrically converted by the plurality of phototransistors. A plurality of access transistors that output currents, and a plurality of reset transistors that are respectively provided in the plurality of phototransistors and reset the plurality of phototransistors. Individual electrodes of the phototransistors are fixed through the reset transistors. A configuration is used in which a reset power supply is connected and disconnected at intervals of an accumulation time, and a bright image signal and a dark image signal are sequentially output from a common signal line according to a scanning pulse via an access transistor.

According to the present invention, the light source is turned on only during the blanking period after the end of reading the stored charges of the image sensor in the dark cycle. The lighting time can be the same, and the same line is scanned twice, and the bright image signal and the dark image signal can be obtained only by scanning once each time without having to discard the previous image signal. When the video signal can be obtained. As a result, an inexpensive image sensor can be used, and high-speed reading can be performed even in a portable image input device that needs to remove disturbance light (background light).

[0018]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a color image reading method according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a schematic diagram showing a configuration of a color image reading device used in a color image reading method according to an embodiment of the present invention. In FIG. 1, reference numeral 51 denotes a document to be read, and 52, a lens. In this example, a rod lens array of an erect equal-magnification is shown. However, a reduction lens or a reduction lens lens array may be used. Reference numeral 53 denotes a linear image sensor, which has a structure in which a large number of photoelectric conversion elements 55 are arranged and arranged on a substrate 54, and n light receiving element rows 56 are arranged on the light receiving surface of the photoelectric conversion elements 55. Have been. As a document illuminating means, a line light source 57 composed of an LED array or the like is provided.
Means (not shown) for turning on / off the line light source 57 are provided in synchronization with means for scanning the relative position between the light source element 1 and the light receiving element array 56. In FIG. 1, the arrow A indicates the relative movement direction of the document 51 to be read. Note that the image reading apparatus of FIG. 1 schematically illustrates the image reading apparatus shown in FIG.

In the color image reading method using the color image reading apparatus having the above-described configuration, as shown in FIG.
When the light is turned on, the incident light information from the document 51 to be read into the image sensor 53 is photoelectrically converted by the image sensor 53, and the stored charges stored in the image sensor 53 by the photoelectric conversion are read while being scanned, and a bright image signal is obtained. And a light cycle for outputting the image sensor 53 from the document 51 to be read with the line light source 57 turned off.
The image sensor 53 photoelectrically converts the incident light information into the image sensor 53, reads the stored charge accumulated in the image sensor 53 by the photoelectric conversion while scanning, and outputs as a dark image signal. The image information of the document to be read is read based on the difference between the signal and the dark image signal. At this time, the line light source 57 is turned on only during the blanking period after the end of reading the stored charges of the image sensor 53 in the dark cycle.

The above-mentioned blanking period is usually provided to secure a signal processing time after reading,
This is provided in order to reduce the shift of the position of the start bit and the end bit due to the line scanning, that is, to reduce the oblique reading by making the reading time earlier than the accumulation time. With this configuration, the line light source 57 is turned on only during the blanking period after the end of reading the stored charges of the image sensor 53 in the dark cycle.
The light source lighting time within the accumulation time can be made the same, and the same line can be scanned twice, without scanning the same line twice and discarding the previous image signal. A time video signal and a dark video signal can be obtained. As a result, an inexpensive image sensor can be used, and high-speed reading can be performed even in a portable image input device that needs to remove disturbance light (background light).

FIG. 2 shows an embodiment of the present invention in which photodiodes and phototransistors are arranged one-dimensionally, and accumulated charges photoelectrically converted based on image information are sequentially output to a common signal line by a shift register. FIG. 4 is a timing chart when an image is read using an inexpensive image sensor. In FIG. 2, (a) shows a start signal St for instructing the start of reading of one line of image information, a period after the first and third start signals St is a dark cycle, and a second and fourth period are shown. The period after the start signal St is a light cycle. (B)
Indicates image signals from the first bit to the n-th bit output from the image sensor 53 in response to each start signal St, and the level of the dark image signal in the dark cycle is reduced by the absence of illumination, The level of the bright image signal in the bright cycle is higher by the amount of illumination, and if the difference between the two is taken, only the illumination (no background light is included)
Is an image signal. (C) shows the on / off state of the light source, and the image sensor 53 in the dark cycle.
Illuminated only during the blanking period after the end of reading the stored charge. The light source is turned on / off by receiving a start signal, generating a lighting timing pulse for defining a lighting timing having a predetermined time width, and a trailing edge (falling edge) of the lighting timing pulse.
To generate a lighting pulse, and the light source is turned on during the generation period of the lighting pulse.

