JPS5986362A - Close-contact sensor - Google Patents

Abstract

PURPOSE:To record a color original with high speed and high resolution by reading out a signal plural CCDs by using one end of the CCDs as a reference. CONSTITUTION:A CCD chip 21a consists of a dummy area 41, 6-bit black reference light shield area, 1,908-bit of effective reading area 42, and a dummy area 44. The light shield area is bits shielding the light of a photodetector and an output from this area has a reference voltage being a voltage attended to a dark current. The CCD21a is driven by using a 2-phase clock, a scanning synchronizing signal, reset signal and an output signal, and its output signal is A/D- converted to a form suitable for picture information by an analog processing circuit including the correcting section and an integrating part. This output signal is formed into a suitable signal for the recording of a recorder by a digital processing system.

Description

[Detailed Description of the Invention] [Technical Field of the Invention] This invention is image 1#! The present invention relates to a color contact sensor for reading a color document, and in particular to a color contact sensor for reading a color document.

[Technical background of the invention and its problems]

For information-rich manuscript reading devices, divine resolution,
There is a demand for miniaturization, colorization, etc.

A solid-state scanning method is a technology that meets these requirements.

For example, auto diode array and MOS for optical conversion
There are two types of devices: one in which the device is completely combined with a multi-function device, and one in which the device itself uses a semiconductor functional element that has both a pixel decomposition function and an optical information storage function.

Since such solid-state sensor devices are highly integrated,
The size of the image was much smaller than the original, and a reduction optical system had to be used. In particular, in the lower part of the second left column of Japanese Patent Application Laid-open No. 54-79511, it is stated that: [Since commercially available photodiode arrays and CCD sensors are integrated, a reduced optical system must be used in order to read the original. This inevitably increases the size of the optical system, which is both extremely inconvenient and unfavorable when it comes to making a small document reading device. So, it's not as if it could be made smaller just by adopting a 1:1 optical system, specifically a light-receiving element array that is as large as the width of a document, with a 1:1 optical system. If you try to make a reading circuit using a single semiconductor switch t1 sold by the city, it will be quite large and expensive, judging from the size and number of parts. ” is written. This description fully represents the state of affairs in this field, and it is with this recognition that dense layer sensors are being developed.

On the other hand, single-sided scanning of color originals has been realized using a dichroic mirror (dichroic m1rror). As is well-known, dichroic 1' is made up of layers with different refractive indexes, and uses the property of increasing the reflectance in the desired wavelength range to separate the three colors of 几, G, and B. One image reduced by a lens system is read simultaneously by different photoelectric conversion devices. It ended up being roughly three times the size of a single-column loading device for black and white originals. In addition, the resolution in this case is limited by the size of the sensor and the number of elements, since the number of lenses is reduced by the lens thread.

On the other hand, as a color dense layer sensor, there is a sensor published by Tajiri et al. that uses the entire Cd-8e sensor. (
(1981 National Conference of the Gatai Electronics Society ``Study of Color Photoelectric Conversion Circuits for Close-Contact Sensors'') This method is noteworthy as a color converter for close-contact sensors.
There is a limit to the light response speed of -8e, and the reading time for one image is limited to 20 m see/fine per single color.

Also, with this method, red (
1t) and an RlG LED (
Ligbt Emitting Diode), Blue (B
) is used as a light source, and the light source is switched for each scan, R, G,
An output of H is obtained. Therefore, a color-separated image of three colors? To obtain this, it takes three times the scanning time of one monochromatic image,
A reading time of approximately 60 msec/line was required.

Furthermore, the frequency per screen was limited to 8 lines/mm due to manufacturing constraints of a sensor using CD-8EF.

[Purpose of the invention]

The object of the present invention is to eliminate the above-mentioned drawbacks and to provide a color contact sensor V which is fast and has high resolution.

[Summary of the Invention] This invention realizes colorization by providing a plurality of photoelectric conversion element arrays each having minute photoelectric conversion elements, and by providing three color filters on the elements themselves. ing. On the other hand, a SELFOC lens is provided for each element row so that all the element rows cover the same line tm without reducing the image source, thereby achieving high resolution in one image.

[Embodiments of the invention]

Next, an embodiment of the present invention will be described with reference to all the drawings. The color contact sensor in this embodiment is capable of reading an A4 size color document.

This color contact sensor, as shown in Figure 5g1,
A sensor package (11) on which a plurality of CCD chips (not shown in FIG. 1) are mounted, two focusing rod lens arrays (12) arranged on this sensor package (11), and this focusing rod lens array (12). Sex rod lens array (12)
It has an integral structure with a linear light source (13) provided near the side surface. However, in Figure 1, a linear light source (13
) is shown, but in reality, there is one more converging rod lens array (121k) sandwiched between them.

Briefly, the function of this color adhesion sensor is that light from a light source (13) is irradiated onto a document (not shown), and the reflected light is transmitted to a sensor package (finishing CCD) by a focusing rod lens array (12). The important point here is that the focusing rod lens array (12) focuses the reflected light from the document onto the multiple CCI chips in a one-to-one relationship without reducing the reflected light in any way. It is to let

The optical image thus formed on the CCD chip is C
The charge is converted into VC due to the light weight conversion ability of OD.

