JPS59215171A - Picture reader - Google Patents

Abstract

PURPOSE:To attain correction with high accuracy by changing over a face arranged with a reference color for correcting shading and a face having a marker part for correcting joint at each correction with each other and arranging them on a reading position to eliminate the effect of temperature change and mechanical vibration or the like. CONSTITUTION:The position of an original A to be read inserted to an original inserting inlet 1 is detected by an original sheet detecting switch 2, the original A is fed to the reading position by paper feeding roller pair 3a, 3b and the position is regulated by guide plates 4a, 4b. This original reading position is provided with a reading window 5 made of transparent glass plate and a reference member 6 for correction of roller shape by clipping an original passing face. Further, the position of the original A after reading is regulated by guide plates 7a, 7b and the original A is discharged by discharge paper roller pair 8a, 8b. Then, a reference white color referenced at the shading correction before scanning of the original is arranged to a circumferential face 6a of the member 6. Moreover, a marker section 6b such as black color is provided at the center in order to use two solid-state image pickup elements, and the marker part 6b is used as a joint area of the solid-state image pickup elements.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention (d) relates to a device used in a facsimile machine, etc., which converts a sheet-like original image into an electrical signal using a plurality of solid-state image sensors and reads it, particularly shading correction and splice correction. The present invention relates to an image reading device that electrically processes images.

Generally, in an image reading device using a CCD'4 solid-state image sensor, it is difficult to obtain a CCD with a large number of bits due to the current situation, so in order to read a wide range of documents with high pixel density, it is necessary to It is necessary to use CCD1 on the algebraic side corresponding to . However, when a plurality of CCU'i are used, optical discontinuity occurs at the boundary of the reading area of each CCI, even though the document is the same. Therefore,
During manufacturing, the mounting position of each CCD is mechanically adjusted in advance, but since the size of one bit of a CCD is usually more than ten microns, it is natural that the mounting position of each CCD is adjusted mechanically.
/It is necessary to adjust the position according to the order. Therefore, simply setting the initial mounting position of each CCD at the time of manufacturing will result in temperature changes, mechanical vibrations, etc. over time. Under the influence of various factors such as c, subtle changes occur in the positional relationship. This method is called full splice correction, which eliminates the optical discontinuities mentioned above and successfully connects the scanned images of the same document, but as mentioned above, mechanically performing splice correction has many problems in terms of accuracy. .

On the other hand, there are variations in the sensitivity of each CCD, and in addition to this, there are variations in the amount of light and illuminance due to the illumination lamp and optical system, so in order to obtain a high-quality image, it is necessary to correct and smooth these variations. In other words, it is necessary to perform shading correction before the document is moved. but,
This is also affected by various factors such as the life characteristics and temperature characteristics of the 1-'I=5 lighting lamp over time, so it is not possible to overcome the inconveniences caused by these effects by simply setting the initial settings at the time of manufacturing. Can not.

Therefore, in the conventional image reading device for sheet-like originals, by making the original guide plate provided above the image reading unit white, the white color of this guide plate is used for shading correction before scanning the original. The shading correction is performed electrically using the image as a reference color every - times of image original reading operation. However, if we try to perform seam correction electrically, we need to set the area that each CCD reads overlappingly, that is, the seam area, in the image reading section.
The only method that can be considered is to fix a marker section of black or other color on the white surface of the document guide plate above it, and to perform electrical processing by reading this marker section redundantly with each CCI. do not have. However, in such a case, when a transparent document is read, a problem arises in that even the fixed marker portion is read together with the document image.

Therefore, when reading normal sheet-like original images, shading correction can be processed electrically as described above, but seam correction is initialized mechanically using a separate marker. There was only one. For this reason, for the reason mentioned above, the x9 connection area is likely to shift, and as a result,
It has been difficult to obtain a high-quality blemish image because the images in the connecting areas overlap or are separated from each other.

The present invention has been made in order to solve the above problem, and it is possible to correct the problem by connecting a marker part of a color different from the standard color, for example, black, in an Aij in which a standard color such as white for 7 aging correction is arranged. It has a reference member for correction, such as a hole-shaped or flat plate-shaped correction member, which is provided for the purpose of adjusting the reference color surface to the original reading position during the seven-step scanning correction before scanning the original to be read, and the marker to the original reading position during the transition correction. By operating and setting the correction reference member at the position where the f8i image is to be read, that is, the one-way area, the shading correction and connection correction are electrically performed for each reading operation. An object of the present invention is to provide an image reading device which can perform high quality image processing and read images of high quality without unevenness.

