JP2003230016A - Color image reader - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a color image reader capable of correcting color slippage with a small scale circuit configuration and at a low cost. <P>SOLUTION: Full pixel color slippage data calculation circuits 112b (113b, 114b) calculate color slippage data of all pixels from color slippage data at a plurality of positions which are measured at adjustment and stored in color slippage memories 112a (113a, 114a) of the color image reader 1, and color slippage correction circuits 112c (113c, 114c) correct the color slippage of an image signal on the basis of the calculated color slippage data of all the pixels. The full color slippage data calculation circuits 112b (113b, 114b) consist only of adders and subtractors and employ an algorithm of digital differentiation analysis (DDA) to calculate the color slippage data of all the pixels from the color slippage data at a plurality of the positions. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color image reading device.

[0002]

2. Description of the Related Art As one of image reading apparatuses used in image scanners, copying machines, facsimile machines, etc.
There is one using a one-dimensional image sensor such as a linear CCD. An image reading device using a one-dimensional image sensor,
As shown in FIG. 6, by illuminating an original with a light source, converging the reflected light by an optical system and forming an image on a one-dimensional image sensor (electrical scanning), an image for one line in the main scanning direction can be obtained. Documents, mirrors, or light sources in the reading and sub-scanning directions,
A two-dimensional image is read by performing mechanical scanning by moving a mirror, an optical system, a carriage having a one-dimensional image sensor, or the like.

In the case of a color image reading device, 3 lines C
A color image sensor such as a CD is used. Figure 7
As shown in, the 3-line CCD has R (red) and G
Sensors of respective colors (green) and B (blue) are arranged at predetermined intervals in the sub-scanning direction. Therefore, in order to obtain the image signals R, G, and B of the original image at the same position as the image signals of the same line, the image signal B corresponds to the interval of two lines of the sensor and the image signal G corresponds to one line of the sensor. By performing the line-to-line correction of delaying the output by the interval, the time shift in the sub-scanning direction is corrected.

Further, in the 3-line CCD, since the sensors for the respective colors are provided at different positions in the sub-scanning direction, color misregistration also occurs in the main scanning direction due to sensor twisting and chromatic aberration of the optical system. Therefore, it is necessary to perform color misregistration correction and output an image signal having no color misregistration. As a technique relating to such color misregistration correction, Japanese Patent Laid-Open No. 6-33
In Japanese Patent No. 4814, since the deviation amount at the left and right ends of the R sensor and the B sensor is large,
It is divided into three in the center and left, and the shift amount of all pixels of the left and right R and B sensors is measured in pixel units and stored in memory in correspondence with the address of each pixel in the main scanning direction. Describes a technique for correcting color misregistration.

[0005]

However, when measuring the deviation amount of all pixels and storing it in the memory, it takes a lot of time and labor for the measurement and a large capacity memory is required. In addition, if the shift amount of each pixel is calculated from the shift amounts at a plurality of locations at the end and center of the sensor, a large-scale arithmetic unit such as a multiplier is required. Therefore, the structure for performing the color misregistration correction is costly.

An object of the present invention is to perform color misregistration correction in a color image reading apparatus with a low-cost and small-scale circuit configuration.

[0007]

In order to solve the above-mentioned problems, the invention according to claim 1 is arranged in parallel with a light source for irradiating a document image surface with light in the main scanning direction, and the light source from the document image surface is provided. Equipped with multiple sensors that detect reflected light for each primary color,
In an image reading apparatus that reads a document image by moving the plurality of sensors in the sub-scanning direction with respect to the document image surface, the deviation amount of the reading position of each sensor in the main scanning direction is set to a plurality of positions in the main scanning direction. Of the main scanning direction detected by each of the sensors based on the shift amount of the reading position corresponding to the address of the plurality of locations in the main scanning direction stored in the storage unit and the storing unit that stores the main scanning direction. Based on the color shift data calculating means for calculating the shift amount of the read positions of all the pixels constituting the color shift data, and the color shift data of all the pixels in the main scanning direction calculated by the color shift data calculating means, the reading is performed. And a color shift correction unit that corrects a color shift due to a shift in the reading position of the original image in the main scanning direction.