Here, a specific example of the photoelectric conversion element 55 in FIG. 1 will be described below. The photoelectric conversion element 55 includes a sensor IC using the bipolar phototransistor and the thyristor shift register shown in FIG.
A sensor IC in which a phototransistor or a shift register is formed by the OS, or an IC element using a MOS amplifier and a MOS shift register for the photodiode in FIG. 5 is used.

First, the photoelectric conversion device shown in FIG.
This will be described with reference to FIG. In FIG. 3, 1a to 11
The base and collector of the NPN transistor up to a are 1b
To the collectors and bases of the PNP transistors 1 to 11b, respectively.
a and the PNP transistors 1b to 1b form eleven thyristors.

The NPN transistors 1c to 10c are relay transistors that transmit ON / OFF of 11 thyristors each composed of NPN transistors 1a to 11a and PNP transistors 1b to 1b to the next thyristor. Function as a reset transistor. The PNP transistors 2e to 9e serve as current switches for transmitting the ON / OFF of the thyristor to the phototransistors 2f to 9f.

The start signal ST shown in FIG. 4B is input to the NPN transistor 12, and the clock signals φ1 and φ2 shown in FIGS. 4C and 4D are supplied to the NPN transistors 1c to 10c and the emitters of the NPN transistors 1a to 11a. , The ON / OFF of the thyristor is transmitted to the next thyristor. Phototransistor 2f ~
The 9f emitter outputs even number and odd number directly from the output terminals 33 and 34 through two common output lines,
Alternatively, the signal is output from output terminals 35 and 36 through a preamplifier. The output terminal 32 of the last thyristor is connected to the signal input terminal SI of the next-stage photoelectric conversion element (not shown), and ON / OFF of the thyristor is passed to the next-stage photoelectric conversion element.

The diodes 2g to 9g are for charging the phototransistors 2f to 9f at a constant voltage at a high speed. In the timing diagram of FIG. 4, the external clock E shown in FIG.
For clock signals φ1 and φ2 generated from X-CK, signals Q1 to Q6 from (e) to (j) indicate ON / OFF of the thyristors from the first stage to the sixth stage. , The high level is off and the low level is on. The signal Sig (1) in FIG. 4K indicates the output signal of the odd-numbered phototransistor, and the signal in FIG.
g (2) indicates an output signal of the even-numbered phototransistor.

Using a photoelectric conversion element as shown in FIG.
A sensor IC element having a length of 16 mm was constructed in which 128 phototransistors were arranged in a line at a pitch of 5 μm. The sensor IC elements were arranged and connected on a substrate with 14 chips seamlessly to obtain an image sensor. This image sensor had an S / N ratio of 32 dB or more with a sensitivity of 25 V / lx · s in sensor surface illuminance after integrating and amplifying the output current with white illumination. When scanning with an external clock EX-CK of 0.5 MHz and an accumulation time of 10 msec using this image sensor, it takes 40 msec to read one line in the conventional reading method.

On the other hand, in the image reading method according to the embodiment of the present invention shown in FIGS. 1 and 2, a reading period of 3.6 msec and a blanking period 6.4 after reading are completed.
One line can be read in a total of 20 msec in an LED lighting period of 5 msec out of msec. However, at this time, since the lighting period is half that of the conventional example, the light source L
The pulse peak current of the ED required twice the current of the conventional example. In addition, the read image was able to reproduce good image information without being affected by disturbance light.

Although the photodiodes used here are one row, another row of light-shielded phototransistors is prepared.
By simultaneously accessing and taking the difference signal between the two output signals, dark noise can be reduced and the load on the preamplifier and the external amplifier can be reduced. That is, in only one column, spike noise at the time of access is large, so that S / N
Can't get big. Therefore, a spike noise can be removed by providing a dummy phototransistor, extracting only a dark signal, and performing an operation of a bright signal-dark signal. As a result, it is not necessary to remove noise in the preamplifier or the external amplifier, and the burden on them can be reduced.

The same is true not only when a shift register using a thyristor is used but also when a decoder circuit disclosed in US Pat. No. 4,567,529 using a bipolar transistor is used. In the case of this sensor, the dynamic range is relatively large, and it is 1000 to 30 in a general office.
The signal does not saturate even under illumination light of about 00 lx,
It was possible to remove disturbance light.