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a schematic diagram showing essential parts of an embodiment of the present invention.

Here, ■ is the document insertion slot υ used when moving a sheet-shaped original to be read and performing the reading operation, and this insertion slot l
Then, the document inspection switch 2 detects the document position. 3a, 3
b is a pair of paper feeding rollers for guiding the original A to the original reading position, and guide plates 4a and 4b regulate the position of the original A.

At the document reading position, a reading window 5 made of a transparent glass plate or the like and a roller-shaped correction reference member 6, which is a feature of the present invention, are provided with the document passing surface in between. Reference numerals 7a and 7b are guide plates that regulate the position of the original A after reading, and 8a and 8b are a pair of discharge rollers that eject the original A.

Here, the correction reference member 6 will be explained in more detail with reference to FIG. 2. On the circumferential surface 6a of the correction reference member 6,
For example, white is used as a reference color for reference during shading correction before scanning the original. Also, for example, CC
When reading an image using two solid-state image sensors such as D, the position corresponding to the joint area during joint correction before reading, that is, the center position in the longitudinal direction of the correction reference member 6. Marker sound IL such as black elongated in the circumferential direction
Ci l) is provided. The correction reference member 6 with the colored surfaces as described above has a center protruding from both ends! q
It is rotatably supported by bearings (not shown) at I16c and 6d, and by rotating the correction reference member 6 by a predetermined amount before scanning the original, the marker portion 6b is arranged within the connecting area of the solid-state image sensor. Installed.

A pair of document illumination lamps 9 are provided below the number window to illuminate the document surface or the circumference of the correction reference member 60. The reflected light is reflected by the reflecting mirror 10. The light is guided through a lens 11 to a solid-state image sensor 12 such as a CCD, and the optical signal is converted into an electrical signal.

Next, the operation of this embodiment will be described. When the original A is inserted into the original insertion slot 1 and detected by the original detection switch 2, the correction reference member 6 moves its reference color surface so that it faces the original reading position, that is, the center of the reading window 5. , the position is set by a drive source (not shown). Therefore, the reference color plane is the original illumination lamp 9.
The solid-state image sensor 12 reads the reference color.
Convert to electrical signal. This signal includes the document illumination lamp 9.
The signals include Xi unevenness, light intensity unevenness due to the lens 11, and sensitivity unevenness of each bit of the solid-state image sensor 12. Therefore, based on this signal, shading correction is performed by a shading correction circuit as described later. It will be done.

When the shading correction process is completed, 1. The correction reference member 6 is rotated by a predetermined angle by a drive source (not shown) and positioned so that the surface covered by the black marker portion 6b faces the reading position, that is, the center of the reading window 5. Then, before the original is scanned, a cross-stitch correction circuit as described later operates to complete the joint correction process for the plurality of solid-state image sensors, and the correction reference member 6 returns to its previous state 9. The reference color plane faces the reading position again. In this manner, when the shading correction and seam correction processing are completed, the document is transferred to the paper feed roller pair 3a, 3b at the reading position, that is, the reading window 5.
The document is fed into the center of the document, where image reading, that is, scanning of the document, is performed by receiving light from the J-type illumination lamp 9, and then the document is discharged by a pair of discharge +lfa rollers Ba and 8b.

Next, other embodiments of the present invention will be described. FIG. 3 is a schematic diagram showing the result. This uses 20 plate-shaped correction reference members for the A6-shaped correction reference part shown in FIG. 1 or 2. This will be explained in more detail with reference to FIG.

For example, white is arranged on one plane 20a of the correction reference member 20 as a reference color for reference during the 7-day correction before scanning the original. In addition, on the platform where two solid-state image sensors such as COD are used to read images, there is a long and narrow black mark at the position corresponding to the seam area during seam correction before Harane scanning, that is, at the center position in the longitudinal direction. marker section 20
1) will be provided. Further, the correction reference BI sheet 20 having a flat surface colored as shown in FIG. The marker portion 20'b is removed so as to enter the connecting area of the solid-state image sensor by sliding the height of the steam. Here, the correction reference member 20 is attached to the solenoid It is driven by 21.