According to the first aspect of the present invention, a plurality of sensors are arranged in parallel in the main scanning direction for illuminating the original image surface with light, and a plurality of sensors for detecting the reflected light from the original image surface for each primary color. In a color image reading device that reads a document image by moving the plurality of sensors in the sub-scanning direction with respect to the document image surface, the main scanning can be performed by determining the deviation amount of the reading position of each sensor in the main scanning direction. It is stored in the storage means in association with the addresses of a plurality of locations in the direction, and the shift amount of the read positions of all pixels is calculated as color shift data based on the shift amount of the read positions corresponding to the stored addresses of the plurality of locations, Based on the calculated color shift data, the color shift due to the shift of the reading position in the main scanning direction of the read original image is corrected. Therefore, as compared with the method of storing the color shift data of all pixels in the storage unit, the storage amount in the storage unit is significantly reduced, and the cost can be reduced.

According to a second aspect of the invention, in the first aspect of the invention, the color misregistration data calculation means is configured to have only an adder and a subtractor as an arithmetic unit.

According to a second aspect of the present invention, in the first aspect of the present invention, the color shift data of all pixels is calculated by using only an adder and a subtracter as an arithmetic unit. Therefore, since the color shift data of all pixels can be calculated without using a complicated circuit configuration such as a multiplier and a divider, the circuit scale can be reduced and the cost can be reduced.

According to a third aspect of the present invention, in the first or second aspect of the invention, the color misregistration data calculating means is provided with read positions corresponding to a plurality of addresses in the main scanning direction stored by the storage means. The shift amount is linearly interpolated by a digital differential analysis to calculate the shift amount of the read position of all pixels constituting the main scanning direction detected by each sensor as color shift data.

According to a third aspect of the present invention, in the first or second aspect of the present invention, the deviation amount of the reading position corresponding to a plurality of addresses in the main scanning direction stored by the storage means is linearly obtained by digital differential analysis. By performing the interpolation, the shift amount of the read position of all the pixels forming the main scanning direction detected by each sensor is calculated as the color shift data. Therefore, the color shift data of all pixels can be calculated easily and at high speed.

[0013]

BEST MODE FOR CARRYING OUT THE INVENTION The present embodiment will be described below in detail with reference to the drawings. In this embodiment, color misregistration data memories 112a, 113a, and 11 to be described later are provided.
Reference numeral 4a denotes all-pixel color shift data calculation circuits 112b, 113b, and 1 as storage means described in the claims of the present invention.
Reference numeral 14b has a function as a color shift data calculating means, and the color shift correcting circuits 112c, 113c, and 114c have a function as a color shift correcting means.

First, the structure will be described. FIG. 1 is a block diagram showing a functional configuration of a main part of the image reading device 1 according to the present embodiment. As shown in FIG. 1, the image reading device 1
Is an R sensor 101, a G sensor 102, a B sensor 103
Sensor 100, A / D converters 104 to 106,
The shading correction circuits 107 to 109, the inter-line correction circuits 110 and 111, the color misregistration correction units 112 to 114, and the control unit 115 are configured.

The sensor 100 has three CCDs (Charge Coupled Devices) of an R sensor 101, a G sensor 102, and a B sensor 103 arranged in parallel at predetermined intervals in the sub-scanning direction.
Is a 3-line CCD image sensor in which is arranged. The R sensor 101 is input via an optical system (not shown),
R (red) from the reflected light of the light emitted from the light source to the color original
The light quantity of the component is converted into an analog electric signal. G sensor 1
Reference numeral 02 converts the amount of light of the G (green) component from the reflected light of the light emitted from the light source to the color original, which is input through an optical system (not shown), into an analog electric signal. B sensor 103
Converts the light amount of the B (blue) component from the reflected light of the light emitted from the light source to the color original, which is input through the optical system (not shown), into an analog electric signal.

The A / D converter 104 converts the R component analog electric signal input from the R sensor 101 into a digital signal, and the A / D converter 105 converts the G component analog input from the G sensor 102. The electric signal is converted into a digital signal, and the A / D converter 106 converts the B component analog electric signal input from the B sensor 103 into a digital signal.

Shading correction circuits 107-109
For each R, G, B digital signal output by the A / D converters 104 to 106 due to variations in the sensitivity of the sensor 100 between pixels, non-uniformity of the light source illuminance, reduction of the peripheral light amount of the optical system, etc. Correct the unevenness that occurs. The shading correction circuit 107 outputs the R digital signal output from the A / D converter 104, and the shading correction circuit 108 outputs A digital signal.
The shading correction circuit 109 performs shading correction on the G digital signal output from the A / D converter 105 and the B digital signal output from the A / D converter 106, and outputs R, G, and B image signals. To do.

