JPH0918712A - Image processor and image read sensor used for same - Google Patents

Abstract

PURPOSE: To improve the image quality of a printed image in the image processor conducting shading correction. CONSTITUTION: An A/D converter means 4 at a pre-stage of a shading correction means 6 is configured to have 10-bit resolution and an image signal after shading correction takes true 256 gradation by selecting a bit width of an input signal of a shading correction means 6 so as to be larger than 8-bit being a bit width of an output signal so that a value far much more than the number of the value taken by an image signal at an output terminal of the shading correction means 6 is expressed by the image signal at an input terminal. Furthermore, a photoelectric conversion means 2 and an analog adjustment means 3 are configured on the same wafer to reduce the number of analog elements in use thereby reducing the effect due to noise indruding between the elements and a temperature change in each element.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus for forming (recording) a print image on a sheet based on information of a read original, and particularly to an analog image signal and a digital image signal. It is suitable for use in an image processing apparatus that performs processing for correcting the level variation of each pixel by converting

[0002]

2. Description of the Related Art Conventionally, an image processing apparatus such as a copying machine for converting an analog image signal generated by an image reading sensor into a digital image signal and correcting the variation in the level of each pixel of the sensor has been proposed. Has been done.

In this type of image processing apparatus, first,
With respect to the RGB signals obtained by the image reading sensor, the overall variation in the level of each R (red), G (green), and B (blue) image signal is adjusted by the analog signal processing circuit (gain adjustment and offset adjustment).

Then, the analog image signal whose overall variation is adjusted is converted into an 8-bit digital signal by using an A / D converter, and so-called shading correction is performed to correct the variation of each pixel. Further, an 8-bit (same as the number of output bits of the A / D converter) image signal is obtained.

In the conventional device as described above, since an image that can be represented by 8 bits has 256 gradations, first, the maximum value of the output of the A / D converter is 200 so as not to exceed it. Adjust with an analog system so that it is about the same level. In the shading correction after digital conversion, for example, 200 gradations are normalized to 256 gradations to obtain an 8-bit signal of 256 gradations.

[0006]

However, the shading-corrected signal is an 8-bit signal in terms of numerical values, and it is possible to obtain 256 gradations, but in reality, before normalization. It was possible to obtain only about 200 gradations, and it was not possible to express the true 256 gradations. Therefore, there is a problem that a good image quality cannot be sufficiently obtained.

Further, since the semiconductor manufacturing process used in the image reading sensor is different from the semiconductor manufacturing process used in the analog signal processing circuit, the image reading sensor and the analog signal processing circuit are simply the same silicon wafer. Could not be configured on. Therefore, the configuration from the reading of the image to the shading correction cannot be made smaller than at least the configurations of the image reading sensor, the analog signal processing circuit, the A / D converter, and the shading correction circuit.

Further, since many analog-related elements are used, noise may be added to the analog image signal between the elements, or the elements may change in temperature due to the large number of steps. There were many. Therefore, there is a problem in that the quality of the image signal and eventually the quality of the printed image may be degraded.

The present invention has been made in order to solve such a problem, and makes it possible to obtain a true number of gradations that can be expressed by 8 bits and reduce the number of analog elements used. It is an object of the present invention to improve the quality of a printed image by making it possible to reduce the influence of noise mixing and temperature change.

[0010]

The image processing apparatus of the present invention is an image processing apparatus having shading correction means for shading-correcting an input digital image signal, wherein the bit width of the input signal of the shading correction means is the shading. It is characterized in that it is made larger than the bit width of the output signal of the correction means.

Another feature of the present invention is image reading means for reading an image and generating an analog image signal, and A / D conversion for converting the analog image signal output from the image reading means into a digital image signal. Means and
An image processing apparatus having shading correction means for shading-correcting a digital image signal output from the A / D conversion means, wherein the A / D conversion means outputs the analog image signal to the shading correction means. It is characterized in that it is converted into a digital image signal having a bit width larger than the bit width of the signal.

Another feature of the present invention is that the value of the digital image signal at the input end of the shading correction means is changed from the value when the white reference image is read by the image reading means to the black reference image. It is characterized by comprising adjusting means for adjusting so that the value obtained by subtracting the value when read is equal to or larger than the maximum value that can be represented by the bit width of the output signal of the shading correction means.

Another feature of the present invention is that the image reading means is a photoelectric conversion means for converting an optical signal obtained at the time of reading an image into an electric signal, and photoelectrically converted by the photoelectric conversion means. Amplifying means for amplifying the electric signal, clamping means for clamping the electric signal amplified by the amplifying means, and electric signal clamped by the clamping means for the A / D
And level shift means for shifting the level so as to match the signal level of the conversion means.

Another feature of the present invention is that at least a photoelectric conversion means for converting an optical signal obtained at the time of reading an image into an electric signal, and an amplification for amplifying the electric signal photoelectrically converted by the photoelectric conversion means. Means, clamp means for performing clamp processing on the electric signal amplified by the amplifying means, and level shift so that the electric signal clamped by the clamp means is adapted to the signal level of the A / D conversion means. And an A / D conversion unit for converting an analog image signal generated by the image reading unit into a digital image signal, the photoelectric conversion unit comprising: The means, the amplifying means, the clamping means and the level shifting means are put in one package. And wherein the door.

Another feature of the present invention is that the photoelectric conversion means, the amplification means, the clamp means and the level shift means are formed on the same wafer.

Another feature of the present invention is a photoelectric conversion means for converting an optical signal obtained at the time of reading an image into an electric signal, and an amplification means for amplifying the electric signal photoelectrically converted by the photoelectric conversion means. , Clamp means for clamping the electric signal amplified by the amplifying means, and level shifting the electric signal clamped by the clamp means so as to match the signal level of the A / D converting means. An analog process composed of two elements, an image reading means in which level shift means is formed on the same wafer, and an A / D converting means for converting an analog image signal generated by the image reading means into a digital image signal. Means for converting the optical signal of the image into a digital electric signal by the analog processing means. The features.

Another feature of the present invention is that the photoelectric conversion means, the amplification means, the clamp means and the level shift means are formed on the same wafer as the image reading means and the A / D conversion means. And are mounted on the same substrate.

Another feature of the present invention is that in an image processing apparatus having image signal processing means for processing an input digital image signal, the bit width of the input signal of the image signal processing means is the bit of the output signal. It is characterized in that it is made larger than the width.

The image reading sensor of the present invention adjusts the level of the photoelectric conversion means for converting an optical signal obtained at the time of reading an image into an electric signal and an analog image signal obtained by photoelectric conversion in the photoelectric conversion means. The analog adjusting means is provided on the same wafer.

[0020]

According to the present invention constructed as described above, the digital image signal at the input end of the shading correction means expresses a number of values far larger than the number of values that the digital image signal at the output end can take. Therefore, all the values represented by the bit width at the output end of the shading correction means can actually be obtained, and when shading correction is performed, the digital image signal at the output end can be represented by the bit width. A true gradation number can be obtained.

According to another feature of the present invention, since the adjusting means for adjusting the value of the digital image signal is provided, the number of possible values of the digital image signal used for normalizing the shading correction is corrected. It is possible to always increase the number of values represented by the bit width of the subsequent digital image signal, and it is not possible to obtain the true number of gradations for the digital image signal normalized by shading correction. Is reliably prevented.

According to another feature of the present invention, since the image reading means is composed of the photoelectric conversion means and the analog adjusting system including the amplification means, the clamp means and the level shift means, only the photoelectric conversion means is used. The number of analog elements used is reduced compared to the conventional image processing device, in which the circuit for reading the image is configured and the analog adjustment system includes the circuit for gain adjustment and offset adjustment. It is possible to reduce the influence of generated noise and temperature change in each element.

Further, according to another feature of the present invention, the photoelectric conversion means, the amplification means, the clamp means and the level shift means are formed on the same wafer, and the image reading means and the A / D constructed as described above are arranged. Since the conversion means and the analog processing means are mounted on the same substrate, the mounting area of the analog circuit is reduced, the number of analog elements used is significantly reduced compared to the conventional one, and noise mixed between the elements is reduced. It is possible to further reduce the influence of temperature change in each element.

