JPH08265506A - Image reading device - Google Patents

Abstract

PURPOSE: To provide an image reading device within can read an image while continuously changing the image resolution in the subscarming direction and also can prevent the occurrence of the color slippage. CONSTITUTION: An RGB position shift correction circuit 19 decides the relative shift speed between every linear image sensor 13 and a color original 1 when the desired image resolution is inputted, so that the value (divisor value N) obtained by dividing the reading position shift extent measured in the subscanning direction among plural sensors 13 put on the original 1 by the scanning line pitch set on the original 1 is equal to the integer value. Based on this decided relative shift speed, an image reading device reads the original 1. Furthermore, the cicuit 19 performs the line number conversion processing of the image data after reading the original 1 ana thus generates the image data of the desired image resolution. Thus the image reading device can continuously change the desired image resolution without causing any color slippage.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image reading device,
More specifically, the present invention relates to an image reading apparatus that compensates a reading position deviation when scanning a color original in a scanning line sequential manner by a plurality of linear image sensors.

[0002]

2. Description of the Related Art In a plane scanning type plate-making color scanner or the like, three units are provided corresponding to read light color components of red, green and blue (hereinafter referred to as "R", "G" and "B", respectively). The linear image sensor of is provided. Then, the R component, G component, and B component of the original image are read for each pixel by these three linear image sensors, and the image data of each color component read in this way or the correction processing thereof is performed. A recorded image (color separation image or duplicate color image) is formed based on the data.

In the image reading apparatus as described above, it is important to ensure the consistency between the image reading positions of the plurality of linear image sensors. This is because if there is no match between the image reading positions of the multiple linear image sensors, the color component information at positions that are offset from each other on the original image will be used, so that the color image to be reproduced will be colored. This is because a gap or the like will occur.

The most typical method for coping with such a situation is, as shown in FIG. To the light α by this color separation optical system 3.
Is decomposed into each color component light and given to the linear image sensors 4a, 4b, 4c for detecting each color component. However, in this method, since the image reading light is dispersed, the amount of light reaching each linear image sensor is drastically reduced, and the image quality is degraded. Moreover, since the highly accurate color separation optical system 3 must be prepared, the size and cost of the image reading apparatus as a whole are increased.

Therefore, in order not to reduce the image reading light amount, the color separation optical system is not used, and the output timing of the image data from each linear image sensor is adjusted so that the deviation between the image reading positions is eliminated. Techniques to compensate are proposed. For example, JP-A-61-1
In the technique in 08,253 No. disclosed, as shown in FIG. 13, the respective linear image sensors 4a~4c for detecting different color components, different reading line W _a
Each image of W _c is read.

In this case, if the original image 1 is moved in the (-Y) direction shown in the figure to perform sub-scanning, one pixel P
Image information of the first image is read by the linear image sensor 4a, and then the other linear image sensors 4b, 4
It is read in the order of c. Therefore, each color component data of the pixel P is output from each of the linear image sensors 4a to 4c at different timing.

Therefore, in this method, the color component data from the other linear image sensors 4a and 4b which read the pixel P earlier than the linear image sensor 4c is supplied to the buffer memories 5a and 5b. Then, by delaying the output timing of the color component data from the linear image sensors 4a and 4b by using the delay function of the buffer memories 5a and 5b, the same pixel (same scanning line)
The color component data S _a , S _b , and S _{c of the above} are output at the same timing.

[0008]

By the way, in an apparatus which does not use a color separation optical system like the apparatus shown in FIG. 13, each of the linear image sensors 4a to 4c detects an instantaneous reading line W at the same time. _{a to} W _c (see FIG. 14) are separated by a predetermined distance A. This distance A mainly depends on the distance between the linear image sensors and the lens 2
It is determined by the optical magnification of. On the other hand, the width d in the sub-scanning direction of the scanning lines L ₀ , L ₁ , ... To be read is determined according to the resolution required for the recorded image. However, when the linear image sensors 4a to 4c are formed by the CCD cell array, the "scan line" referred to here means that the original image 1 continues to move in the (-Y) direction during one charge accumulation time. by the instantaneous reading line W _a to _W-c is pointing to a band-like region to be scanned on the original 1 in the charge accumulation time. In FIG. 14, the width of the scanning line is exaggerated.

As described above, since the distance A and the width d are determined from different viewpoints, the relationship between the two is generally
It is possible to change variously, and the distance A is not always an integral multiple of the width d. In particular, in a device that can continuously change the resolution, the value that the width d can take is continuous, and inconsistency between the distance A and the width d will inevitably occur.