The interline correction circuits 110 and 111 are
R sensor 101, G sensor 102, B sensor 103
It corrects a temporal shift in the sub-scanning direction caused by the two sensors being arranged at a predetermined interval in the sub-scanning direction.
Specifically, the pixel signals R, G, B of the original image at the same position
In order to obtain the image signal of the same line, the inter-line correction circuit 111 delays the image signal B by two line intervals of the sensor and the inter-line correction circuit 110 delays the image signal G by one line interval of the sensor. Output.

The color misregistration correction units 112 to 114 are provided in the sensor 1.
The color misregistration in the main scanning direction caused by the twist of 00 and the chromatic aberration of the optical system is corrected. Color misregistration correction is R from the absolute standard
The correction may be performed using the deviation amount of the pixels of each color of GB, or the reference color may be determined and the correction of the address of each pixel in the main scanning direction from the pixel position of the same address of the reference color may be performed. Good. The color shift correction unit 112 performs color shift correction on the image signal R, the color shift correction unit 113 performs the image signal G, and the color shift correction unit 114 performs the color shift correction on the image signal B.

The CPU (Central Processi) of the control unit 115
ng Unit) 115a reads a program for controlling the image reading device 1 stored in advance in the storage unit 115b, and controls the entire image reading device 1 according to the read program. When reading an image, the CPU 115a sets the A / D converters 104 to 106 to the A / D conversion block 11, the shading correction circuits 107 to 109 to the shading correction block 12, and the interline correction circuit 1.
10 to 111 are the line-to-line correction block 13, and the color misregistration correction units 112 to 114 are the color misregistration correction block 14, and the operation of each unit is controlled by generating a timing signal to the sensor 100 and each block. Storage unit 115b
Is an EEPROM (Electrically Erasable Programmab
le ROM) and the like, and stores a program for controlling the image reading device 1.

Next, the internal structure of the color misregistration correction unit 112 will be described in detail with reference to the block diagram of FIG. As shown in FIG. 2, the color shift correction unit 112 includes a color shift data memory 112a, an all-pixel color shift data calculation circuit 112b, and a color shift correction circuit 112c. The all-pixel color-misregistration data calculation circuit 112b calculates the color-misregistration data of all pixels based on the color-misregistration data stored in the color-misregistration data memory 112a, and the color misregistration data is calculated based on the calculated color-misregistration data. The misregistration correction circuit 112c performs color misregistration correction.

The color misregistration data memory 112a is an EEPR.
It is configured by a non-volatile memory such as OM (Electrically Erasable Programmable ROM). When the deviation amount is adjusted in advance at the time of shipment of the apparatus, main scanning from a reference position of a plurality of points of the R sensor 101, for example, the right end, the center, the left end, the middle point between the right end and the center, and the five points at the middle point between the left end and the center. A direction shift amount (hereinafter, referred to as a color shift amount) is measured, and the measured color shift amount is stored in the color shift data memory 112a as color shift data in association with the main scanning direction address of each pixel. The reference position may be an absolute reference,
A reference color may be provided and the amount of deviation of the reference color from the sensor in the main scanning direction may be measured. When the reference color is provided, the color misregistration correction unit relating to the reference color is not required. Therefore, for example, when R is the reference color, the color misregistration correction unit 112 is not required.

It should be noted that the color shift amount (color shift data) of the pixels between the measurement points is calculated by linearly interpolating by the all-pixel color shift data calculation circuit 112b described later, and the color shift data of all the pixels is calculated. Therefore, it is preferable that the measurement point of the color misregistration amount is determined based on an average pattern of deviation of the reading position in the main scanning direction due to twisting of the R sensor 101, distortion at the time of attachment, and the like. Further, four or more measurement points of the color shift amount are necessary in order to make the color shift data of each pixel closer to the actual one.

The all-pixel color shift data calculation circuit 112b is
Color shift data for all pixels is calculated from the color shift amounts (color shift data) stored in the color shift data memory 112a at a plurality of locations. Specifically, the x-axis direction is the address in the main-scanning direction (main-scanning address), and the y-axis direction is the color misregistration amount. The color shift amount of each pixel in between is calculated. This is performed for all the measurement points to calculate the color shift data of all pixels. One unit (of the color shift amount) in the y-axis direction is the accuracy of the color shift data recorded in the color shift data memory 112a, such as 1/8 pixel or 1/16 pixel.