This charge is forward-transferred by the charge transfer capability of the COD.
-(becomes an image signal.Such signal conversion is performed as the color contact sensor moves relative to the original +f4VC.

Now, let's explain each part in detail. As shown in FIG. 2, the sensor cage (11) includes a ceramic substrate (22) on which eight CCD chips (21a) to (21h) arranged in a staggered manner are provided, and this ceramic substrate ( 22) It consists of a container (23) that covers the.

The CCD chips (21a) to (21h) are TCD102C-1'i manufactured by Tokyo Shibaura Electric Co., Ltd.

The light receiving portions of the CCD chips (21a) to (21h) are composed of Pn photodiodes. The size of one bit of this light receiving section is 14μ×14μ. One CCD chip (21a) to (21h) consists of a light receiving area of 2048 bits (approximately 28.7 mm1).

In this invention, such CCD chips (21a) to (21h) are arranged in two rows in a staggered manner as shown in FIG. At this time, the two rows of CCD chips (21a) to (21h) are arranged substantially parallel to each other with a distance mff between the centers of the light receiving sections, as shown in FIG. Further, these CCD chips (21a) to (21h) are arranged so as to overlap each other along the arrangement direction, allowing dk.

However, the overlap d is not allowed unlimitedly, but CC
From the tip of the D chip (21a) to the CCD chip (21h)
) The length L of the rear edge blade is set to correspond to the width of the document to be read. In this example, the width of A4 paper is 2.
Aim for 10 mm Vc4'.

Furthermore, all of the light receiving sections of the CCD chips (21a) to (21h) are not used for reading one image. As aforementioned,
One CCD chip (21a) to (21h) has 20
It consists of a 48-bit light receiving section. These light receiving sections consist of, from the left end, a dummy area (41), a 1908-bit reading effective area (42), and an overlapping area (43) with the next stage CCD chip. The effective reading area (42) is a little over 210 nm wide in A4 size.

A focusing rod lens array (12
), one line on the document can be reduced to 1 x 1 without being reduced. CCD chips (21a) to (21h)
I sat on top of Yui 11. That is, the image information on the same line is the CCD chip (21a) shown in FIG.
21c), (21e), (21g) columns and CCD
Chips (21b), (21d), (21f), (
21h).

To play all images 1a and 1a on the same line of the manuscript, use all C
The full output from the CD chips (21a) to (21h) is required, but as mentioned above, they are arranged in a cylindrical shape to allow heavy FX, so the CCD chips (21a) to (21h)
) information is also important. This is reflected in the existence of the dummy area (41) and the overlap permissible area (43). However, the dummy area (41), ! Take effective area (
42).

The overlap permissible area (43) is not marked on the light-receiving parts of the CCD chips (21a) to (21h), but merely refers to differences in signal processing, which will be described later.

Glass filters are installed in the light receiving parts of such CCD chips (21a) to (21h). Mechanical precision is important in installing glass filters.

In this embodiment, from the viewpoint of positional accuracy during mounting, the overlap region (43) is allowed to have a width of up to 1.3 mm.

The color filter can be attached by directly printing the filter onto the light receiving section. According to this method, it is possible to print the filter with very high precision only on the light receiving section. In this method, the burning technique is the key.

Next, the filter array will be explained. In this example,
The filter array was white (W), yellow (Y), and cyan (c). One gold filter was attached to a 2-bit light receiving part, resulting in 3 colors and 6 bits.

As will be described later, 6 bits were used as one pixel during reading.

Next, the spectral characteristics of the filter will be explained. The spectral characteristics referred to here are the relationship between wavelength and transmittance. As shown by the curve (51) in FIG. 5, the W filter has a transmittance of approximately 100 cm over all wavelengths. The transmittance of the C filter shows a peak around 500 nm, as shown by curve (52), and increases again around 700 nm. The transmittance of the Y filter increases rapidly from around 500 nm, as shown by curve (53). An important point in the spectral characteristics of these filters is that the transmittance does not become zero even for wavelengths of about 700 nm, which are outside the human visual sensitivity range.

Although the filter here has such characteristics, the filter and the CCD chips (21a) to (21h) perform a function similar to that of the human eye. CCD chip (
As shown in FIG. 6, the spectral characteristics of the light receiving sections 21a) to (21h) have a +mouth-to-sensitivity tube that reaches a maximum at a wavelength of about 700 nm and is limited up to 11,000 nm or more.

In the end, the CC to which the color filters in this example were attached
The light receiving section of D chips (21a) to (2H+) is 700
A response also exists for wavelengths of nm or more.

On the other hand, as is well known, the visual magnification of the human eye is zero for wavelengths of 700 nm or more. Therefore, simply the CCD chips (21a) to (zth) and W,
When used in combination with Y and C filters, it is not the same as the human eye. In order to solve this problem, in this embodiment, the light source is specified as described later.