Since the reading timing itself of each of the linear image sensors 4a to 4c is common, for example, one linear image sensor 4c scans the scanning lines L ₀ , L ₁ ,
, While the other linear image sensor 4b sequentially reads the image line information of the scanning lines L ₀ ′, L ₁ ′ ... Go. The remaining linear image sensor 4a is also misaligned.

Therefore, even if the output timings of the color component data from these linear image sensors 4a to 4c are adjusted by the buffer memories 5a and 5b of FIG.
Since there is a deviation between the read scanning lines in the first place, there is a problem that accurate image data cannot be obtained.

In Japanese Patent Laid-Open No. 61-108253, the output timing is corrected based on the reading of the test chart, but in this case, the same problem as described above occurs.

The color shift caused by such a situation is
Since it occurs within the range of the scanning line pitch d or less, it does not pose a problem in ordinary image recording. However, particularly when highly accurate image recording is required or when recording is performed on a portion where the density or color tone changes abruptly on the original image, the influence of such scanning line deviation is effectively prevented. Becomes important.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide an image reading apparatus capable of reading an image while preventing the occurrence of color shift and continuously changing the resolution in the sub-scanning direction.

[0015]

The invention according to claim 1 is
A plurality of linear image sensors extending in the main scanning direction are arranged in parallel in a plurality of rows, and while the linear image sensors and the color original are relatively moved in the sub scanning direction, the color original of the color original is detected by each linear image sensor. An image reading apparatus for reading each color component of an image in a scanning line-sequential manner, the optical system including a lens, for forming an image of a color original on a light receiving surface of each of a plurality of linear image sensors by a predetermined magnification. Input means for inputting resolution, such that a value (divisor value N) obtained by dividing the reading position shift amount of the plurality of linear image sensors on the color original in the sub-scanning direction by the scanning line pitch on the color original becomes an integer value. Determining means for determining the relative moving speed in the sub-scanning direction between each linear image sensor and the color original, and the relative moving speed determined by the determining means Based on the above, the speed adjusting means for adjusting the relative moving speed in the sub-scanning direction between each linear image sensor and the color original document, the storage means for storing the image data obtained by reading the color original document, and the storage means. There is provided a line number conversion means for converting the number of lines of the image data stored in to generate the image data of the target resolution input by the input means.

According to a second aspect of the present invention, in the first aspect of the invention, the relative moving speed in the sub-scanning direction and the relative moving speed such that the divisor value N becomes an integer value in the state where the optical magnification in the optical system is fixed. The determining means further includes a table memory storing a plurality of corresponding resolutions, and when the target resolution is input from the input means, the determining means searches the table memory for a resolution that is equal to or higher than the target resolution and is the closest.
It is characterized in that the relative moving speed in the sub-scanning direction corresponding to the retrieved resolution is determined as the moving speed given to the speed adjusting means.

[0017]

According to the first aspect of the present invention, when the target resolution is input, the value obtained by dividing the reading position shift amount of the plurality of linear image sensors on the color original in the sub-scanning direction by the scanning line pitch on the color original. The relative moving speed in the sub-scanning direction between each linear image sensor and the color original is determined such that the (divisor value N) is an integer, and the color original is determined based on the determined relative moving speed. I am trying to read. By the way, under the condition that the divisor value N becomes an integer value, the resolution changes discretely,
The image data obtained by reading does not always match the target resolution. So, after reading the color original,
By performing the line number conversion processing of the image data, the image data of the target resolution is generated. This makes it possible to continuously change the target resolution without causing color misregistration.

According to the second aspect of the present invention, the relative moving speed in the sub-scanning direction and the corresponding resolution such that the divisor value N becomes an integer value while the optical magnification in the optical system is fixed are stored in advance in a table. Multiple sets are stored in the memory. When the target resolution is input, the table memory is searched for a resolution that is equal to or higher than the target resolution and is the closest to the target resolution, and the table memory is used so that reading can be performed at the searched resolution. The relative moving speed between each linear image sensor and the color original is adjusted based on the relative moving speed in the sub-scanning direction. This simplifies the procedure for determining the relative movement speed, which facilitates the arithmetic processing.

[0019]

【Example】

(1) Overall Configuration of Embodiment FIG. 1 is a diagram showing the overall configuration of an image reading apparatus according to an embodiment of the present invention. In FIG. 1, on the document table 6,
A document 1 and a reference plate 7 for shading correction are placed. This document table 6 is moved by the motor 8 (-Y).
Is moved in the direction. A light source 9 for transmission and a light source 10 for reflection are provided above and below the document table 6, respectively. Further, below the document table 6, there is provided a mirror 11 for reflecting the reading light transmitted through the document 1 or reflected by the document 1 in a predetermined direction. The reading light reflected by the mirror 11 is incident on the lens 2 via the diaphragm 12. The multi-row image sensor 13 is arranged so as to be located on the image plane of the lens 2.