According to another feature of the present invention, the bit width of the input signal of the image signal processing means is made larger than the bit width of the image signal processing means.
The accuracy of the image processing is improved, and the output image quality can be improved.

[0025]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of an image processing apparatus of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a main configuration of an image processing apparatus according to an embodiment of the present invention. FIG.
Is an image reading unit, and a photoelectric conversion unit that converts an optical signal obtained at the time of reading an image into an electric signal and an analog image signal obtained by photoelectric conversion in the photoelectric conversion unit are amplified (amplification of several stages). Only the rate can be selected) and the level is adjusted by analog adjusting means.

The photoelectric conversion means has photoelectric conversion elements by a number corresponding to, for example, a plurality of pixels arranged linearly. The analog adjusting unit roughly adjusts the variation of each pixel signal level in the photoelectric converting unit by selecting the amplification factor of several stages and outputs the adjusted signal from the image reading unit.

Although not shown, the analog adjusting means, for example, an amplifying means for amplifying the electric signal photoelectrically converted by the photoelectric converting means, and a clamp process for the electric signal amplified by the amplifying means. It is composed of clamping means for applying and level shift means for level shifting the electric signal clamped by the clamping means.

Is an A / D conversion means, which converts the analog image signal output from the image reading means into a digital image signal. The A / D conversion means has a resolution of 10 bits or 12 bits,
The input analog image signal is converted into a 10-bit or 12-bit digital image signal.

As is apparent from FIG. 1, the photoelectric conversion means and the analog adjustment means which form the image reading means are formed on one wafer. In addition, a circuit for reading an image, converting the image into an analog electric signal, adjusting the level of the analog image signal and performing A / D conversion, that is, an image reading means configured on one wafer, Analog processing means including A / D conversion means is mounted on one board.

As described above, since the image reading means is constructed by providing the photoelectric conversion means and the analog adjusting means having the above-mentioned configuration on the same wafer, these circuits are provided on separate wafers and the gain adjustment is performed. The number of analog elements used can be reduced compared to the conventional image processing device in which a circuit for performing offset adjustment is included in the analog adjustment system, and the effects of noise between elements and temperature changes in each element can be reduced. Can be made smaller.
Therefore, it becomes possible to suppress the deterioration of the quality of the analog image signal.

Further, since the analog processing means is constituted by mounting the image reading means and the A / D converting means on the same wafer on the same substrate, the mounting area of the analog circuit can be further reduced. Since the number of analog elements used can be significantly reduced as compared with the conventional case, the influence of noise mixed between elements and the temperature change in each element can be further reduced.

The shading correction means normalizes the input 10-bit or 12-bit digital image signal to an 8-bit digital image signal. Here, the shading correction is for correcting the sensitivity variation for each pixel of the image reading unit and the illumination unevenness of the illumination light source.

In the correction method, first, a black reference image is read without turning on the illumination light source. Next, the illumination light source is turned on to read the white reference image, and the so-called white correction value is obtained by subtracting the value when the black reference image is read from the value when the white reference image is read. obtain.

When the document is read by the image reading means, the digital image signal of the read document is multiplied by the white correction value to carry out normalization to obtain a shading-corrected digital image signal. That is.

As described above, in this embodiment, the bit width (for example, 10 bits) of the digital image signal at the input end of the shading correction means is larger than the bit width (8 bits) of the digital image signal at the output end. ,
A significantly larger number (1024) of values than the number of values (256) represented by the bit width of the digital image signal at the output end of the shading correction means is obtained at the input end.

Therefore, at the input end of the shading correction means, all values represented by the bit width of the corrected digital image signal can be actually obtained, and the digital image signal at the output end of the shading correction means is
It is possible to obtain the true number of gradations that can be expressed by the bit width. For this reason, it is possible to obtain sufficiently good image quality as compared with the conventional device in which the true number of gradations could not be obtained.

Is an adjusting means for subtracting the value when the black reference image is read from the value when the image reading means reads the white reference image for the digital image signal value at the input end of the shading correction means. The adjusted value is adjusted to be greater than or equal to the maximum value that can be represented by the bit width of the output signal of the shading correction means.

That is, in the case of this example, the adjusting means has a value obtained by subtracting the digital signal value when the black reference image is read from the digital signal value when the white reference image is read, in 8 bits. Adjust so that the maximum value that can be expressed is 255 or more.

As the adjusting method, for example, the value of the digital image signal may be adjusted by adjusting the light amount of the illumination light source used when the image is read by the image reading means, or the value may be output from the image reading means. The value of the digital image signal may be adjusted by adjusting the amplification factor of the analog image signal thus obtained by the analog adjusting means.

As described above, since the adjusting means for adjusting the digital image signal value is provided, the number of values that the digital image signal used for normalizing the shading correction can take is
It is possible to always increase the number of values represented by the bit width (8 bits) of the corrected digital image signal, which is 255, and it is possible to reliably obtain the true number of gradations that can be represented by 8 bits. become able to.

Next, a copying machine which is an embodiment of the image processing apparatus of the present invention will be described with reference to the drawings. First, the contents of image processing in a copying machine which has been generally used conventionally will be described with reference to FIGS. 2 and 3.

In the configuration of the image reading apparatus shown in FIG. 2, there are two areas on the original glass 52, an original reading area for setting an original and an area having a standard white plate 53 used for shading correction. is there. Information on the original placed on the original platen glass 52 and the standard white plate 53 is read as follows.

That is, as shown by the dotted line in FIG. 2, the original illumination lamp 54 is turned on and the light from the original illumination lamp 54 is placed on the original table glass 52.
The original 3 and the light reflected by the standard white plate 53 are guided to the CCD (image reading sensor) 1 through the self-hook lens 55. Then, the light guided to the CCD 1 is C
The CD1 converts an optical signal into an electric signal.

The above-mentioned CCD 1, self-hook lens 55 and original illumination lamp 54 are arranged in the main scanning carriage 51, and the main scanning carriage 51 moves to a position for reading an image in the original reading area, whereby the image of the original is read. Is configured to read.

In the configuration of the image processing block shown in FIG. 3, the image signal converted into an electric signal by the CCD 1 is
It is input to the next analog signal processing circuit 2. In the analog signal processing circuit 2, the input image signal is the A / D of the next stage.
It is amplified to fit the input dynamic range of the converter 3.

Conventionally, as the analog signal processing circuit 2, a circuit using a plurality of operational amplifiers or a dedicated IC is used.
Was used, and it was configured so that any amplification factor could be selected. On the other hand, in this embodiment, as will be described later, only the amplification factor having a certain fixed value can be selected.

The image signal amplified by the analog signal processing circuit 2 is then input to the A / D converter 3. A
The / D converter 3 converts the input analog image signal into a digital image signal. The image signal converted into a digital signal is input to the shading correction circuit 4, where the light amount distribution unevenness of the document illumination lamp 54, CC, and the like.
The variation in the image signal due to the difference in sensitivity between the read pixels D1 is corrected.

Next, the shading-corrected image signal is input to the input masking circuit 5, where the CCD 1
The image signal based on the color space of the color filters (R, G, B) is converted into the image signal based on the standard color space of color. Then, the image signal converted into the color standard color space is scaled by the scaling circuit 6 to a target image size.

The image signal scaled by the scaling circuit 6 is converted by the log conversion circuit (density conversion circuit) 7 into R,
From the G and B luminance signals, C (cyan), M (magenta),
It is converted into a Y (yellow) density signal. The image signal converted into the C, M, and Y density signals is then input to the output masking / UCR (Under Color Removal) circuit 8,
Here, the image signal based on the standard color space of color is converted into the image signal based on the color space in consideration of the printing characteristics of the printer. That is, the density signals of the three colors C, M, and Y are converted into the density signals of the four colors C, M, Y, and K (black plate).