The movement of the hooked lever 22 may be transmitted to the correction reference member 20 by, for example, forming a protrusion on a part of the correction reference member 20 and forming a groove on one end of the hooked lever 22. This is carried out by means such as engaging the protrusion and the groove.

Next, the operation of this embodiment will be explained, but since it is similar to the previous embodiment and the 11th embodiment, a brief explanation will be provided. That is, when the insertion of the original A is detected by the original detection switch 2, the correction reference member 2, which is positioned on the reference color plane before scanning the original, is activated.
Shading correction is performed by reading 0. When the seven-gage lottery process is completed, the correction reference member 20 slides to a predetermined position by driving the solenoid 21, the marker section 20b enters the reading position, and the connection correction is performed before scanning the document. When the skew correction process is completed, the solenoid 21 is driven, the reference color plane of the correction reference member 20 faces the reading position again, and then the original image is read, that is, the original is scanned.

In the above-described embodiment, shading correction is started from the operation of the original detection switch 2 in the original insertion slot 1, but it may also be started when the power of the apparatus is turned on or when the copy start key is activated.

Therefore, an electric circuit for performing shading correction and connection correction in the above-described embodiment will be described below.

FIG. 5 is a block diagram schematically showing the flow of electrical signals when the correction reference member or the original image undergoes shading correction, seam correction, and image reading. Here, two solid-state image sensors (CCIXI) 12 a
It is assumed that the CCD (2) 12b is lined up in a line in the main scanning direction, that is, in the direction intersecting the original moving direction, to carry out skinning. First, when shading correction is performed in row 4c5, the original from the reference color plane located at reading position B is transmitted to CCD (1) 12 a,
The CCD (2) 12b reads the data and converts it to electrical 1M code. The signal from CCD (1) 12a is sent to the amplifier (AMIF
) 30a, and converted into a digital signal by the A/D conversion circuit 31a, and then sent to the shading correction circuit 32.
Enter a. The input signal is temporarily stored as a shading value by the operation described below. Also, CCD
(2) Similarly, the signal from 12b is also amplifier (AMP) 3
0 b and is temporarily stored as shading 1 shift in the shading correction circuit 32b via the A/D conversion circuit 31b. Next, the position of the correction reference member is changed, and the CCD (1) 12a and CCD (2) 12b capture the marker portion located in the connecting region C by *+i and convert it into an electrical signal. The signals are each sent to an amplifier (AMP) 30
a, 30b, Az'D conversion circuits 31a, 31b and sent to shading correction circuits 32a, 32b. Here, the signal and the shading value are then subjected to arithmetic processing in the following manner to perform shading correction, and the connection correction circuit 33 performs connection correction on these signals.

The connection correction circuit 33 will be described in detail later.

In this way, electrical settings for shading correction and connection correction are performed. After that, the original image is moved to reading position B.
By moving the image, the image is transferred to the CCD (1) 12a.
and read by CCD (2) 12b. The read signal is sent to amplifiers (AMP) 30a, 30b and A/D conversion circuit 3.
1a and 31b, the shading correction circuits 32a and 321) perform 7 aging corrections, and after the connection correction is performed using the connection correction circuit 33, the image is displayed as an image number on a printer 34 or the like.

Next, the aforementioned shading correction circuit will be explained in detail. FIG. 6 shows an example of the shading correction circuit, showing only the device 4 that processes the output of one CCD (, 1112a. In FIG. 6, when starting the shading correction, the selector 40 is switched to the A side. .Bit serial image signal from CCD (1) 12a (amplified by Ij amplifier 30a, A/1) Transformer: 3
1a, this digital signal is converted into a digital signal by a random access memory (hereinafter referred to as ItAM).

) written in 41. Address 7i of this RAM41
Both f11 and CCD (IH2a) use a clock signal that is in synchronization with a clock signal of a clock generator 45 (not shown in FIG. 5) for IH2a.
It is done through. Next, when scanning the portion including the marker section 4 and scanning the document, the selector 40 is switched to the B side. The portion including the marker portion 6b or 20b and the image data obtained by reading the document are stored in a multiplication read-only memory (hereinafter read-only memory is referred to as ROM).
42. Multiplication It OrA /l 2
A signal from the RAM 41 containing the shading values obtained in the shading sequence described above is given to the other input of the shading sequence. Address control of the multiplier 110M42 is performed via the address counter 43ff using a clock signal that is synchronous with the clock signal of the clock generator 45. The multiplication ROM 42 is programmed to output a code m/n when the input from the RAM 41 is n and the image data is m, and shading is corrected by this logic. The output of the multiplication ROM 42 is connected to the output of the dither ROM 46, which is a threshold value determination circuit, and the comparison circuit 4.
9, binarized image data is obtained and input to the shift memory (1) 50.