A CCD chip (21
a) to (21hl) are arranged on the ceramic substrate (22) and then covered with a container (23).
3) is a glaze without a bottom surface and has a CCD chip (21a).
to (21h) Protected from dust, etc. However, C
CD chips (21a) to (21h) are irradiated with light 1
Therefore, the CCD chip (
Windows are provided at locations corresponding to 21a) to (21h).

The surface where this window is provided is covered with glass. In this way, COD (21a) to (21h)
Only light is supplied to the device, and no dust or the like will adhere to it.

Next, the convergent rod lens array (12) will be explained.

The focusing rod lens array (12) is widely sold under the trade name Selfoc Lens (manufactured by Nippon Sheet Glass) and is easily available. The convergent rod lens array (12) in this embodiment is on the light incident side.

The distance between the focal points on the exit side, that is, the conjugate length Tc is 54rr
I used one with an equivalent F number of approximately 3.15.

Furthermore, as shown in FIG. 7, the angle formed by the optical axis of the convergent rod lens array (12) is approximately 15°, and the document surface (71) is located at the point where the optical axes coincide. do. There are CCD chips (21a) to (21h) f between this document surface (71) and the reflection direction, and a CCD chip row (72) that divides the two into two.
exists.

With these settings, the document surface (71) and the top of the CCD chip row (72) are in a relationship. In other words, the document surface (
71) The above image is erected in a one-to-one manner (imaged as an elephant on the CCD chip row (72). However, the manuscript surface (71) is uneven when used for normal one-story printing. However, the focusing rod lens array (12) has a deep depth of focus and is resistant to changes in the position of the original.
The imaging relationship with the CD chip (72) is unlikely to deteriorate.

Furthermore, the tilt error of the convergent rod lens array (12), the deviation in parallelism in the longitudinal direction of the convergent rod lens array (12), etc. have important effects on the imaging relationship.

In other words, accurately setting the position of the focusing rod lens array (12) has a great impact on reading accuracy.
give. One adjustment means is shown. This adjustment means, as shown in FIGS. 8 and 9, consists of a plate-shaped angle adjustment plate (81) provided along which the focusing rod lens (12) is aligned; Angle adjustment plate (81
) and the focusing rod lens array (12), and a screw (82) for fixing the focusing rod lens array (12)' to the angle adjustment plate (81).

It will be explained in further detail. This vJ4 adjusting means is provided in a pair so as to be narrower than the converging rod lens array (12). This adjustment means has the following structure so that the inclination of the angle adjustment plate (81) can be finely adjusted.

On the substrate (83), clamping parts (84) for holding the angle adjustment plate (81) are fixedly provided at both ends B, b.

However, the angle adjustment plate (81) is rotatable around the point held by the clamping part (84). However, this holding point is located approximately at the center of both end surfaces of the angle adjustment plate (81).

A regulating protrusion (85) is provided parallel to the holding portion (84) and near the center of the substrate (83).

On the other hand, near both ends of the side surface of the angle adjustment plate (81), there are adjustment members (8
6) is fixed. Among these adjustment members (86), +14 adjustment members (8) that are not fixed to the angle adjustment plate (81)
A horizontal protrusion (87) is provided at the bottom of the end of 6). The surface of the adjustment member (86) opposite to the surface on which the horizontal projection (87) is provided is in contact with the side surface of the regulating projection (85).

The adjustment member (86) is fixed to the angle adjustment plate (81) and moves together with the angle adjustment plate (81). Therefore, when the angle adjusting plate (81) is rotated about the holding point of the holding part (84), the adjusting member (86) also moves. In particular, the horizontal protrusion (87) performs a downward movement m on the road. Conversely, when the horizontal protrusion (87) is moved up and down, the angle rA adjustment plate (81) changes its inclination with respect to the vertical direction. An angle adjustment screw (88) is used to move this horizontal protrusion completely up and down. This angle adjustment screw (88) control member (89) 't-watasu. This regulating member (89) is fixed to the holding portion (84) and the regulating projection (85).

This regulating member (89) extends above the horizontal protrusion (87). This part is threaded and provided with an angle adjustment screw (88). The lower end of this angle adjustment screw (88) is in contact with the horizontal protrusion (87).

Therefore, when the angle 14 is advanced toward the tl-f side, the horizontal protrusion (87) is pushed downward, and as seen from the direction indicated by the arrow (90) in FIG. The angle adjustment plate (81) rotates. Angle adjustment screw (88)
' When the horizontal protrusion (87) is advanced upward, the spring (8
9a) always applies an upward force, causing it to move upward.

Then, the angle adjustment plate (81) rotates in one direction in the opposite direction.

Once the angle of the angle adjustment plate (81) has been determined using the angle adjustment screw (88) e, turn the adjustment screw (82)
The focusing rod lens array (12) is fixed by

The angle adjustment plate (81) has farts (
90a) to (90f) are provided. Among these holes (90a) to (9of), hole (90b),
(90a).

(90f) has a tap meeting. Hole (90a), (
90c).

(90e) is not tapped.