The image signal converted into the density signals of the four colors of C, M, Y and K is converted into the target density by the γ conversion circuit 9. The image signal converted to the target density by the γ conversion circuit 9 is converted by the binarization processing circuit 10 from an 8-bit multivalued signal to a binary signal for a printer. Then, the image signal converted into the binary signal as described above is printed on the paper by the printer 11.

Next, various points improved in the present embodiment with respect to the general copying machine configured as described above will be described in order. First, the operation of the printing means used in the printer 11 according to this embodiment and the image reading means corresponding to the printing means will be described.

Printer 11 used in this embodiment
Is a so-called on-demand type ink jet printer in which a plurality of print nozzles are linearly arranged and an image is formed by ejecting ink from the plurality of print nozzles according to an image to be printed.

That is, in the ink jet printer of this embodiment, as shown in FIG.
The print head unit 12 is equipped with a print head 202 that prints in a line by 512 print nozzles arranged in a straight line with 12 nozzles. Then, by a head drive control means (not shown), the print head portion 12 is moved in the direction of the arrow A in FIG. 4 and printing is sequentially performed in a linear manner to form a band-shaped image. .

The print resolution of the ink jet printer is
It is determined by the arrangement pitch of the nozzles forming the print head 202 and the accuracy of movement in the direction of arrow A in FIG. For example, if the arrangement pitch of the nozzles of the print head 202 is 0.0635 mm, a printer having a resolution of 400 dpi can be configured. Of course, in this case, a movement accuracy of 0.0635 mm pitch is required in the direction of arrow A.

Assuming that the number of nozzles of the print head 202 of this ink jet printer is 512 as described above, the print width of one band is 0.0635 × 512 = 3.
It becomes 2.512 mm. Therefore, when the printing of one band is completed, the printing paper is moved to the direction of arrow B.
The paper is fed by 512 mm and the next band is printed.
The inkjet printer prints in a desired print range by repeating the above print control.

Needless to say, the optimum image reading method used in the image reading apparatus for such an ink jet printer (see FIG. 2; hereinafter, this image reading apparatus is referred to as a scanner) is a band corresponding to the band printing of the printer. It is a reading method.

That is, in this band reading system,
As shown in FIG. 5, a desired image is obtained by a scanner including a reading unit 212 in which a CCD (image reading sensor) 1 including 512 pixels linearly arranged from a first pixel to a 512th pixel is mounted. Read.

Similar to the print head 202 shown in FIG. 4, this scanner has a pixel array pitch of 0.0635 m.
A scanner having a resolution of 400 dpi is configured by setting the moving pitch of the reading unit 212 in the direction of arrow A to 0.0635 mm.

Next, the structure of the image reading sensor used in this embodiment will be described. This image reading sensor has a different structure from the generally known CCD image sensor (CCD 1 shown in FIG. 3). That is, the image reading sensor of this embodiment is developed in-house, and hereinafter, this will be referred to as a color image reading sensor 201 to distinguish it from the conventional CCD 1.

The color image reading sensor 201 is
It is configured as shown in FIG. 6, and the original image illuminated by the original illumination system shown in FIG. 2 is color-separated into analog image signals of three colors of R, G, B, and the color-separated analog image signal is obtained. Amplify and output.

In FIG. 6, a portion surrounded by a thick line is a color image reading sensor 201 constructed on one base (for example, semiconductor silicon). The color image reading sensor 201 is mounted on the image reading board 252. The output signal of the color image reading sensor 201 is supplied to the A / D converter 203 which is also mounted on the image reading substrate 252, and is converted into, for example, a 10-bit digital image signal.

Here, the color image reading sensor 2
The details of 01 will be described. In 201-1, 512 pixels of photosensors are arranged in an array. The photo sensor 201-1 is a photoelectric conversion element that converts an optical signal into an electric signal, and includes a photodiode and a photo transistor.

Reference numerals 253 and 254 denote photosensors 20.
1-1 is a group of switches for sequentially taking out the electric signal generated in 1-1 for each pixel.
The number corresponding to 1-1 is provided.

The opening / closing timing of each switch of the switch groups 253 and 254 is controlled by the clock controller 255. That is, the clock control unit 255 turns on each switch of the switch group 253 at the same time by a periodic signal from the outside, and after a specified time, the switch group 25 is turned on.
Turn off each switch of 3 at the same time. After that, each switch of the switch group 254 is controlled to be sequentially turned on / off according to the external clock.

The switch group 253 includes the photo sensor 2
It has a function of transferring the electric charges generated in 01-1 at a certain time to a charge holding unit 256 composed of, for example, a plurality of capacitors. The capacitors forming the charge holding unit 256 are provided in a one-to-one correspondence with the pixels of the photo sensor 201-1.

As shown in FIG. 7, the switch group 253 is turned on for a certain period of time by a pulse φSH having a period of a certain period corresponding to the reading period of one line (when the pulse φSH in FIG. 7 is "Hi"). Turn on). Switch group 2
When each switch of 53 is turned on, the charge accumulated in each pixel of the photo sensor 201-1 is stored in the charge holding unit 2.
The charge is transferred to 56 and is held until the next charge is transferred.

The charges held in the charge holding unit 256 are the pulses φCLK1 to φCLK5 shown in FIG.
The switches of the switch group 254 are sequentially turned on / off in synchronism with 12 to be sequentially output to the amplifier 257. When all the charges accumulated in the charge holding unit 256 are transferred to the amplifier 257 by opening / closing the switch group 254, the charge holding unit 256 is regulated by a reset mechanism (not shown) before the switch group 253 is turned on again. Set to voltage.

As described above, the circuit from the photosensor 201-1 to the switch group 254 shows the structure for converting the optical image signal into the electric image signal, which is referred to as the photoelectric conversion unit 201-3.

The above-mentioned amplifier 257 is the amplifier circuit control unit 2
The amplification factor can be controlled by the control signal output from 59. However, the amplification factor that can be set is not something that can be selected during the term in order to be able to configure it within one wafer, and it is limited to standard, standard + 10%, standard -10%, for example. There is.

The signal level of the image signal amplified by the amplifier 257 is shifted to a specified level by the clamp circuit 258. Then, the clamped image signal is output from the color image reading sensor 201. As described above, the circuits from the amplifier 257 to the clamp circuit 258 show the configuration for adjusting the analog image signal, which will be referred to as the analog adjusting unit 201-2.

The output signal of the color image reading sensor 201 is converted into a digital image signal by the A / D converter 203. As is apparent from FIG. 6, in the case of this embodiment, the output of the color image reading sensor 201 is A /
It is directly connected to the input of the D converter 203.

Next, a 512-pixel line sensor having 400 dpi (the above-mentioned color image reading sensor 201)
The configuration at the wafer stage of will be described. The total length of each light receiving pixel of the line sensor in the unity-magnification imaging system (FIG. 2), which is the simplest optical system configuration, is 32.
It is 512 mm.

However, it is extremely difficult with the present technology to construct this line sensor on a silicon wafer, take out one chip in one cutting process, and manufacture it in one photomask exposure process. Further, even if it becomes technically possible, the price of the line sensor becomes expensive because it is necessary to invest a huge amount of money into the manufacturing equipment.

Therefore, in order to realize a line sensor having a chip length of 32.512 mm, FIG.
5 inch master silicon wafer 231
8 is provided with a plurality of cutout regions 232 of about 20 mm × 20 mm, and FIG.
As shown in (b), a cutout area 23 having a length of about 17 mm
There is no choice but to adopt the method of providing 3,234.

Therefore, the photoelectric conversion unit 2 shown in the configuration of FIG.
01-3 is divided into two. That is, the cutout region 233 of FIG. 8B is a silicon chip in which the first pixel to the 256th pixel are formed on silicon, and the cutout region 234 is formed of the 257th pixel to the 512th pixel in silicon. It is a silicon chip. Further, the clock control unit 255, the amplifier 257, and the clamp circuit 25
8 is divided into two cutout areas 233 and 234 to form one of the cutout areas.