FIG. 3 is a circuit diagram showing an all-pixel color shift data calculation circuit 112b for calculating color shift data of all pixels using a digital differential analysis (DDA) algorithm.
The operation of the all-pixel color shift data calculation circuit 112b will be described below with reference to the block diagram of FIG. 3 and the flowchart of FIG. The digital differential analysis (DDA) algorithm is different when the difference in the x-axis direction is larger than the difference in the y-axis direction and when it is smaller. In the present embodiment, since the amount of color misregistration is corrected by less than ± 1 pixel, an explanation will be given using an algorithm when the difference in the x-axis direction is larger than the difference in the y-axis direction. A calculation circuit using an algorithm in which the difference in the direction is smaller than the difference in the y-axis direction may be additionally provided.

As shown in FIG. 3, first, when the main scanning of the sensor 100 is started, a line start signal is input to the all-pixel color shift data calculation circuit 112b, and this signal is AC.
It becomes an LR signal and all pixel color shift data calculation circuit 112b
Is initialized. Next, the CPU 115 shown in FIG.
By a, the data is read from the color shift data memory 112a from the two measurement points (measurement points) having the smallest addresses and input to the all-pixel color shift data calculation circuit 112b. That is, if the smaller one of the two measurement points is the first measurement point and the larger one is the second measurement point, the address of the first measurement point is the address initial value, and the second measurement point and the second measurement point. The address difference at the first measurement point is the address difference, the difference in color shift amount between the first measurement point and the second measurement point is the color shift amount difference, and the color shift amount at the first measurement point is the initial color shift amount. It is input to the pixel color shift data calculation circuit 112b.

When the necessary data stored in the color misregistration data memory 112a is input, the all-pixel color misregistration data calculation circuit 112b starts the inter-measurement-point color misregistration data calculation process shown in FIG. It should be noted that in the first steps S1 to S7 after the start of the process of calculating the color shift data between measurement points, the color shift amount at the first measurement point is calculated based on the input initial value, and the color shift is performed in step S8. The data is output as data, the process proceeds to the next address in step S9, and the process returns to step S1 again to calculate and output the color misregistration amount of the next address. Hereinafter, the address of the pixel whose color shift amount is calculated in steps S1 to S8 will be referred to as the current address.

The all-pixel color shift data calculation circuit 112b is
½ of the address difference is calculated and set as A (step S
1) Then, the color shift amount difference is subtracted from the calculated value A, and the subtraction result is set to A (step S2). Next, when the value of A is negative (step S3; YES), the all-pixel color shift data calculation circuit 112b adds the address difference to the value of A and sets the addition result to A (step S4), and determines the color shift amount. If the difference is positive (step S5; positive), the color misregistration amount is incremented by 1 (step S6), and if the color misregistration amount difference is negative (step S5; negative), the color misregistration amount is decremented by 1. (Step S7). Then, the process proceeds to step S8, and the color shift amount calculated in step S6 or step S7 is output as color shift data corresponding to the current address (step S
8).

On the other hand, if the value of A is not negative in step S3 (step S3; NO), the all-pixel color shift data calculation circuit 112b proceeds to step S8 and the previously calculated color shift amount is used as it is as the current address. Is output as color shift data corresponding to (step S8).

The all-pixel color shift data calculation circuit 112b is
After the color shift data is output in step S8, the process proceeds to the next address (step S9) until the calculation and output of the color shift data corresponding to the addresses of all the pixels between the two measurement points are completed (step S10; NO), steps S1 to S
9 is repeatedly executed. When the calculation of the color shift data of all pixels between the two measurement points is completed (step S10; YES),
Although the color shift data calculation process between the measurement points is completed, the all-pixel color shift data calculation circuit 112b calculates the color shift data at the next measurement point interval.

In the block diagram of FIG. 3, DOTCNT
The address initial value and the address difference input to the all-pixel color shift data calculation circuit 112b are input to the register,
Steps S1 to S1 of the color shift data calculation process between the measurement points
When S8 is executed and the process moves to the next address in step S8, the DOTCNT counter is counted down, and when it reaches 0, the NEXT signal is generated. NEX
The T signal matches when YES in step S10 of the color misregistration data calculation process between measurement points, which means that the calculation of the amount of color misregistration between the two input measurement points is completed.