On the other hand, the focusing rod lens array (12) has holes (90a), (90c),
At the position corresponding to (90e), a hole (91a),
(91b), (91c) 'frFa get. These holes (91a), (91b), (91c)
is tapped.

Note here that the hole (91a), (9'lb
), 1Hc), the CCD chips (21a) to (21h) are not located, and the hole (9
1a), (91b), (91cl.

CCD chips (21al to (21h)
d! It is the point where it is located. That is, FIG. 3(b)
As shown in , the focusing rod array (12) is not used effectively at all, but only at the locations where the CCD chips (21a) to (21h) are located. Use C
Holes for glAl adjustment (91a) and (91
b) , (91cl t”) was provided.

In such a configuration, the holes (90a), (90
c).

(90e) through the adjustment screw (82) and the hole (91a).
).

At (91b) and (91c), they are screwed together.

Here, for example, the distance between the hole (90a) and the hole (91b) is adjusted by the amount of screw engagement in the hole (91a). Hole (
Distance between hole (90c) and hole (91b), hole (90e)
The same applies to the distance between the hole (91c) and the hole (91c).

Also, holes (9ob), (9oa), (9or
), the adjustment screw (82) is inserted into the
) and the angle adjustment plate (81) are screwed together. At this time, the other end of the adjustment screw (82) is connected to the focusing rod lens array (
Reach the side of 12), hole (9ob), (90d)
, (90f), the distance between the convergent rod lens array (12) and the angle adjustment plate (81) is defined.

The focusing rod lens array (12) is sensitive to heat and is very flexible, but in this way, the distance 1lvI't between the angle adjustment plate (81) and the focusing rod lens array (12) can be adjusted at multiple points. - this deflection is corrected. Moreover, when specifying this distance, the hole (91g).

In terms of (91b) and (91c), the focusing rod lens a)C generates a force directed from the angle adjustment plate (81) toward the focusing rod lens array (12), so it can be more completely Corrected.

Also, by adjusting each adjustment screw (82),
Focusing rod lens array (12) and adjusting plate (8)
1) can be made parallel, that is, the two converging rod lens arrays (12) can be made parallel.

Next, the light source (13) will be explained. The light source (13) in this example is comprised of a fluorescent lamp. In this example, the fluorescent lighting option is an electric lamp.

The functions required of a dense layer sensor as a reading device are:
This is the ability to read images, especially colors, with the same sense as the human eye. The image information obtained by the adhesion sensor is used to visualize it on a display device.

For contact sensors, colors that humans perceive as black must be read as black. A contact sensor should be no better or worse than the human eye when it comes to reading color.

Figure 10 shows the fr curve obtained by Thomson (!: Wrrfht). This curve shows all the visibility characteristics of the human eye according to color. Is it related to wavelength? j'+ + #
The 2 + Is curves correspond to 几, G, and 13, respectively. As is clear from this curve, the human eye has 7
It is not sensitive to light with long wavelengths of 00 nm or more.

On the other hand, as mentioned above, the spectral characteristics of the light receiving parts and color filters of the CCD chips (21a) to (zth) are 700 nm.
It has a finite sensitivity value even to light with a long wavelength of m or more. Therefore, such a color filter, CCI) chip (
When white light is incident on the light receiving sections 21a) to (21h), it is perceived as light with a long wavelength of 700 nm or more.

Therefore, in this embodiment, the fluorescent light of the fluorescent lamp was selected to match the human eye's sensitivity.

In this example, two fluorescent lamps with different spectral wavelengths are used, and the phosphor combinations are White and Btu.
It is.

(1) Wbitc +Blue (Overall 5PD
-11J) The spectral characteristics of these phosphors are shown in FIGS. 11 and 12. Why ite's phosphor is 3Ca
3 (P 04 ) t'Ca (F, Ct)2
: Sb, Mn The spectral characteristics of these are in the wavelength region close to 7000 m, as shown in Figure 11 (this region can be called the long wavelength region in the currently required wavelength region of 700 nm or less). It has a peak at

BAue's phosphor is 3Sr3 (PO4)2 ・Ca
C62: Bu, as shown in Figure 12, 4
It peaks in a wavelength region (which can be called a short wavelength region) close to 5 (lnm).

In this way, these phosphors are composed of two types of phosphors, one having a peak in the short wavelength region and the other having a peak in the long wavelength region, and by increasing the amount of phosphors, C
CI) This compensates for the fact that the sensitivity of the light-receiving parts of the antennas (21a) to (21h) is low in the short wavelength region, and the decrease in the light amount of the filter itself.

A fluorescent lamp having such a phosphor is a type of linear light beam, but due to the influence of the filament, non-uniformity in brightness occurs in the length direction of the pipe.

In this embodiment, in order to obtain uniform illuminance over the width of four people, the tube length of the fluorescent lamp (13) is made long as shown in FIG. 14, and the distance between the filaments is 129.
It was set to 6mm. As a result, uniformity of illuminance within ±5 inches was achieved within A4 size.

Further, the fluorescent lamp in this embodiment is a fluorescent lamp with a reflective film and has an aperture.