Actually, in the process of forming the line sensor, each silicon chip is not shown so that the 256th pixel of the silicon chip in the cutout region 234 is connected immediately after the 256th pixel of the silicon chip in the cutout region 233. A method of arranging it inside the package and wire-bonding the necessary portion is used.

Another method is to perform one photomask exposure step twice on one silicon chip,
There are possible ways of constructing the line. That is, from the first pixel to the 256th pixel in the first photomask exposure process.
A portion corresponding to the pixel is formed on silicon, and the 257th pixel to the 512th pixel are formed in the second photomask exposure step.
A portion corresponding to a pixel is formed on silicon. And
In this method, an image reading sensor formed on one silicon wafer is arranged in a package.

Next, the difference between the conventional shading correction method and the shading correction method of this embodiment will be described in detail. First, conventional shading correction will be described with reference to FIGS.

FIG. 9 shows an image reading sensor (CCD) 1.
It is a figure showing the composition of the analog processing block from to A / D converter 3. The optical signal from the document is converted into R, G, B image signals by the CCD 1 and output. CC
The R, G, and B image signals output from D1 are subjected to predetermined analog processing by the analog processing circuits 101 to 103 for the respective colors, and are supplied to the selector 104.

Then, by the selector 104, R, G,
By sequentially selecting the B signals, the input R, G, B parallel image signals are converted into R, G, B serial image signals, which are supplied to the A / D converter 3. It The A / D converter 3 converts the R, G, B analog image signals into 8-bit digital image signals.

The processing contents of the analog processing circuits 101 to 103 are as shown in FIG. That is, CC
The output signal of D1 is first supplied to the S / H (sample and hold) circuit 106. The S / H circuit 106 samples the analog image signal for a fixed period, and
The image signal is sampled by holding the sampled image signal for a certain period.

The image signal sampled by the S / H circuit 106 is amplified 3.8 times by the amplifier 107. The magnitude (amplification factor) of the amplified signal is adjusted by the next gain adjusting circuit 108. The gain adjusting circuit 108 has a function of setting an arbitrary amplification factor of 0 to 1 times, and 0.5 times is a standard value.

Here, the adjustment value in the gain adjusting circuit 108 will be described. FIG. 11A shows an example of the output signal of the CCD 1 when the original illumination lamp 54 of FIG. 2 is turned on and the standard white plate 53 is read. Originally, when reading an ideal white subject, R, G, B
Each signal level in should be the same. However, due to the deviation of the spectral characteristics of the standard white plate 53 and the original illumination lamp 54 from the ideal shape, the influence of the spectral characteristics of the R, G, B optical filters used in the CCD 1, and the like, R, G, B
The respective signal levels of may not match.

Therefore, in the output of CCD 1, R, G,
The analog processing circuit 1 calculates the difference between the B signal levels.
The gain adjustment circuit 108 for each color in 01 to 103 makes the R, G, and B signal levels match at the input of the A / D converter 3. FIG. 11B shows an input signal of the A / D converter 3 when the original illumination lamp 54 is turned on and the standard white plate 53 is read. As is clear from this figure, in the input stage of the A / D converter 3, the R, G, and B signal levels match.

The signal thus gain-adjusted is
Clamp processing is performed by the next clamp circuit 109. The clamp process operates so as to keep the optical black signal level included in the image signal constant. The clamped analog image signal is then doubled by the amplifier 110 and supplied to the offset adjustment circuit 111. In the offset adjustment circuit 111,
The analog image signal is output after being level-shifted by a predetermined amount.

Here, the adjustment value in the offset adjusting circuit 111 will be described. FIG. 12A shows an example of the output signal of the CCD 1 in the case where the original illumination lamp 54 is turned off so that no light enters the CCD 1. In this case, ideally, the R, G, and B signal levels should be the same. However, in an actual circuit, the R, G, and B signal levels may not match due to variations in the elements used in the CCD 1. In addition, the influence of variations in the elements used in the analog processing circuits 101 to 103 is also added.

Therefore, in the output of CCD 1, R, G,
The analog processing circuit 1 calculates the difference between the B signal levels.
01-103 offset adjustment circuit 111 for each color
A / D converter 3 by adjusting the offset value with
The signal levels of R, G, and B are matched at the input of. FIG. 12B shows A in the case where the original illumination lamp 54 is turned off so that the CCD 1 is not exposed to light.
The input signal of the / D converter 3 is shown. As is clear from this figure, in the input stage of the A / D converter 3, R, G, B
It can be seen that the respective signal levels of are in agreement.

Further, the offset adjusting circuit 111 is
It also has a function of adjusting the level of the image signal to the input range (dynamic range) of the D converter 3. That is,
The A / D converter 3 is used when the input voltage range is 2 to 4V. Therefore, the offset adjustment circuit 111
Is an input range of the A / D converter 3 while eliminating variations in the signal levels of R, G, and B when the original illumination lamp 54 is turned off and no light enters the CCD 1.
The offset value is determined in consideration of setting it to V or more.

Similarly, the gain adjusting circuit 108
It also has a function of adjusting the level of the image signal to the input range of the A / D converter 3. That is, the document illumination lamp 54
Is turned on and the standard white plate 53 is read,
In addition to eliminating variations in the R, G, and B signal levels, R, G, and B within the input range of 4 V, which is the input range of the A / D converter 3,
The gain is determined in consideration of the maximum values of the G and B signals. That is, FIGS. 11B and 12B show signal levels when the gain adjustment and the offset adjustment are performed, respectively.

As described above, the output signal of the CCD 1 is the analog processing circuits 101 to 1 having the configuration shown in FIG.
After being subjected to a plurality of stages of analog processing in 03, it is output to the selector 104 of the next stage.

FIG. 13 shows an example of conventional shading correction. 13A is a block diagram showing the configuration of the shading correction circuit 4 for performing the above shading correction, FIG. 13B is a diagram showing an example of the signal level at each stage, and FIG. FIG. 9 is a diagram showing correction coefficients used when performing white correction. Note that FIG.
In (A), the image processing circuit 114 includes circuits after the input masking circuit 5 shown in FIG.

In FIG. 13A, the analog image signals of R, G and B are 8 bits (0 to 0) by the A / D converter 3.
It is quantized into a digital image signal of 255 levels). The following processing in the black correction circuit 112, the white correction circuit 113, and the image processing circuit 114 will be described based on the signal level shown in FIG.

Here, in the output of the A / D converter 3, the original illumination lamp 54 is turned on to turn on the standard white plate 53.
The following description will be made assuming that the signal value (hereinafter, referred to as a white signal level) when reading the document is 210, and the signal value (hereinafter, referred to as a black signal level) when the document illumination lamp 54 is turned off is read.

First, the black correction circuit 112 subtracts the correction value 10 for black correction from the digital image signal. As a result, in the output of the black correction circuit 112, the white signal level becomes 200 and the black signal level becomes 0. Next, in the white correction circuit 113, the correction coefficient shown in FIG. 13C is applied to the black-corrected digital image signal (the slope of the coefficient is 255).
/ 200). As a result, the white signal level is 25
5, the black signal level becomes 0.

As described above, the shading correction is performed for each pixel of the CCD 1 (coefficients for black correction and white correction are determined for each pixel), and the original illumination lamp 54 is turned on to read the standard white plate 53. When the signal value is 255,
The signal value when the original illumination lamp 54 is turned off and read is normalized to 0. The normalized image signal is subjected to predetermined image processing in the image processing circuit 114.

As described above, by shading-correcting the digital image signal, the 8-bit signal is normalized, and a signal having 256 gradations (8 bits) can be obtained.

However, the signal width before shading correction 1
When a 200-tone signal of 0 to 210 is normalized to 256 tones of a signal width of 0 to 255 by performing shading correction, a signal value of 0 and a signal value of 255 are obtained, but a signal value of 1 to 254 is 198. Only one signal value actually exists, and the signal value is in a missing state.