Further, the ADRCNT register shown in FIG.
One half of the address difference is input as an initial value, and the Sub subtractor subtracts the color misregistration amount difference from the value in the ADRCNT register (corresponding to steps S1 and S2 in FIG. 4). Then, when the subtraction result becomes negative (step S3 in FIG. 4;
A CHG signal is generated, and the address difference is added to the value of the ADRCNT register by the Add adder (corresponding to step S4). CHG signal becomes a pulse,
If the color difference difference is positive (step S5; corresponding to positive),
The color shift amount stored in the ZURE register is counted up (corresponding to step S6), and when the color shift amount difference is negative (step S5; corresponds to negative), the color shift amount stored in the ZURE register is calculated. Count down (corresponding to step S7). Then, the count-up or count-down color shift amount is output as all-pixel color shift data (corresponding to step S8).

FIG. 5A is a diagram showing the relationship between the color shift data obtained by measuring the color shift amounts of all pixels, with the main scanning address as the x-axis and the color shift amount as the y-axis. In FIG. 5B, the x-axis is the main scanning address, the y-axis is the color shift amount, and the color shift amount at a plurality of locations is calculated by linearly interpolating between the measurement points by the color shift data calculation circuit 112b. It is the figure which represented the color shift data which was shown with the line graph. FIG. 5C is a diagram showing the difference between FIG. 5A and FIG. 5B on the same graph in order to show the difference.
The dotted line is FIG. 5 (a) and the solid line is FIG. 5 (b). Figure 5
As shown in (c), all pixel color shift data calculation circuit 112
The color misregistration data calculated by b can take a value substantially close to the color misregistration data obtained by measuring all pixels.

The color misregistration correction circuit 112c performs resampling or the like on the basis of the color misregistration data of all pixels output by the all-pixel color misregistration data calculation circuit 112b.
The color misregistration due to the misalignment of the reading position in the main scanning direction, which is caused by the twist of the sensor, the distortion when the sensor is attached, and the like, is corrected.

The color shift correction unit 113 includes a color shift data memory 113a and a color shift data calculation circuit 113 for all pixels.
b, the color shift correction circuit 113c is provided, and the same operation as the color shift correction for the image signal R by the color shift correction unit 112 described above is performed on the image signal G. The color shift correction unit 114 includes a color shift data memory 114a, an all-pixel color shift data calculation circuit 114b, and a color shift correction circuit 114.
The image signal G is subjected to the same operation as the color misregistration correction by the color misregistration correction unit 112 described above.

As described above, according to the image reading apparatus 1, the color shift data memory 112a (113a, 114).
All pixel color shift data calculation circuit 112b from the color shift data stored in a) at a plurality of positions measured at the time of adjustment.
The color shift data of all pixels is calculated by (113b, 114b), and the color shift correction circuit 112c (113c, 114c) corrects the color shift of the image signal based on the calculated color shift data of all the pixels. All pixel color shift data calculation circuit 1
The reference numeral 12b (113b, 114b) is composed of only an adder and a subtractor, and calculates the color shift data of all pixels from the color shift data at a plurality of locations by using a digital differential analysis (DDA) algorithm.

Therefore, the labor for measuring the color shift data of all the pixels can be saved, and the work at the time of adjustment can be simplified. Further, the storage amount stored in the color shift data memory 112a (113a, 114a) is significantly smaller than that in the case where the color shift data of all pixels is stored, so that the memory usage amount can be reduced. Further, the all-pixel color shift data calculation circuit 1
In 12b (113b, 114b), since the arithmetic unit is composed of an adder and a subtractor and does not use a multiplier and a divider, the circuit scale can be small and the color shift data of all pixels can be calculated at high speed. it can.

The description in the above embodiment is a preferred example of the image reading apparatus 1 according to the present invention.
It is not limited to this. For example, although the color misregistration data memory corresponding to each of the R, G, and B sensors is provided in the above-described embodiment, one memory has R, G, and
The color misregistration data of B may be collectively stored. In addition, the detailed configuration and the detailed operation of the image reading apparatus 1 can be appropriately changed without departing from the spirit of the present invention.

[0039]

According to the first aspect of the present invention, a plurality of light sources for irradiating light to the original image surface and a plurality of light sources arranged in parallel in the main scanning direction to detect reflected light from the original image surface for each primary color. In a color image reading apparatus that includes a sensor for moving a plurality of sensors in the sub-scanning direction with respect to the document image surface to read a document image, the deviation amount of the reading position of each sensor in the main scanning direction is It is stored in the storage means in association with a plurality of addresses in the main scanning direction, and the shift amounts of the read positions of all pixels are calculated as color shift data based on the shift amounts of the read positions corresponding to the stored addresses of the plurality of places. Then, on the basis of the calculated color shift data, the color shift due to the shift of the reading position of the read original image in the main scanning direction is corrected. Therefore, as compared with the method of storing the color shift data of all pixels in the storage unit, the storage amount in the storage unit is significantly reduced, and the cost can be reduced.