However, as shown in FIG. 7, the coincidence point P of the optical axes of the two focusing rod lens arrays (12) and the fluorescent lamp (13)
The angle formed with the center Q of is set to 45°.

With the above groove path, the focusing rod lens array (12)
The image on the document surface (11) is erect and one-to-one
The crystals are formed on the ccu chips (21a) to (211s) with a size of .

That is, the image (elephant) on the document (71) is directly copied onto the CCD chips (21a) to (2111).CC
The light receiving area of the D chips (21a) to (21h) is approximately 14
With a pitch of μm, this equates to 72 bits per mm.

Therefore, when reading in black and white, it is possible to read 72 lines/mm'c' of the image on the document surface (71).This invention is a color contact sensor, and as shown in FIG. The same color filter is installed in the light receiving section of the W, C, and YK. In the end, one pixel is formed by 6 bits, and the resolution is 3
This is equivalent to 6 lines/11'lnl+12 lines/mm (one pixel is 84 μm) during three-color separation.

Next, the electrical system will be explained. 'Electrical system is CCDf
A drive circuit to operate, an analog processing circuit system that converts the output signal from the COD into a form suitable for the image information, and a signal from the analog processing circuit system that converts the signal from the analog processing circuit system into a signal suitable for the recording format of the recording device. It consists of a digital processing circuit system.

First, the drive circuit will be explained. However, in the following explanation,
This is a driving circuit for a CCD chip (21a).

As shown in FIG. 15, this drive circuit has a two-phase clock φ1. φ2. Scan synchronization signal SH, 17 set signal R
8, handles the output signal O8 and sampling hold signal SP.

Inverters (1, 7) are connected to the input terminal of the clock signal φ1.
1) is connected. The output terminal of this inverter (171) is connected in parallel with a resistor (1721) and a speed-up capacitor (17'd). The terminal of this parallel element is connected to the input terminal of the inverter (174). The output terminal of this inverter (174) is a resistor (1
75). The resistor (175) is further connected to the CCD
It is connected to the φ terminal of the chip (21a).

The input terminal of clock signal φ2 and the CCD chip (21a
) is the same as in the case of φ1.

An inverter (171
), resistor (172), capacitor (173),
An inverter (1, 74) is connected. A resistor (176) is connected to this inverter (174). This resistor (17, 6) is the S H of the CCD chip (2Xa)
Connected to the terminal.

The output signal O8 terminal is a PnP transistor (178)
is connected to the emitter of and a load resistor (179).

The collector of the PnP transistor (1781) is a resistor (1
80). The other end of this resistor (180) is grounded. The base of the PnP ) transistor (178) is
It is connected to the O8 terminal of the CCD chip (21a).

In addition, the operation of the CCD chip (21a) 'power supply ■ All terminals (1
81). This terminal (181) is connected to one end of a grounded capacitor (182). One end of this capacitor (182) is connected to the aforementioned resistor (179) and the OD terminal of the CCD chip (21a).

In the above configuration i1, inverter (171) i'
, i zSN7104 were used, two inverters (174) were used as a set, and DSOO26 was used.

This DSOO26 operation signal is supplied from the terminal (181). That is, it is supplied to one end of a capacitor (182) via a resistor (183).

2-phase clock φ1. φ21rJ', CCD chip (2
This is a signal necessary to transfer the weight generated in each bit of 1a). Since this 'I:lj' load is transferred at high speed, there is a possibility that waveform distortion will occur. As a countermeasure to this, the resistance values of resistors (175) and (176) were set to optimal values through experiments.

The scan synchronization signal SH is a signal that distinguishes one scan from the other in transferring charges of the CCD chip (21a).

, IJ upper set number R8 is the first number for erasing the pit after charge has been transferred.

Signal O8 is an output signal from the CCD chip (21'a). According to the specifications, this CCI) chip (21a)
There are 2048 bits of valid signals, dummy No. 1 and CC
The reference black level signal output from D is output. The focus positions of these signals are set accurately. As will be described later, the reference point level 1d is the dark level of the light receiving section, and is used to obtain the true output corresponding to the color.

Next, the analog processing circuit will be explained. This analog processing circuit is connected to each CCD chip (21a) to (21h).
provided for each Here, as a representative CCD chip (2
The steps for 1a) will be explained.

This circuit t, as explained in FIG.
An integrating section (162) that separates and integrates the number 1a for each purple color, an amplifying section (163) that amplifies and adjusts the number 1a from this integrating section (162) for each color, and a color output from this amplifying section Ωtan and A, -D conversion unit (16
4) and 1, which stores the signals from the A/D converter (164), consist of a section Uμ.

The correction unit (161) includes a zero rung circuit (1 (311)) and an amplifier (1612).

The integrating section (162) includes a multiplexer (1621) that separates the output of the amplifier (162) for each color, and integrating circuits (1622a) to (1622c) that divide the output of each color from the multiplexer (1621) into 16 circuits. ).

The amplifier section (163) includes integral circuits (1G22a) to (1G22a) to (1G22a) to (1G22a).
622c) provided for each amplifier (1631a) to (
1631G).