The output signal of the multi-row image sensor 13 is amplified and sample-held by the amplifier & sample-hold circuit 15, and then converted into a digital signal by the A / D converter 16. The output of the A / D converter 16 is given to the shading correction circuit 17 and is subjected to known shading correction. The output of the shading correction circuit 17 is R
It is given to the GB position shift correction circuit 19. The RGB position shift correction circuit 19 switches the moving speed (the feed speed in the sub-scanning direction) of the document table 6 according to the input target resolution, and performs the line number conversion processing of the reading line signal. In order to switch the moving speed in the sub-scanning direction, RG
The output of the B position shift correction circuit 19 is given to the motor 8. The output of the RGB misalignment correction circuit 19 is given to the post-processing circuit 18 and various processing (for example, γ correction) is performed.
Is applied.

FIG. 2 is a block diagram showing a more detailed structure of the RGB position shift correction circuit 19 in FIG. In FIG. 2, this RGB position shift correction circuit 19 has a CP
U 191, ROM 192, RAM 193, input device 194, table memory 195, motor driver 196, DSP (digital signal processor) 19
7 and 7. The operation program of the CPU 191 is stored in the ROM 192. The CPU 191 operates according to this operation program. RAM193
Stores various data necessary for the data processing of the CPU 191. The input device 194 includes a keyboard operated by an operator, a mouse, etc., and inputs various instructions and data. The table memory 195 is the CPU 1
91 stores table data required when determining the feed speed in the sub-scanning direction, and details thereof will be described later. The motor driver 196 is a drive circuit for operating the motor 8 of FIG. 1, and includes an amplifier and the like.

FIG. 3 is a diagram showing an optical system of the image reading apparatus shown in FIG. FIG. 4 is a perspective view showing a detailed configuration of the multi-row image sensor 13 shown in FIG. As shown in FIG. 4, each of the multi-row image sensors 13 has a C
It has three linear image sensors 132B, 132G, 132R including a one-dimensional array of CD light receiving cells 131. These linear image sensors 132B, 13
The 2G and 132R are integrally formed on a single substrate SB in a state of being arranged in three rows in parallel with each other. In addition, the B color filter 133 is provided on the surface of each cell row.
B and G color filters 133G and R color filters 133R are fixed.

Returning to FIG. 3, the lens 2 is arranged in front of the light receiving surface of such a multi-row image sensor 13. Then, the document 1 to be read is placed at a position facing the multi-row image sensor 13 with the lens 2 interposed therebetween. As is well known, linear image sensors 132B, 1
The 32G and 132R perform main scanning by reading an image for each pixel in the longitudinal direction. Therefore, the longitudinal direction of the linear image sensors 132B, 132G, 132R coincides with the main scanning direction X in the image reading of the document 1.

The original 1 is sent at a predetermined speed V in the (-Y) direction perpendicular to the main scanning direction X by the moving mechanism driven by the motor 8 shown in FIG. Therefore, the linear image sensors 132B, 132G, 132R and the document 1 relatively move in the (−Y) direction at the speed V, and the Y direction becomes the sub-scanning direction.

With the above arrangement relationship,
The linear image sensors 132B, 132G, 132R arranged in multiple rows along the sub-scanning direction Y have the same timing and the light α _B , α _G from the instantaneous reading lines W _B , W _G , W _R on the original 1. , Α _R (only the center plane of the optical path is shown in the figure). Then, while moving the document 1 in the (−Y) direction, each linear image sensor 132B, 132G,
The charge accumulation and charge transfer of the 132R CCD light receiving cell are repeated. Since the color filters 133B, 133G, and 133R are provided, the linear image sensor 132
B, 132G, and 132R respectively receive the B color component, the G color component, and the R color component of the document 1 and photoelectrically convert them.

(2) Principle of color misregistration correction Before describing the operation of the above embodiment, the principle of color misregistration correction of this embodiment will be described.

FIG. 5 shows a reading state when color misregistration occurs. That is, in FIG. 5, the linear image sensors 132B, 132G, 132R on the original 1 are arranged.
Instantaneous reading line W _B , W _G , W _R mutual interval A
Is a non-integer multiple of the pitch d of the scanning lines L ₀ , L ₁ , ... In such a state, no matter how much the output signals from the linear image sensors 132B, 132G, and 132R are timing-adjusted in the buffer memory, the reading position shift remains and color shift occurs. This has been described above.