A signal in synchronization with the clock signal of the clock generator 45 is input to the main scanning counter 47 for the dither ROM 46, and the main scanning counter 47 is input to the sub-scanning counter 48.
A count pulse is input at every fixed count of , and the synchronization with the sub-scanning of the document is maintained.

Next, five means for automatically correcting the connection of the output signals of the CCD (1) 12a and the C'CD (2) 12b will be explained.

In the same way as in FIG. 6, the output of CCD (1) 12a is stored in the shift memory <1) 50.
(2) The output of 12b is stored in shift memory (2) (shown in FIG. 9). Figure 7 shows the principle of V-batch correction, where a is the scanning area of CCD (1) 12 a, and b is the CC
The scanning area of D(2)12b is shown, and the scanning direction is indicated by Y in the figure. When the original illumination lamp 9 illuminates the part including the r+1i+1i black reference or 2o marker section 6b or 20b, the last 128 bits of the bit serial signal from the CCD (1) 12a are shown in FIG.
2) The first 12 bits of a 12b serial signal
8 bits are shown in Figure d. Here, 128 bits is 16
8 mm on the document surface when the resolution is 7 mm
Corresponds to the width. Then, each of these 128 bits is extracted and expanded into RAM. This state is shown in figure e and f.
As shown in the figure above, if the ranges scanned by the two optical systems are set to overlap as described above, white bits will always appear before and after the 128 bits developed in ftAM, and the marker part described above The black bit corresponding to 6b or 20b is sandwiched.

Therefore, the number of lower white bits (L, W, ) in the output of CCD (1) 12a and the number of upper white bits (U, W, >) in the output of CCD (2) 12 b (7) and the number of black bits (B
) is added to the number of bits when reading it from the soft memory.
and CCD (2H2b) (takes out the D output as a continuous signal.

Next, FIG. 9 shows a connection correction circuit that executes the above operation. FIG.
The manner in which the first 128 bits are extracted from the output signal of b. The basics of this extraction and 9-processing are to read data from both sides 16 times in 8-bit units, and to develop this into IIAM. The device operates on a write clock signal and a read clock 1d number 03. First, the central processing unit (hereinafter referred to as CPU) sets the write address counter (1) and read address counter (2) to a down count mode. Note that the read address counter (1) is always preset to up count mode.

Then, the I10 register (1) is set to address 2592 (A20H in hex code), and the I10 register (3) is set to address 8 (same <8H). In this circuit, when the CPU activates the test terminal T of the I10 register (4), the read circuit (not shown) generates a video enable signal (EN in FIG. 8) indicating one main scanning period once; CC1) (1) and CCD (2) are designed to send out bit-serial signals.

The video enable signal EN is input to each set terminal S of the flip-flops (11 and (2)) in FIG.
(1) The video signal of the duck is the same as the shift memory (1) (shift memo January 1 in FIG. 6) 50. ) and 8-bit shift register (1), and the video signal from C0D (2) is sent to shift memory (2) and 8-bit shift register (21).
K is input. 8-bit shift register (1) and (
2) both start the shift operation upon activation of the video enable signal EN, and end the same operation in response to the count completion signal RPC1 from the write address counter (1) and read address counter (2), respectively. Note that this operation is performed by flip-flops (1) and (2), respectively.
The process is carried out via the line t.

Therefore, when the CPU activates the test terminal 'r mentioned above, when the video enable signal is completed, the last 8 bits of the signal from CCD (1) remain in the 8-bit shift register (IJ), and the 8-bit Shift register (2
), the first 8 bits of the CCD (2) signal remain. Next, the CPU reads the contents of these shift registers and expands them onto the RAM. To process the next 8 bits, write (2592
-8, ) address is set in the I10 register (3) and (8+8 address) is processed in the same manner as above. These processes are 1
It repeats six times and extracts 128 bits.