That is, even if the digital image signal is normalized to 8 bits (256 gradations) by shading correction, it is obvious that the actual value of the normalized digital image signal is actually only 200 gradations. is there.

Next, the shading correction according to this embodiment will be described with reference to FIGS. FIG.
From the color image reading sensor 201 of this embodiment
It is a figure showing the composition of the analog processing block to the / D converter 203.

In FIG. 14, the photoelectric conversion unit 201-3
Is for converting an optical signal into an electrical signal. The optical signal from the original is converted into R, G, and B image signals by the photoelectric conversion unit 201-3 and output. Photoelectric conversion unit 201-
The R, G, and B image signals output from No. 3 are subjected to analog processing by the analog processing circuits 118 to 120 for the respective colors and supplied to the selector 104.

Then, the selector 104 selects R, G,
By sequentially selecting the B signals, the input R, G, B parallel image signals are converted into R, G, B serial image signals, which are supplied to the A / D converter 203. It The A / D converter 203 converts the R, G, B analog image signals into, for example, 10-bit digital image signals.

Thus, in this embodiment, the A / D
The output of the converter 203 is 10 bits. The difference between the conventional A / D converter 3 and the A / D converter 203 of the present embodiment is that the output of the A / D converter 3 is 8 bits and the A / D converter 2 is
The output of 03 is 10 bits, which means that the output bit width is different.

The processing contents of the analog processing circuits 118 to 120 are as shown in FIG. That is, the output signal of the photoelectric conversion unit 201-3 is first supplied to the S / H (sample and hold) circuit 106. S / H circuit 1
Reference numeral 06 samples the analog image signal for a certain period and holds the sampled image signal for a certain period to sample the image signal.

The image signal sampled by the S / H circuit 106 is amplified 3.8 times by the amplifier 257. The amplified signal is used in the next clamp circuit 1
The clamp processing is performed according to 09. The clamp process operates so as to keep the optical black signal level included in the image signal constant. The clamp circuit 109 is the clamp circuit 258 shown in FIG.
Is the same as

The analog image signal clamped by the clamp circuit 109 is transferred to the next level shift circuit 1.
17 is supplied. In the level shift 117, the input analog image signal is level-shifted so as to match the input level of the A / D converter 203. Then, the image signal thus level-shifted is sent to the selector 10 of the next stage.
4 is output.

As described above, the photoelectric conversion unit 201-
The output signal of No. 3 is subjected to a plurality of stages of analog processing in the analog processing circuits 118 to 120 having the configuration shown in FIG. 15 and output to the selector 104 of the next stage.

Next, in the block diagram shown in FIG. 14, A when the image is read by the photoelectric conversion unit 201-3
An example of the image signal at the input terminal of the / D converter 203 is
This will be described with reference to FIG.

FIG. 16A shows the input signal level of the A / D converter 203 when the original illumination lamp 54 is turned on and the standard white plate 53 is read. That is, the output signal of the photoelectric conversion unit 201-3 shown in FIG. 14 is amplified 3.8 times and the level thereof is shifted by a predetermined amount.

FIG. 16B shows the original illumination lamp 5
4 shows the input signal level of the A / D converter 203 when 4 is turned off and no light enters the photoelectric conversion unit 201-3. That is, the output signal of the photoelectric conversion unit 201-3 shown in FIG. 14 is amplified 3.8 times and the level thereof is shifted by a predetermined amount.

As is apparent from comparison between FIG. 16 (A) and FIG. 11 (B) and between FIG. 16 (B) and FIG. 12 (B), respectively, the conventional A / D conversion shown in FIG. At the input terminal of the device 3, when achromatic color (white, black) information is read, the R, G, and B signals have the same level. On the other hand, the A / D converter 203 of this embodiment shown in FIG.
At the input terminal of, the R, G, and B signal levels are different even when the information of achromatic color (white, black) is read.

However, the difference between the R, G, and B signal levels is very small in an actual device, and according to the shading correction of this embodiment, the level difference can be substantially corrected. It doesn't matter. FIG.
As described above, the amplifier 257 of No. 2 is configured so that the amplification factor can be selected from three types of standard value and standard value ± 10% by selecting the resistance value, and R, G, and B signals can be selected. It is possible to make small adjustments independently.

Here, the conventional analog processing block shown in FIG. 10 is compared with the analog processing block of the present embodiment shown in FIG. As is clear from both figures,
The same S / H circuit 106 and clamp circuit 109 are used.

Further, the amplifier 10 used in FIG.
7 (amplification factor 3.8 times), amplifier 110 (amplification factor 2 times), and amplifier 257 used in FIG. 15 (amplification factor 3.8 times in the figure, but amplifier 107 is an amplification factor). (The rates may be different) are all amplifiers of the same configuration. Therefore, in the analog processing block of the present embodiment shown in FIG. 15, the number of amplifiers is one step less than in the conventional case, and the mounting area is small.

The offset adjustment circuit 111 used in FIG. 10 and the level shift circuit 117 used in FIG. 15 change the level (offset) of the image signal to change the level of the image signal. Has the same function in that is adjusted to the input level of the A / D converter. However, the offset adjustment circuit 111 and the level shift circuit 117 are different in the following points.

That is, the level shift circuit 117 is
Since the process of eliminating the difference between the R, G, and B signal levels is not performed, the image signal may be shifted by the voltage value determined in advance. Therefore, the voltage value to be shifted may be fixed.

On the other hand, the offset adjusting circuit 111
Eliminates the difference between the R, G, and B signal levels. Therefore, a circuit for selecting a voltage value to be shifted for each R, G, B signal and a circuit for setting or storing each R, G, B voltage value are required. Therefore,
The offset adjustment circuit 111 includes a level shift circuit 117.
The configuration is more complicated than that of, and the number of elements used is significantly increased.

In addition to the circuit for amplifying the level of the image signal, the gain adjusting circuit 108 used only in FIG. 10 has R, G, and B to eliminate the difference between the signal levels. A circuit for selecting the gain for each of G and B,
A circuit for setting or storing the R, G, and B gains is required.

As described above, the configuration of the analog processing block according to the present embodiment shown in FIG. 15 does not have a gain adjusting circuit and the amplifier has one stage as compared with the configuration of the conventional analog processing block shown in FIG. Few. Further, the structure of the level shift (offset adjustment) circuit is simplified.

As a result, in the present embodiment, the number of stages of blocks for processing an analog image signal can be markedly reduced as compared with the conventional case, and the influence of external noise and variations in analog elements can be reduced to analog. It is possible to make it difficult for the image signal to be received. This improves the quality of the analog image signal.

Further, as the number of elements is reduced, the cost of the elements is reduced, and the area of the printed board on which the elements are mounted can be reduced, so that the cost of the printed board is reduced. In addition, the reduced number of elements reduces the power (current) consumed by the analog processing circuit, simplifying the power supply circuit for supplying power to the analog processing circuit and generating it in the power supply circuit. The heat generated can be reduced, and the power supply circuit can be downsized and the cost can be reduced.

Therefore, by adopting the configuration of the analog processing block shown in FIG. 15, it is possible to simultaneously reduce the number of elements of the analog signal processing circuit and the heat generated, and as a result, FIG. The illustrated photoelectric conversion unit 20
It becomes possible to form 1-3 and the analog adjustment part 201-2 on the same silicon wafer.

FIG. 17 shows an example of shading correction of this embodiment. FIG. 17A is a block diagram showing the configuration of a shading correction circuit that performs the above shading correction, FIG. 17B is a diagram showing an example of the signal level at each stage, and FIG. FIG. 7 is a diagram showing correction coefficients used when performing white correction.

The signal waveforms of R, G and B as shown in FIGS. 16A and 16B are inputted to the input terminals of the A / D converter 203 shown in the block diagram of FIG. 17A. . The analog image signal input to the A / D converter 203 is quantized by the A / D converter 203 into a 10-bit (0-1023 level) digital image signal. Black correction circuit 1 below
21, the processing in the white correction circuit 122 and the image processing circuit 114 will be described based on the signal levels shown in FIG.