Next, FIG. 6 is a timing chart of each pulse and output of the example of the photoelectric conversion element shown in FIG. 5 in which current is amplified by a MOS transistor and image information is read out while sequentially accessing a shift register using a photodiode. It will be described using FIG. In FIG. 5, the photoelectric conversion element includes a photodiode 40, an amplifying MOS field effect transistor 41 and an access MOS field effect transistor 42 which operate by receiving the voltage of an individual electrode (anode) of the photodiode 40 at a gate. A plurality of pixels 44 each including a resetting MOS field effect transistor 43 for returning the individual electrodes 40 to the initial potential;
, A scanning circuit 45 for generating scanning signals Y1, Y2, Y3,.
A pixel signal output line 46 having the source of the field effect transistor 42 connected in common and a reset power source 47 having the source of the reset MOS field effect transistor 43 corresponding to the photodiode 40 of each pixel connected in common.
And a reset line 48 for connecting to
Created by OS process.

The scanning signals Y1, Y2, Y3,... And the NAND signal of the reset pulse RS of the photodiode 40 are input to the gate of the reset MOS field effect transistor of the photodiode 40. The built-in amplifier is formed as required, but the amplifier 49
A grounded-gate amplifier including an S field effect transistor 1, a load MOS field effect transistor 2, a p-channel MOS field effect transistor as a reset switch 3, a capacitor 4, and a source resistor 5 may be formed in the same chip.

Here, the relationship between the operation of the above-described photoelectric conversion element and the connection of the individual electrodes of the photodiode to the reset power supply at a fixed storage time interval via the reset MOS field effect transistor will be described. That is, it is necessary to read out a signal after a certain storage time, read out a charge (read out a voltage corresponding to the charge stored in the photodiode as a current), and then reset the stored charge to the initial voltage. For that purpose, it is connected to the reset power supply as described above.

FIG. 6 shows the operation timing of the photoelectric conversion element shown in FIG. 5, in which the clock pulse CK of (a) and the start pulse ST of (b) supplied from the outside are output from the scanning circuit 45 ( c), (d) and (e) scanning signals Y1, Y2, Y3, and (f) photodiode 4
A reset pulse RS of 0, a pixel signal current Iin of (g) output from the pixel signal output line 46 and input to the amplifier unit 49, and a pixel signal voltage Vout of (h) output from the amplifier unit 40, respectively. ing.

The first pixel is accessed by the scanning signal Y1, and the photodiode 40 of the first pixel is reset by the scanning signal Y1 and the NAND signal of the reset pulse. Therefore, a pixel signal is output between the rising of the scanning signal Y1 and the rising of the reset pulse RS, and the dark signal, that is, the reference signal Ire, is output from the falling of the reset pulse RS to the falling of the scanning signal Y1.
f and Vref are output. In this way, the pixel signal and the reference signal are continuously output according to the scanning signal.

For example, photodiodes are arranged one-dimensionally at intervals of 62.5 μm, and a sensor chip having 256 pixels and a length of 16 mm is formed by a CMOS process. The electric charge corresponding to the image information accumulated during the accumulation time was reset to the reference voltage after reading, so that there was no afterimage and good photoelectric conversion characteristics were exhibited. The sensor chip was connected to 14 chips to form an A4 size image sensor.
With a clock frequency of 2 MHz and an accumulation time of 2.5 msec, an S / N ratio of 35 dB or more could be obtained with a sensitivity of 24 V / lx · s.

When a scan as shown in FIG. 8 of a conventional example is performed using this image sensor, GaP having a wavelength of 565 nm is used.
When switching LED light source, 10msec for one line scanning
c reading time is required. On the other hand, in the reading method of FIG. 2 of the present invention, the reading period of 1.8 msec and the blanking period of 0.7 msec (middle, 0.
5 msec is the LED lighting period), 2.5 msec,
One line is read twice, a signal when the LED is not lit is taken out, and the difference between the two is taken, so that 2.5 msec × 2
Thus, one line can be read in a total of 5 msec. In addition, the read image was of course good without being affected by external light.

Here, when the general illumination light is 3000 lx or more, the output voltage may be saturated in the sensor of FIG. For this reason, by adhering a band-pass type optical filter that transmits the same wavelength (565 nm) to the image sensor 53 side of the lens 52 in FIG. 1, this can be largely improved, and the filter can be saturated even under illumination of 5000 lx. It could be confirmed that it operates normally without any problem.

In the above embodiment, the reading of a color original has been described. However, the case of reading a black and white original can be similarly considered.