A-D converter (1641 is a selector (1641) that switches and outputs the amplified signal for each color, and an A-D converter (1642) that converts the output from this selector (1641) into a gold digital signal. Consists of.

The memory section has storage areas (1651), (1652), (1653) provided for each w, y, and 'c.
).

The above is the main configuration of the analog processing circuit provided corresponding to the CCD chip (21a), and in addition to these elements, several other control elements are provided. These will be explained together with the operation of the circuit.

First, the CCD chip (21a) is connected to a CCD pulse generator (16) that adjusts the pulse signal of the oscillator (166) as appropriate.
7) is supplied with a group of signals. This number 1 group is C
The scanning synchronization signal 8H, IJ upper set number R8, and two clocks φ1. It is φ2.

A CCD chip (21a
As mentioned above, the output signal from the Q4 is determined by the signal output of each bit forming the light receiving section.

The CCD chip used in this example was TCDIO2C-
1, and the configuration of each bit of the light receiving section of the CCD chip is: from the end, a dummy area, a 6-bit black reference light-shielding area,
1' is an effective reading area of 108 bits, and is again a dummy area.

Here, the light-shielding area is a bit where the light-receiving part is light-shielded,
The output from this region has a charge f associated with the dark current. This voltage is called the reference voltage. The output signal of the effective reading area is a minute amount with respect to this reference voltage. For example, T.C.D.
In 102C-1, the base voltage is 6 to 8 volts, and the output signal of the effective reading area is ioowv with respect to this reference voltage.
It is of the order of.

Therefore, it is not preferable in terms of signal processing to directly absorb the output signal of the effective reading area. Therefore, it is desirable to remove the reference voltage from this signal and output it in accordance with the W, C, and Y filters.

In this embodiment, in order to realize this, a clamp circuit (1611) is provided.岑Clamp circuit (16
11) outputs the difference from this 11L based on the reference 'compression pressure'.

Although the output signal of this clamp circuit (1611) is a very small signal, it is amplified by a certain number of signals by an amplifier (1612). This output No. 16 includes output No. 1 of the valid reading area and an output signal of the dummy area following it.

The effective reading area is w, y as shown in FIG.
, c are pasted on two bits consecutively.

Therefore, when the original is blank and the reflected light is also white light, the output of the clamp circuit (1611) or amplifier (1612) is a signal (181W) corresponding to 2 bits of W, as shown in FIG. A signal corresponding to 2 bits of Y (1
81Y), a signal corresponding to 2 bits of C (181C
) are consecutive. However, in FIG. 17, the separation interval between each transfer key code is ignored. The same applies to the drawings below.

The output signal of the amplifier (1612) is sent to the multiplexer (1
621). This multiplexer (1621
) has a control f from the A/D pulse generator (1613).
Wholesale signals are supplied and the human power relationship is switched.

This control signal is a two-phase clock φ1.0 from the CCD pulse generator (167). It is a signal related to φ2, and C
CJ) is synchronized with the signal transfer from the chip (21a). For the 2-bit signal from the beginning, C is connected to the signal line (16
"1" is output to 9a). For the next 2-bit signal, the signal line (169b) K r 1 j is output. f, the next 2-bit signal is connected to the signal line (
169c) Vcr I J is output.

When “1” is input to the signal line (109a), the amplifier (1
The output number 1 of 612+, ) is the integrating circuit (1622a)
(This is an integrating circuit corresponding to the W filter.)

When rlJ is input to the signal line (169b), the integration circuit (
1612) (This is an integrating circuit corresponding to the filter of Y.) The output signal is input to the integrating circuit (x622b) (This is an integrating circuit corresponding to the filter of c.). 1g
When rLJ is input to the line (169c), the integration circuit (1
612) is inputted to the integrating circuit (11322c).

That is, the output of the three terminals of the multiplexer (1621) becomes a signal corresponding to the color of the filter, as shown in FIGS. 18(a) and (b)+(C).

Integrating circuit (1622a), (1622b), (
1622c), such a signal is integrated over time to. Then, as shown in Fig. 19, the solid line (
A constant voltage v such as 201) causes a charge to be held in the capacitor.

Next, the integration circuit (1622a), (1622b)
, (1622c) and the next signal is input, the A/D pulse generator (1622c)
68), integrating circuit (1622a), (1622
b) A control signal instructing discharge is supplied to (1622c). For example, the integration times % corresponding to the filter of W
In (1622a), time t after a certain period of time has elapsed,
(after a constant value is achieved by the product 1 minute) AZ
From D pulse generator (168) to 1g line (182a)
A signal “1” is supplied through the integrator circuit (1622a
) is open. Then, the charge accumulated in the capacitor is discharged, and the integrating circuit (1622a) is initialized.

Such an integration circuit (1622a), (1622b
), (1622c) has the following effects. For example, as shown in FIG.
Consider a case where noise is added to the signal transferred from a).