On the other hand, FIG. 6 shows a reading state when no color shift occurs. That is, in FIG. 6, the mutual interval A between the instantaneous reading lines W _B , W _G , and W _R is an integral multiple (three times in the case of FIG. 6) of the pitch d of the scanning lines L ₀ , L ₁ , ... Has become. In such a state, if the output signals from the linear image sensors 132B, 132G, and 132R are timing-adjusted by a buffer memory or the like, read information at the same position can be obtained at the same timing, and color misregistration does not occur. .

In this embodiment, color misregistration is prevented by always reading the original 1 in the state as shown in FIG. However, one problem is pointed out in such a reading method. Hereinafter, this problem will be described. Let N be the divisor value obtained by dividing the mutual spacing A of the instantaneous reading lines by the scanning line pitch d (N = A / d), and let the divisor value N be a positive integer (that is, the mutual spacing A of the instantaneous reading lines A Is an integer multiple of the scanning line pitch d), the resolution in the sub-scanning direction is fixedly determined for each value of N. In FIG. 7, the resolution when N = 1 is 254.
When the value is set to [dpi], the changing state of the resolution when the value of N is changed is shown. As is apparent from FIG. 7, if the resolution is changed without causing color shift, the change in resolution becomes discrete. With this, the resolution in the sub-scanning direction cannot be freely changed. Therefore, the present inventor has devised a method capable of freely changing the resolution in the sub-scanning direction without causing color shift. Hereinafter, this method will be described.

Referring again to FIG. 6, the scanning line pitch d is
It is determined by the feed speed V of the original 1 in the sub-scanning direction and the line scanning period T of each linear image sensor, and is expressed by the following equation (1). d = VT (1)

On the other hand, the mutual distance A between the instantaneous reading lines
Are linear image sensors 132B, 132G, 13
It is determined by the mutual distance of 2R (hereinafter referred to as the distance between element rows) D (see FIG. 4) and the optical magnification m of the lens 2, and is expressed by the following equation (2). A = D / m (2)

Therefore, from the above equations (1) and (2), the above-mentioned divisor value N is expressed by the following equation (3). N = A / d = D / VTm (3)

The above equation (3) is transformed into the following equation (4). V = D / NTm (4) In the above formula (4), as long as the condition that the divisor value N is an integer value is satisfied, the color shift due to the shift of the reading position does not occur.

Here, the resolution R in the sub-scanning direction (unit is dpi) is given by the following equation (5). R = 25.4 × 10 ⁻³ / d = 25.4 × 10 ⁻³ / VT (5) In the above formula (5), 25.4 × 10 ⁻³ (unit is millimeter) is dpi. Is a conversion coefficient to.

Substituting the above equation (4) into the above equation (5) gives the following equation (6). R = 25.4 × 10 ⁻³ × N × m / D (6)

In the present embodiment, based on the above equations (4) and (6), the feed speed V in the sub-scanning direction that does not cause color misregistration,
And the resolution R corresponding thereto is calculated in advance and stored in the table memory 195. Then, the reading of the document 1 is performed according to the conditions stored in the table memory 195.

FIG. 8 is a diagram showing an example of data stored in the table memory 195. In FIG. 8, the table memory 195 has a default condition (distance between element rows D = 100 [μm], line scanning period T = 10 [m].
s] and optical magnification m = 1), the relationship between the resolution [dpi] and the feed speed V of the document 1 in the sub-scanning direction when the divisor value N is changed is shown. When the target resolution is input, the CPU 191 searches the table memory 195 for a default resolution that is equal to or higher than the target resolution and is the closest, and obtains the feed speed V in the sub-scanning direction corresponding to the resolution. Then, the CPU 191 controls the reading of the original 1 at the acquired feed speed V. Therefore, the reading position does not shift between the linear image sensors, and the color shift does not occur.

As described above, in the present embodiment, the original 1 is once read with discrete resolution without color misregistration. The read image data is stored in a memory (for example, RAM 193) and is converted by the DSP 197 into image data of a target resolution. At this time, the number of lines of the original image data is reduced by performing weighted averaging according to the number of pixels of adjacent lines to be referred to. That is, in this embodiment, image data of the target resolution is obtained by performing so-called line number conversion processing.