Here, in the output of the A / D converter 203, the white signal level when the original illumination lamp 54 is turned on and the standard white plate 53 is read is 840, and the original illumination lamp 54 is turned off to read. The black signal level when it is set to 40.

First, the black correction circuit 121 subtracts the correction value 40 (predetermined in advance) for black correction from the digital image signal. As a result, in the output of the black correction circuit 121, the white signal level becomes 800 and the black signal level becomes 0. Next, in the white correction circuit 122, the correction coefficient shown in FIG. 17C (determined in advance, the slope of the coefficient is 255 for the black-corrected digital image signal.
/ 800). As a result, the white signal level is 25
5, the black signal level becomes 0.

As described above, by performing the shading correction for each pixel of the photoelectric conversion unit 201-3 (the coefficients of black correction and white correction are determined in advance for each pixel), the original illumination lamp 54 is turned on and the standard illumination lamp 54 is turned on. The signal value when the white plate 53 is read is normalized to 255, and the signal value when the document illumination lamp 54 is turned off and read is normalized to 0.
The output of 22 is an 8-bit digital image signal. The normalized digital image signal is the image processing circuit 1
At 14, predetermined image processing is performed.

As described above, by shading-correcting the digital image signal, the 8-bit signal is normalized so that a signal having 256 gradations (8 bits) can be obtained. That is, in the present embodiment, FIG.
In the shading correction shown in the above example, a signal of 800 gradation with a signal width of 40 to 840 before shading correction
By performing shading correction, 2 of signal width 0-255
It is normalized to a 56-tone signal. As a result, signal width 0
With a value of 255, it is possible to obtain a digital image signal having a true 256-tone that does not cause a missing tooth state in the middle of the signal as in the conventional shading correction shown in FIG.

Although the present embodiment has been described using the A / D converter 203 having a 10-bit resolution, the A / D
The same effect can be obtained as long as the number of bits of the converter is larger than the number of bits output from the shading correction circuit.

The A / D converter that can be used for image reading by the reader described in this embodiment is currently CXD2.
There are 310R (10-bit output manufactured by Sony), AD875 (10-bit output manufactured by analog device), MP8791 (12-bit output manufactured by Micro Power System), and the like.

In the future, 10-bit output or 12 will be output.
It is expected that the number of manufacturers producing A / D converters with bit outputs will further increase. Also, at the present time, A / output of 12-bit output or more that can be used for reader image reading
It seems that the D converter product is not on sale, but it will be commercialized soon. It goes without saying that these A / D converters can also be used in the method described in this embodiment.

Next, the shading correction coefficient (black correction coefficient, white correction coefficient) is determined using the explanatory diagram of the shading correction value input shown in FIG. 18 and the flow chart of the shading correction value input shown in FIG. The necessary processing will be described.

In FIG. 19, first, in step S1, initialization of each unit shown in the block diagram of FIG. 18A is performed. That is, the value of the counter 123 is set to 0.
In addition, the output of the counter 123 is stored in the two memories 125-
The selection of the first selector 124 is switched so as to connect to the address bus of No. 1 and 125-2, and the output of the A / D converter 203 is changed to the above two memories 125-1 and 12-2.
The second selector 1 so as to be connected to the data bus 5-2
The selection of 26 is switched.

In the next step S2, black data is input. When inputting the black data, the original illumination lamp 54 is turned off and the image is read. The analog image signal obtained by this reading is converted into a 10-bit digital image signal by the A / D converter 203, and the two memories 125-1 and 125 are passed through the second selector 126.
-2.

In the memories 125-1 and 125-2, the data bus is composed of two SRAMs having a width of 8 bits, and the upper 2 bits of the 10-bit digital image signal are stored in the memory 125-2 and the lower 8 bits. Bit is memory 125-
1 is stored in each.

As described above, each memory 125-
The address buses 1 and 125-2 are connected to the output terminals of the counter 123 via the selector 124. The digital image signal (black data) described above is stored in the addresses of the memories 125-1 and 125-2 corresponding to the output value of the counter 123.

The counter 123 operates as shown in the timing chart of FIG. 18 (B). That is, when the output of the A / D converter 203 is R1 (first pixel of red), the output of the counter 123 is 0, when the output of G1 (first pixel of green) is 1, the output of counter 123 is B1. The output of the counter 123 at the time of (blue first pixel) is 2, the output of the counter 123 at X1 (the first pixel of indefinite data) is 3, and the output at the time of R2 (second pixel of red). The output of the counter 123 is 4, and so on, as the image reading position changes.
It operates so that the output of 3 increases by one.

When the counter 123 counts up to the value 2223 corresponding to one line of the color image reading sensor 201, the count-up operation ends. This completes the writing of the image signal to each of the memories 125-1 and 125-2. As a result, one line of black data is stored in each of the memories 125-1 and 125-2.

In the next step S3, the black correction value is calculated. That is, by switching the selection of the first selector 124 and the second selector 126, the address bus and data bus of each of the memories 125-1 and 125-2 are connected to the CPU 127. As a result, the black data stored in each of the memories 125-1 and 125-2 is CP
It is read by U127.

A memory (not shown) managed by the CPU 127 manages the black data read by the CPU 127.
Is stored as the black correction value in step S4 and the process proceeds to step S4. As described above, in the present embodiment, the read black data is used as it is as the black correction value, but for example, in order to remove the noise component of the black data, the black data is read a plurality of times and the average thereof is taken. In some cases, the arithmetic processing of is performed.

In step S4, the settings necessary for obtaining the white correction coefficient are set. That is, the counter 12
Set the value of 3 to 0. Further, the selection of the first selector 124 is switched so that the output of the counter 123 is connected to the address buses of the memories 125-1 and 125-2, and the output of the A / D converter 203 is changed to the memory 125-.
The selection of the second selector 126 is switched so as to connect to the data buses 1 and 125-2.

In the next step S5, white data is input. When inputting white data, the standard white board 53
The color image reading sensor 201 is moved to the position for reading, and the original illumination lamp 54 is turned on to read the image. The white data obtained by this reading is stored in each of the memories 125-1 and 125 similarly to the processing of step S2.
-2. Then, the process proceeds to the next step S6.

In step S6, the white correction value is calculated as follows. First, by switching the selection between the first selector 124 and the second selector 126, the address bus and data bus of each of the memories 125-1 and 125-2 are connected to the CPU 127. As a result, the CPU 127 reads the white data stored in each of the memories 125-1 and 125-2.

Then, the white data read by the CPU 127 is a memory (not shown) managed by the CPU 127.
To memorize. The black correction value obtained in step S3 is subtracted from the white data stored in this memory.
It is processed as white data that has undergone black correction processing.

Secondly, white correction data (white correction coefficient) is created based on the white data which has been subjected to the black correction processing. As an example, the value of white data after black correction processing is 8
The case of 00 will be described. As described above, in this embodiment, the maximum value of the data after white correction is set to 255. That is, the result of multiplying the white data value 800 after black correction processing by the coefficient should be 255, so the white correction coefficient is 0.31875 (255 /
800).

That is, by multiplying the digital image signal read by using the pixel whose black data value after black correction processing is 800 to the white correction coefficient 0.31875, the digital image signal can be normalized. .

Similarly, the white correction value (coefficient) is obtained by executing the processing for obtaining the white correction coefficient for the other pixels, and the white correction value (coefficient) is stored in the memory (not shown) managed by the CPU 127. Then, the process proceeds to the next step S7, and the input process of the shading correction value is ended.

Next, how to use the shading correction values (black correction value and white correction value) determined as described above will be described with reference to the block diagram of the shading correction circuit shown in FIG.

In FIG. 20, a RAM 128 is a RAM that stores black correction values. The address 0 of the RAM 128 has the black correction value of the pixel R1 and the address 1 has the pixel G1
, The black correction value of pixel B1 at address 2, the black correction value of pixel R2 at address 4, and the pixel G at address 5.
The black correction value of 2, the black correction value of the pixel B2 at the address 6, the black correction value of the pixel R3 at the address 8, and so on are written in order according to the output timing of the pixel of the sensor. ing.