COD noise sources include noise due to charge injection into the device, noise due to fluctuations in the amount of charge during charge transfer, and noise during charge detection. Also, of the analog processing circuit here, the correction section (161) is C
It is provided on the ceramic substrate (22) together with the OD chip (21a). Other than this, it is installed on the main body side and connected with a signal line. Line 1 tends to pick up noise.

These noises cast a huge shadow over the image information.

However, since the process of integration is not a point process in time, but a one-dimensional process in time, the broken line (20
As shown in 2), the average ratio is applied, and the influence is almost eliminated. However, each integrating circuit (1622a).

The outputs from (1622b) and (1622c) do not match.

Each of the 1d numbers from which the influence of noise has been removed in this way is
Amplifier (1631a), (1631b), (1
631c) with individual amplification factors. Here, FIG. 21(a).

(b), (c) K indicates each amplifier (
1631a).

The outputs from (1631b) and (1631c) are the same 1
Each amplification factor is set so that

Here, the points to note are Fig. 22 (a), (b),
As shown in (C), each amplifier (1631a),
(1631b).

The signal from (1631c) is at a completely different time (point being output to the river).

Such a signal is input to the selector (1641). This selector (1b41) is supplied with an I control signal from an A/D pulse generator (168) via a signal line (183) ffi. This system (i1111a
, COD Chisov.

It takes one period for the signal from (21a) to be output for 2 bits.

When this control signal is supplied, -hirefta (1641
) switches the relationship between input and output. flJ,
At a certain time, when a control signal is supplied to the selector (1641) when the output J of the amplifier (1631al) is associated with the output of the selector (1641), the output of the amplifier (1631al) is
The output of the drinker (1631b) becomes the output of the selector (1641).

Next, when a control signal is supplied to the selector (1641), the output of the amplifier (1631c) changes to the selector (1641).
41) is the output.

In other words, the output of the selector (1641) is:
W.

Output signals corresponding to the Y and C filters appear alternately. Such a signal is processed by a 6-bit A-D converter (164
2) is converted into a digital signal. The timing of this conversion is determined by the A-D conversion start signal supplied from the A/D pulse generator (168) via the signal line (184) e.

As described above, 4-pit digital signals corresponding to the W, Y, and C filters appear alternately as the output signal of the AD converter (1642).

Let us now evaluate the number of digital signals.

In this embodiment, the signal tone from the continuous COD chip (21a) is separated into three colors of 2 bits each by a multiplexer (1621). Rainbow, integral circuit (16
22a) to (1622c) integrate the signal of each color 2 bits at a time. Therefore, the number of signals for each color is C
The number of signals from the OD chip (21a) is one-sixth of the number of signals.

For example, the effective reading area in the CCD tick (21a) is 1
908 pits, but the output signal according to one filter is 1/6th of that. That is, there are 318 pits.

Such a signal is stored in the storage section (165).

The storage unit (165) stores each COD chip (21al to (
21h), and storage areas are set according to w, y, and c. For the COD chip (21a), the layer regions (1651), (16521, (16
53) is set.

Here, each record 1.橡 area (1651), (165
2), (1653) is static RAM (Ra
ndom Access Memory)
----11 The capacity of each is 324×6 bits.

Since the output No. 11 of the eight-way converter (1642) is 6 bits, each storage area (16511, (16521, (1
653) is 324 color signals? Can be remembered. Therefore,
Addresses are set from 0 to 323. Here, the color signal is 1g, which is caused by the signal from the light receiving section to which w, y, and c are pasted.

Below, this storage area (1651), (16521, (
1653) and the reading of information, it should be noted that the CCD chips (21a) to (
21h) and signal overlap removal using the convergent rod lens array (12).

Now, storage area (1651), (1652), (1
653), from the A-D converter (1642),
Digital signals are constantly coming in. Therefore, it is necessary to control where the data is stored.

In this embodiment, a memory counter (185), an address counter (186), an address decoder (187), an OR circuit (1881), and a read/write control (189) are provided to carry out this control.

When a write command “1” is issued from the read/write control (189), the memory control (185)
Write pulse kW synchronized with A-D conversion start signal
, Y, C storage area (1651), (1652
).

(1653). By this signal, the storage RAM that constitutes the storage areas (16511, (1652), (1653) is released from the prohibited state of information transfer. In other words, the storage areas (16511, (1-1-
652) and (1653) are supplied with digital signals almost continuously, but there are storage areas (1651), which can be retracted according to the output signal of the A-Df converter (1642).
(1652) and (1653) are selected.

Furthermore, clock pulses are supplied from the A-D pulse generator (168) to the memory control (185).

In this state, the memory control (185) supplies this clock pulse to the address counter (186).

This address counter (186) is a normal counter, and its output is simultaneously supplied to the storage unit U corresponding to all the CCD chips (21a) to (21h).
Used as an address.

On the other hand, the input command "1" from the read/write control (1891) is supplied to the OR circuit (188).The output of the OR circuit (188) is used as a tick selector. When the signal "1" is supplied from the OR circuit (188) to the Σ54-, and at the same time the write signal is supplied, the write signal is supplied, and when the read signal is supplied, the glare signal is output from the address. Follow the instructions.

billion area (1 (351), (1652), (16
531 contains color No. 18 and dummy' due to the effective reading area.
He lives with No. 1 Yo from Uta J Ebisu.