According to the second aspect of the present invention, in the first aspect of the invention, the color shift data of all pixels are calculated only by the adder and the subtracter as the arithmetic unit. Therefore, since the color shift data of all pixels can be calculated without using a complicated circuit configuration such as a multiplier and a divider, the circuit scale can be reduced and the cost can be reduced.

According to a third aspect of the present invention, in the first or second aspect of the present invention, the deviation amount of the reading position corresponding to a plurality of addresses in the main scanning direction stored by the storage means is linearly obtained by digital differential analysis. By performing the interpolation, the shift amount of the read position of all the pixels forming the main scanning direction detected by each sensor is calculated as the color shift data. Therefore, the color shift data of all pixels can be calculated easily and at high speed.

[Brief description of drawings]

FIG. 1 is a block diagram showing a functional configuration of an image reading apparatus 1 according to the present invention.

FIG. 2 is a color misregistration correction unit 112 (113, 114) of FIG.
3 is a block diagram showing the functional configuration of FIG.

3 is an all-pixel color shift data calculation circuit 112b in FIG.
It is a circuit diagram which shows (113b, 114b).

4 is an all-pixel color shift data calculation circuit 112b in FIG.
It is a flowchart which shows the color gap data calculation process between measurement points performed by (113b, 114b).

FIG. 5 is color shift data obtained by measuring color shift amounts of all pixels;
FIG. 9 is a line graph showing the color shift data calculated by linearly interpolating the measurement points by the all-pixel color shift data calculation circuit 112b from the color shift amounts at a plurality of locations.

FIG. 6 is a diagram showing a simplified image reading method of an image reading apparatus using a one-dimensional image sensor.

FIG. 7 is a diagram showing a sensor arrangement of a color image sensor.

[Explanation of symbols]

1 Image Reading Device 11 A / D Conversion Block 12 Shading Correction Block 13 Interline Correction Block 14 Color Misregistration Correction Block 100 Sensor 101 R Sensor 102 G Sensor 103 B Sensor 104-106 A / D Converter 107-109 Shading Correction Circuit 110 , 111 inter-line correction circuits 112 to 114 color shift correction units 112a, 113a, 114a color shift data memories 112b, 113b, 114b all-pixel color shift data calculation circuits 112c, 113c, 114c color shift correction circuit 115 control unit 115a CPU 115b storage Department

   ─────────────────────────────────────────────────── ─── Continued front page    F term (reference) 5B047 AA01 AB04 BB03 CA12 CB05                       DB01 DC20                 5C072 BA01 BA02 BA19 CA02 EA05                       EA08 QA10 QA17 UA11 UA13                       UA18 XA01                 5C077 LL17 MM03 MP08 PP32 PP39                       PP47 PQ12 PQ22 RR19                 5C079 HB01 JA03 JA23 LA24 MA01                       MA11 NA09

Claims (3)

[Claims] 1. A light source for irradiating a document image surface with light, and a plurality of sensors arranged in parallel in the main scanning direction to detect reflected light from the document image surface for each primary color. In an image reading device that reads a document image by moving it in the sub-scanning direction with respect to the document image surface, the deviation amount of the reading position in the main scanning direction of each sensor is associated with a plurality of addresses in the main scanning direction. And a storage unit that stores the data in the main scanning direction stored in the storage unit on the basis of displacements of reading positions corresponding to addresses at a plurality of positions in the main scanning direction. The color shift data calculating means for calculating the shift amount of the reading position as color shift data, and the read based on the color shift data of all the pixels in the main scanning direction calculated by the color shift data calculating means. Color image reading apparatus characterized by comprising by the main scanning direction of the shift of the reading position of the original image and the color shift correcting means for correcting the color shift, was. 2. The color image reading apparatus according to claim 1, wherein the color shift data calculation means is configured to include only an adder and a subtractor as arithmetic units. 3. The color misregistration data calculation means linearly interpolates the deviation amount of the reading position corresponding to a plurality of addresses in the main scanning direction stored by the storage means by digital differential analysis to linearly interpolate each of the sensors. 3. The color image reading apparatus according to claim 1, wherein the shift amount of the read positions of all the pixels constituting the main scanning direction detected by is calculated as color shift data.