[0040]

According to the image reading method of the present invention, when the difference between the bright image signal when the document is illuminated by the light source and the dark image signal when the light source is turned off is read as image information of the document,
Since the light source is turned on only during the blanking period after the end of reading the accumulated charges of the image sensor in the dark cycle, it is possible to read a good image at high speed using an inexpensive image sensor.

[Brief description of the drawings]

FIG. 1 is a schematic diagram of an image reading apparatus that performs an image reading method according to an embodiment of the present invention.

FIG. 2 is a timing chart showing driving of the image reading method according to the embodiment of the present invention.

FIG. 3 is a circuit diagram showing a configuration of a first example of a photoelectric conversion element in an image sensor used in the image reading method according to the embodiment of the present invention.

FIG. 4 is a timing chart showing an operation of the photoelectric conversion element of FIG. 3;

FIG. 5 is a circuit diagram showing a configuration of a second example of the photoelectric conversion element in the image sensor used in the image reading method according to the embodiment of the present invention.

FIG. 6 is a timing chart showing the operation of the photoelectric conversion element of FIG.

FIG. 7 is a perspective view illustrating a configuration of an example of an image reading apparatus according to a conventional example and an embodiment.

FIG. 8 is a timing chart showing driving of an image reading method in a conventional example.

[Explanation of symbols]

 Reference Signs List 1 MOS field effect transistor for drive 2 MOS field effect transistor for load 3 Reset switch 4 Capacitor 5 Source resistance 1a-11a NPN transistor 1b-11b PNP transistor 1c-10c NPN transistor 2e-9e PNP transistor 2f-9f Photo transistor 2g-9g Diodes 2d to 10d Resistance 12 NPN transistor 20 Image reading device 21A Selfoc lens 21B Optical sensor 22 Line light source 23 Sensor driving means 24 Reading control means 24a Background light removing means 25 Scan control means 26 Incident plane 32 Final output of thyristor 33 Common output Line 34 common output line 35 amplifier output 36 amplifier output 40 photodiode 42 amplifying MOS field effect transistor 43 reset Preparative field effect transistor 44 a plurality of pixels 45 scanning circuit 46 pixel signal output line 47 resets the power 48 reset line 49 amplifier unit 51 reads the target document 52 lens 53 image sensor 54 substrate 55 photoelectric conversion element 56 receiving element array 57 line light source

Claims (3)

[Claims] 1. An image sensor for photoelectrically converting incident light information from an original to be read into an image sensor with a light source illuminating the original to be read turned on, and accumulating the information in the image sensor by photoelectric conversion. A light cycle in which the electric charge is read while scanning and output as a light image signal; and light information incident on the image sensor from the document to be read is photoelectrically converted by the image sensor in a state where the light source is turned off. A dark cycle of reading and outputting as a dark image signal while scanning the accumulated charge accumulated in the image sensor while scanning, and based on a difference between the bright image signal and the dark image signal, An image reading method for reading image information of a document, comprising: Image reading method characterized by only turning on the blanking period after completion of the reading of the stored charge in Mejisensa. 2. An image sensor, comprising: a plurality of photodiodes arranged and arranged to photoelectrically convert incident light information; and a plurality of photodiodes respectively provided in the plurality of photodiodes and stored by signals photoelectrically converted by the plurality of photodiodes. A plurality of current-amplifying MOS field-effect transistors each of which converts an electric charge into an impedance; and a plurality of access MOS field-effect transistors respectively provided in the plurality of photodiodes for outputting a holding voltage of the plurality of current-amplifying MOS field-effect transistors. A plurality of reset MOS field-effect transistors provided on the plurality of photodiodes and resetting the plurality of photodiodes, respectively, and the individual electrodes of the photodiodes are connected via the reset MOS field-effect transistors. Are connected to a reset power supply at regular intervals of a storage time, and sequentially output a bright image signal and a dark image signal from a common signal line according to a scanning pulse via the access MOS field effect transistor. The image reading method according to claim 1. 3. An image sensor, comprising: a plurality of phototransistors arranged in a line and photoelectrically converting incident light information; and a plurality of phototransistors provided in the plurality of phototransistors, respectively, and a charge accumulated by a signal photoelectrically converted by the plurality of phototransistors. And a plurality of reset transistors provided in the plurality of phototransistors to reset the plurality of phototransistors, respectively. Connect and disconnect the electrode to the reset power supply at regular intervals of storage time,
2. The image reading method according to claim 1, wherein said image reading method is configured to sequentially output a bright image signal and a dark image signal from a common signal line according to a scanning pulse via said access transistor.