When the calculation of the stitching amount is completed, the document is scanned, but before this, the CPU writes 25 to the I10 register (2).
The address 04 is set in the Ilo register (3) (2592 intermittently). This situation is shown in Figure 7g and 1.
], and g in the same figure shows the output of CCD (1), which consists of 88 bits of blank space from the beginning), 2376 bits corresponding to the information limit of 148.511 m, and 12 bits for connection processing.
It consists of 8 bits totaling 2592 bits, while Figure 11 shows the output of CCD (2), which also has a total of 2592 bits.
The bits are obvious, but the first CON indicates the amount of connection. Note that the video output is output via the OR circuit on the output side of the shift memories (1) and (2).

FIG. 8 shows the video enable signal < EN ), CC
The output word from D(1) (CCD(1)), CCD(
2) shows the time relationship between the output signal ((:'CD(2))) and the read clock (ri,).In addition to the above, the I10 register (4) in FIG. (1) and read address counter (2), and selection of address selectors (1) and (2).
The shading correction circuit shown in the figure can also be controlled, and the RAM, I't, and OM in FIG. 9 can also be used as those in FIG. 6.

As described above, in the present invention, the surface with the reference color for shading correction and the surface with the marker part for seam correction are switched and placed at the reading position for each correction. It is now possible to electrically process various corrections when reading originals, and to perform them for each reading operation. This has the extremely remarkable effect of making it possible to perform highly accurate corrections.

[Brief explanation of the drawing]

FIG. 1 is a schematic diagram showing the main parts of an embodiment of the present invention, FIG. 2 is an enlarged perspective view showing the roller-shaped correction reference member used in FIG. 1, and FIG. 4 is an enlarged perspective view showing the flat correction reference member used in FIG. 3; FIG. 5 is a schematic diagram showing the main parts of the embodiment; FIG. 5 is an electrical circuit diagram used in each embodiment of the present invention; -A block diagram showing an example; FIG. 6 is a block diagram showing an example of a shading correction circuit that is applied to each embodiment of the present invention; FIG. 7 is an explanatory diagram showing the principle of joint correction; FIG. FIG. 9 is a block diagram showing an example of a connection correction circuit used in each embodiment of the present invention. DESCRIPTION OF SYMBOLS 1... Original insertion slot, 2... Original detection switch, 3a, 3b... Paper feed roller pair, 4 a, 4 b, 7 a, 7 b-guide plate, 5... Reading window, 6 ... (roller-shaped) reference member for correction, 6 a - circumferential surface, 6 b ... marker section, 6 c, 6 d ... central axis, 8 a + 8 b ... paper ejection roller pair, 9.
...Original illumination lamp, 10...Reflector, 11, lla
, llb'-lens, 12-... solid-state image sensor, 12 a =-CCD (1)
, 12b... CCI] 2), 20... (flat plate-shaped) correction reference member, 20a... plane, 20b... marker portion, 21... solenoid, 22... hook-shaped lever, 30 a, 30
b-AMP. 31 a, 3 l b-a/D conversion circuit, 32 a,
'321)...Shading correction circuit, 33...
Connection correction circuit. Patent applicant Caine Co., Ltd. (32a Figure 6 Figure 7 Figure I8Wi

Claims (1)

[Scope of Claims] 1. While moving a near-1-shaped original to be taken, images of the original are displayed in a plurality of images arranged side by side in a main scanning direction intersecting the moving direction. In the image+i (C capture/equipment) that is read by a solid-state image sensor, a marker part for joint correction of a color different from the base color is provided in the plane on which the standard color for Shaping IIi is arranged. A reference member is provided, and the reference color surface is set to the original at the JC position 16 at the time of the sonning sleeve before scanning the original.
In addition, during the shift correction, the marker portion is placed at the document reading position where the plurality of solid-state image sensing devices are overlapped, respectively, for the correction month 1! ＜i iqb 1 mountain 11゛禾1n″cJ characterized by setting a shrine
1 to 11(. 2, +11i 11E Ill reference part must be a roller-like member, and the marker part is provided on the circumferential surface of the quasi-color). The image reading device according to claim 1. 3. The image reading device according to claim 1, wherein the reference member for correction is a flat member, and a marker portion is provided on one plane on which the reference color is arranged.