In this way, when the storage of information in the storage unit Ωl is completed, the memory control (185) notifies the read/write control (189).Then, the read/write control (189) stops selecting the input instruction. do.

On the other hand, the memory control (185) controls the storage area (1
651), (1652), and (1653). This V is therefore the first chain acid (1651).

(1 (i52), (1653) All constituent static RAMs are released from the read-inhibited state.

Also, a clock pulse is given from the memory control (185) to the address counter Bs6), and each time a clock pulse is given, the address counter (186)
)'s output is added one by one. And the setting number A (
318≦A≦324) If K is delayed, the contents will be cleared,
Address decoder (187) K Informs this.

In the address decoder (187), the address counter (
If the contents of 1861 have never reached the set number of kV, the memory section (1) corresponding to the CCD chip (21a)
Select sweet purple. - When the set number A is reached, select all the notes (CC) corresponding to the chip (21b).

Furthermore, when the setting is reached again, the CCD chip (2IC)
All the storage units (165) corresponding to are selected.

Below, perform VC in the same manner and select the section 1 corresponding to the CCD chip (21h).

The readout of the original part is performed using the chip select signal from the address decoder (1117),
Address data and 1. This is done by supplying an excursion pulse.

That is, information is read out sequentially from the storage section corresponding to the CCD chip (21a) to the CCD f-lag (21a).
Note 1. corresponding to h) and λ(j is used for part (1) phrase progression. However, information corresponding to the w, y, and c filters is extracted at the same time.

The electrical component here is the aforementioned setting number A.

This setting number A is for each storage area (1651), (1
652).

Of course, I have already decided where within (1653) the address will be exposed. As mentioned above, since the light receiving parts of the CCI chips (21a) to (21h) are arranged with some overlap between them, it is necessary to remove this overlap.

In this example, /7! rStorage/Area Dos+), (
1652).

In contrast to (1653), since the data is stored from the beginning of the effective reading area of the CCD chip (21a), the timmy area other than the effective reading area is an overlapping area and must be removed.

Therefore, the thickness of the filter is determined by the error and the CCD chip (2
1a) - 4 Geometric error of the gap to install (211+)?
Considering this, once the device is completed, we will read out all the information from the storage section Ωj without deciding on the set number A. Then, the effective reading area and the overlapping area can be easily distinguished. Therefore, At is determined from this result. Therefore, note 1. There is a possibility that C1 person will change due to bar<ll (165).

The image information read in this way is
1) It faithfully conveys the image on the same line above.

The linear light sources are only fluorescent lights (, L, BI) (L
tlht Emittin, @Diode) but there is no difference.

The optical system may be formed by arranging a large number of microlenses made of St resin instead of a focusing rod lens array.

This lens makes the image in the provided area (micro area) erect, sharp, and approximately the same size.

As the image sensor, not only the CcD but also any available image sensor can be used.

[Brief explanation of drawings]

FIG. 1 is a perspective view of a contact sensor according to an embodiment, and FIG. 2 is a perspective view of a sensor package. 3 and 4 are diagrams showing the arrangement of CCD chips,
Figure 5 is a spectral characteristic diagram of the filter attached to the light receiving part of the CCD chip, Figure 6 is a spectral characteristic diagram of the light receiving part of the CCD chip, Figure 7 is a side view of the contact sensor, and Figure 8 2
and FIG. 9 show the adjustment means of the converging rod lens array, FIG. 8 is a perspective view thereof, and FIG.
Figure 0 shows the determination of TI+omson1- and Wrifht. A diagram showing a %i curve,
iJG l Figures 1 and 12 are spectral characteristic diagrams of each phosphor in a fluorescent lamp as a light source, Figure 13 is a diagram showing the relative brightness of each part of the fluorescent lamp, and Figure 14 is a diagram showing the relative brightness of each part of the fluorescent lamp. FIG. 15 is a diagram showing the arrangement of filters in the light receiving section of C.
FIG. 16 is a diagram showing the drive circuit of the OD chip, FIG. 16 is a diagram showing the analog processing circuit, and FIGS. 17 to 21 are diagrams showing No. 1g of each part in the analog processing circuit. (I2-1... 11 temples 110..., ToL, ',
S'Tshii. (m-・Flood 5 pressure (21al, CAlbl bar 21c), ()ld,) nail 1la), (λli), (ml)), +1lb)・-
CCD-t-,,7',. l (embedding Cold burial t @+1 Issuing confirmed (Kei Exploration Thin Fig. 1/':S Fig. 2 Fig. 3 (tl, n41 μ Fig. 5 Fig. 10 Country No. ■ Fig. 12 Fig. 1) (nm) Figure 13 Figure 14 VA

Claims (1)

[Claims] A contact sensor comprising a plurality of image sensors arranged in a row, and reading signals from each of the image sensors based on one end of the image sensor.