The RAM 129 is a RAM that stores white correction values. The address 0 of the RAM 129 has the white correction value of the pixel R1, the address 1 has the white correction value of the pixel G1, the address 2 has the white correction value of the pixel B1, the address 4 has the white correction value of the pixel R2, and the address 5 has The white correction value for pixel G2, the white correction value for pixel B2 at address 6, and the address 8
, The white correction value of the pixel R3 is written in order according to the output timing of the pixel of the sensor.

Note that here, RA for storing the black correction value
Although M and the RAM for storing the white correction value are separately described for convenience of description, these are the memory 125-1 and the memory 125-1 in FIG.
This corresponds to 125-2.

For example, when the data (10 bits) of the pixel R1 of the color image reading sensor 201 is output as the output of the A / D converter 203, the count value 0 is output from the counter 123. As a result, the black correction value for the pixel R1 is output from the RAM 128, and the white correction value for the pixel R1 is output from the RAM 129.

As a result, the subtractor 134 causes the pixel R1
The shading correction is performed by subtracting the black correction value for the pixel R1 from the data value of 1. and multiplying the black-corrected data value of the pixel R1 by the white correction value for the pixel R1 by the multiplier 135. This allows
An 8-bit image signal that is shading-corrected for the pixel R1 is obtained.

Other pixels (G1, B1, R2, G2, B
2, R3, ...), the same processing is performed to obtain an 8-bit image signal that is shading-corrected for each pixel.

The shading correction circuit may be constructed as shown in the block diagram of FIG. 21, and the shading correction values (black correction value and white correction value) may be used as described below.

In FIG. 21, the R register 131 is
This is a 10-bit wide register that stores the black correction value of red R. The value stored in the R register 131 is the average value of the black correction values of the pixels R1 to R556.

The G register 132 is a green G register.
It is a 10-bit wide register that stores the black correction value of. The value set in the G register 131 is the pixel G
It is the average value of the black correction values of 1 to pixel G556.

The B register 133 is a 10-bit wide register that stores the black B black correction value. The value set in the B register 133 is the pixel B
It is the average value of the black correction values of 1 to pixel B556.

The selector 130 includes an R register 131,
One of the black correction values stored in the G register 132 and the B register 133 is selected. The selector 130 is configured to select a black correction value corresponding to the color of the data output from the A / D converter 203.

That is, the selector 130 selects the R register 131 when the R signal is output from the A / D converter 203, selects the G register 132 when the G signal is output, and selects the B signal. Is output, the B register 133 is selected.

Therefore, for example, when the image signal of the pixel R1 (10 bits) is output from the A / D converter 203, the selector 130 selects the R register 131 and outputs the red R black correction value. To be done. This allows
The black correction value for red R is subtracted from the data value of the pixel R1 by the subtractor 134.

Next, the image signal subjected to the black correction is R
It is input to the OM 136, and white correction processing is performed here. That is, in this embodiment, the white correction is performed using the ROM table conversion technique.

The ROM 136 stores conversion data (FIG. 17) for correcting variations in assumed image sensor characteristics.
(Corresponding correction coefficient shown in (C)) is written in advance. Further, for example, the sensor of the pixel R1 is shown in FIG.
Assuming that the characteristic is as shown in (C), the selection signal for selecting the conversion table having the characteristic as shown in FIG. 17C is written in the address corresponding to the pixel R1 of the RAM 137.

At the timing when the A / D converter 203 outputs the image signal of the pixel R1, a selection signal for selecting the conversion table corresponding to FIG. 17C in the ROM 136 is output from the RAM 137. , That is ROM
It is input to the upper address of 136. On the other hand, the image signal of the pixel R1 subjected to black correction by the subtractor 134 is input to the lower address of the ROM 136.

The image signal of the pixel R1 input to the lower address is converted based on the conversion table of FIG. 17C and output from the ROM 136. As a result, an 8-bit digital image signal that is shading-corrected for the pixel R1 is obtained.

Next, with reference to the flow chart of the image signal adjustment shown in FIG. 22, the processing necessary for adjusting the light source for illuminating the original will be described. First, in step S11, the processes of steps S1 to S5 shown in the flowchart of FIG. 19 are performed. As a result, the black correction value is obtained and the image signal (white data) obtained by reading the standard white plate 53 is input. Next, in step S12, the above step S1
The black correction value is subtracted from the white data input in 1, and the process proceeds to the next step S13.

In step S13, it is determined whether or not the value obtained by subtracting the black correction value from the white data is within the range of all pixels 0-254. If the value obtained by subtracting the black correction value from the white data is within the range of 0 to 254, the image signal value when the standard white plate 53 is read at the input end of the A / D converter 203 is the target. It is determined that the value is less than the value, and the process proceeds to the next step S14.

At step S14, the A / D converter 203
At the input end of, the image signal value when the standard white plate 53 is read is adjusted to be larger than the target value. That is, the light amount of the document illumination lamp 54 that illuminates the standard white plate 53 is adjusted (the document illumination lamp 5
Increase the voltage applied to 4).

When the standard white plate 53 is read at the input end of the A / D converter 203 by selecting the gain so that the gain of the amplifier 257 of the analog processing block shown in FIG. 15 increases. The image signal value may be larger than the target value.

After the image signal value when the standard white plate 53 is read is adjusted to be larger than the target value as described above, the process returns to step S12, and the same process is repeated.

On the other hand, if the value obtained by subtracting the black correction value from the white data is 255 or more, it is determined that a true 256 gradation can be obtained when the shading correction is performed, and step S1
The process proceeds from step S3 to step S15, and the light amount adjustment process ends.

Although the shading correction has been described as an example in the present embodiment, the present invention is not limited to this, and the input bit width may be made larger than the output bit width in other signal processing. With this configuration, the output image quality can be improved.

[0172]

As described in detail above, according to the present invention, the bit width of the input signal of the shading correction means (for example, 10 or 12 bits) is the bit width of the output signal of the shading correction means (for example, 8 bits). bit)
Since it becomes larger, it becomes possible to obtain a true gradation number (for example, 256 gradations) in the output of the shading correction means. As a result, the number of gradations that can be used in the image processing circuit after the shading correction means is increased, so that the accuracy of the image processing is improved and the output image quality can be improved.

According to another feature of the present invention, since the adjusting means for adjusting the value of the digital image signal is provided, the number of possible values of the digital image signal used for normalizing the shading correction is corrected. Can always be more than the number of values represented by the bit width of the digital image signal of
The conventional inconvenience that the true number of gradations cannot be obtained in the output of the shading correction means can be surely prevented, and the output image quality can be surely improved.

According to another feature of the present invention, the image reading means is composed of the photoelectric converting means and the analog adjusting means including the amplifying means, the clamping means and the level shifting means. It is possible to significantly reduce the number of analog elements used compared to.
As a result, it is possible to reduce the influence of noise mixed between each element of the analog circuit and the temperature change in each element, improve the quality of the analog image signal, and further improve the output image quality. became.

Further, according to another feature of the present invention, the photoelectric conversion means, the amplification means, the clamp means and the level shift means are formed in the same package or on the same wafer, and the image thus formed is formed. Reading means and A /
Since the analog processing means is configured by mounting the D conversion means on the same substrate, the cost of the element is reduced by reducing the number of elements used in the analog circuit and the cost of the board is reduced due to the reduction of the board area for mounting the element. Can be realized at the same time. Furthermore, by reducing the mounting area of the analog circuit, it is possible to further reduce the amount of external noise that is mixed into the analog circuit, and because the number of stages of analog elements is reduced, the heat generation of the analog circuit is reduced. It is possible to reduce the influence of temperature drift. As a result, the quality of the analog signal related to the image is improved, and the output image quality can be improved.

According to another feature of the present invention, the bit width of the input signal of the image signal processing means is made larger than the bit width of the image signal processing means.
The accuracy of image processing is improved, and the output image quality can be improved.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a main configuration of an image processing apparatus according to an embodiment of the present invention.

FIG. 2 is a diagram showing a configuration of an image reading apparatus.

FIG. 3 is a diagram showing a configuration of an image processing block generally used in a copying machine.

FIG. 4 is a diagram showing a configuration of a print nozzle and a moving direction thereof.

FIG. 5 is a diagram showing a configuration of an image reading sensor and a moving direction thereof.

FIG. 6 is a diagram showing a configuration of an analog processing unit (color image reading sensor and A / D converter) adopted in this embodiment.

FIG. 7 is a timing chart showing the timing of each clock for controlling the image reading unit generated in the clock control unit.

FIG. 8 is a diagram for explaining a chip configuration of a line sensor on a silicon wafer.

FIG. 9 is a diagram showing a configuration of a conventional analog processing block.

10 is a diagram showing a configuration of each analog processing circuit shown in FIG.

11 is a characteristic diagram for explaining an example of gain adjustment in the gain adjustment circuit shown in FIG.

12 is a characteristic diagram for explaining an example of offset adjustment in the offset adjustment circuit shown in FIG.

FIG. 13 is a diagram for explaining an example of conventional shading correction.

FIG. 14 is a diagram showing a configuration of an analog processing block according to the present embodiment.

15 is a diagram showing a configuration of each analog processing circuit shown in FIG.

16 is a characteristic diagram showing an example of a signal input to the A / D converter shown in FIG.

FIG. 17 is a diagram for explaining an example of shading correction according to the present embodiment.

FIG. 18 is a diagram for explaining an input operation of shading correction values (black correction value and white correction value) used in the present embodiment.

FIG. 19 is a flowchart showing an operation of inputting a shading correction value used in this embodiment.

FIG. 20 is a block diagram showing a configuration example of a shading correction circuit according to the present embodiment.

FIG. 21 is a block diagram showing another configuration example of the shading correction circuit according to the present embodiment.

FIG. 22 is a flowchart showing the contents of image signal adjustment processing when a standard white plate is read.

[Description of Reference Signs] Image reading means Photoelectric conversion means Analog adjustment means A / D conversion means Analog processing means Shading correction means Adjustment means Illumination light source 106 Sample / hold circuit 109 Clamp circuit 117 Level shift circuit 118 R signal analog processing circuit 119 G Signal analog processing circuit 120 B Signal analog processing circuit 121 Black correction circuit 122 White correction circuit 123 Counter 124 Selector 125-1, 125-2, 128, 129, 137 R
AM 127 CPU 131 Register for R 132 Register for G 133 Register for B 134 Subtractor 135 Multiplier 136 ROM 201 Color image reading sensor 201-1 Photosensor 201-2 Analog adjustment unit 201-3 Photoelectric conversion unit 203 A / D conversion Device 212 Read unit equipped with color image reading sensor 231 Master silicon wafers 232, 233, 234 Cutout region 252 Read substrate 253 Switch group 254 Switch group 255 Clock control unit 256 Charge holding unit 257 Amplifier (amplifier) 258 Clamp circuit 259 Amplification circuit Control unit

Claims (16)

[Claims] 1. An image processing apparatus having shading correction means for shading-correcting an input digital image signal, wherein a bit width of an input signal of the shading correction means is larger than a bit width of an output signal of the shading correction means. An image processing device characterized in that 2. An image reading unit for reading an image to generate an analog image signal, an A / D converting unit for converting an analog image signal output from the image reading unit into a digital image signal, and the A / D converting unit. And a shading correction means for shading-correcting the digital image signal output by the A / D conversion means, wherein the A / D conversion means converts the analog image signal into a bit width of the output signal of the shading correction means. An image processing apparatus, which converts a digital image signal having a large bit width. 3. The image processing device according to claim 1, wherein the bit width of the input signal of the shading correction means is 10 bits, and the bit width of the output signal of the shading correction means is 8 bits. . 4. The image processing apparatus according to claim 1, wherein a bit width of an input signal of the shading correction means is 12 bits, and a bit width of an output signal of the shading correction means is 8 bits. . 5. The value of the digital image signal at the input end of the shading correction means is a value obtained by subtracting the value when a black reference image is read from the value when a white reference image is read by the image reading means. 3. The image processing apparatus according to claim 2, further comprising an adjusting unit that adjusts so as to be equal to or larger than a maximum value that can be represented by a bit width of the output signal of the shading correcting unit. 6. The adjustment means subtracts a value when the black reference image is read from a value when the white reference image is read, and is a maximum value of 2 that can be represented by 8 bits.
The image processing apparatus according to claim 5, wherein the image processing apparatus is adjusted to be 55 or more. 7. The adjusting means adjusts the value of the digital image signal by adjusting the amount of light of an illumination system used when the image is read by the image reading means. Image processing device. 8. The image according to claim 5, wherein the adjusting unit adjusts the value of the digital image signal by adjusting the amplification factor of the analog image signal output from the image reading unit. Processing equipment. 9. The image reading means includes a photoelectric conversion means for converting an optical signal obtained at the time of reading an image into an electric signal, and an amplifying means for amplifying the electric signal photoelectrically converted by the photoelectric conversion means. Clamping means for clamping the electric signal amplified by the amplifying means, and level shifting the electric signal clamped by the clamping means so as to match the signal level of the A / D converting means. The image processing apparatus according to claim 2, wherein the image processing apparatus comprises a level shift unit. 10. A photoelectric conversion means for converting at least an optical signal obtained at the time of reading an image into an electric signal, an amplification means for amplifying the electric signal photoelectrically converted by the photoelectric conversion means, and an amplifier amplified by the amplification means. Clamping means for clamping the electric signal, and A / D for the electric signal clamped by the clamping means.
An image reading means composed of a level shift means for level shifting so as to match the signal level of the converting means; and an A / D conversion means for converting an analog image signal generated by the image reading means into a digital image signal. An image processing apparatus including: the photoelectric conversion means, the amplification means, the clamp means, and the level shift means in one package. 11. The image processing apparatus according to claim 10, wherein the photoelectric conversion means, the amplification means, the clamp means and the level shift means are formed on the same wafer. 12. A photoelectric conversion means for converting an optical signal obtained at the time of reading an image into an electric signal, an amplification means for amplifying an electric signal photoelectrically converted by the photoelectric conversion means, and an electric power amplified by the amplification means. Clamping means for clamping the signal and level shifting means for level shifting the electric signal clamped by the clamping means so as to match the signal level of the A / D converting means are formed on the same wafer. And an analog processing means composed of two elements, that is, the image reading means and the A / D converting means for converting the analog image signal generated by the image reading means into a digital image signal. The image processing device is characterized in that the optical signal of the image is converted into a digital electric signal. 13. The image reading means comprising the photoelectric conversion means, the amplification means, the clamp means and the level shift means on the same wafer, and the A / D.
The image processing apparatus according to claim 12, wherein the conversion unit and the conversion unit are mounted on the same substrate. 14. The image processing according to claim 9, wherein the amplification means is configured to perform amplification processing only at a predetermined amplification factor. apparatus. 15. An image processing apparatus having image signal processing means for processing an input digital image signal, wherein the bit width of the input signal of the image signal processing means is made larger than the bit width of the output signal. An image processing device characterized by: 16. A photoelectric conversion means for converting an optical signal obtained at the time of reading an image into an electric signal and an analog adjusting means for adjusting the level of an analog image signal obtained by photoelectric conversion in the photoelectric conversion means are the same. The analog adjusting means provided on a wafer, the amplifying means for amplifying the electric signal photoelectrically converted by the photoelectric converting means, and the clamp means for performing a clamp process on the electric signal amplified by the amplifying means, An image reading sensor comprising: a level shift unit that shifts a level of an electric signal clamped by the clamp unit to match a signal level of an A / D conversion unit.