In FIG. 9, the image data read at the resolution J [dpi] is subjected to the line number conversion processing to obtain the target resolution K.
An example of the relationship between an input data sequence and an output data sequence when converting to image data of [dpi] is shown. However, K ≦ J. In FIG. 9, in the present embodiment, the input data strings are grouped for each J / K, weighted according to the ratio of pixels included, and the average value is output. For example, focusing on the output data Q (2), it can be seen that the input data string between x (1) and x (2) may be referred to. Specifically, the input data P
(2), P (3) and P (4) will be referred to. By the way, the position of x (1) is a from the beginning of P (2).
(1) It is deviated, and the position of x (2) is P (4)
Is shifted by a (2) from the beginning. Therefore, P (2), P
When averaging (3) and P (4), P (2) and P (4)
It is necessary to reduce the weighting of P (3) by an amount corresponding to the above shift as compared with the weighting of P (3).

For example, J = 800 [dpi], K = 50
If it is 0 [dpi], the values of Q (1) and Q (2) are as in the following equations (7) and (8), respectively. Q (1) = (5/8) P (1) + (3/8) P (2) (7) Q (2) = (2/8) P (2) + (5/8) P ( 3) + (1/8) P (4) (8)

Paying attention to the above equation (7), the weighting for P (1) is (5/8) and the weighting for P (2) is (3/8). Also, paying attention to the above equation (8), the weighting for P (2) is (2/8), the weighting for P (3) is (5/8), the weighting for P (4) is (1 / 8).

The above line number conversion is performed by integration in order to satisfy the sampling theorem. Now resolution J
It is assumed that the input image of [dpi] is converted into the output image of the target resolution K [dpi]. At this time, the output image Q (t) is represented by the following expression (9).

[Equation 1]

However, in the above equation (9), Q (t) ... Output image of the t-th line (t = 1, 2, ...) P (x) ... Input image of the x-th line (x = 1, 2, ...) …)

Digitally describing the above equation (9), the following equations (10) to (12) are obtained.

[Equation 2]

In the above equations (10) to (12),
The various quantities are as follows. t ... Line number of output image (t = 1, 2, ...) x (t) ... Line number of input image to be referred (x (t) =
1, 2, ...) a (t) ... Head position correction amount of reference range (a (t) = 1,
2, ..., J-1) J ... resolution of input image (J = 1, 2, ...) K ... resolution of output image (K = 1, 2, ...; K ≦ J)

Here, the above equation (11) is the output data Q
(T) is an expression for obtaining a range that refers to the input data P (x). That is, the input data P (x) between x (t-1) and x (t) becomes the referenced input data.

Further, the above equation (12) is expressed by x (t-1), x
It is an expression for obtaining how much (t) deviates from the beginning of the corresponding P (x). That is, a (t-1) for x (t-1) becomes a (t) for x (t).

The first term of the above equation (10) is the term for the input data P (x) at the beginning of the reference range, that is,
This is a term for x (t-1). Also, the above equation (10)
The third term of is a term for the input data P (x) at the end of the reference range, that is, a term for x (t). The second term of the above equation (10) is the input data P of the reference range.
When (x) extends over 3 lines, P (x (t-1)
This is a term for input data P (x) existing between (+1) and P (x (t) -1).

Generally, the input data of the reference range is the head reference data ... x (t-1) intermediate reference data ... x (t-1) +1, x (t-1) +.
2, ..., X (t) -1 end reference data ... X (t).

By the line number conversion processing as described above, the output data is generated once per J / K line of the input data.

(3) Operation of the Embodiment FIG. 10 is a flowchart showing the operation of the above embodiment. Hereinafter, the operation of the above embodiment will be described in more detail with reference to FIG.

First, the operator operates the input device 194 to input the target resolution R into the CPU 191 (step S1). Next, the CPU 191 uses the table memory 1
By referring to 95 (step S2), it is determined whether or not the resolution matching the target resolution R exists in the table memory 195 (step S3). If a resolution matching the target resolution R exists in the table memory 195, CP
U 191 reads the target resolution R from the table memory 195.
The feed speed V in the sub-scanning direction corresponding to is acquired (step S4). After that, the CPU 191 sets the feed speed V in the sub-scanning direction acquired in step S4 to the motor driver 19
6 is set (step S5). Thereby, the motor 8 moves the document 1 in the sub-scanning direction at the set speed. After that, the image reading device scans (reads) the document 1 (step S6). At this time, the CPU 19
1 puts the DSP 197 in the through state and outputs the image data input from the shading correction circuit 17 to the post-processing circuit 18 without performing the line number conversion process.

On the other hand, when it is determined in step S3 that the resolution that matches the target resolution R does not exist in the table memory 195, the CPU 191 sets the closest default resolution that is equal to or higher than the target resolution R.
The table memory 195 is searched (step S7).
Next, the CPU 191 acquires the feed speed V in the sub-scanning direction corresponding to the resolution retrieved in step S7 from the table memory 195 (step S8). Next, CPU1
The 91 sets the target resolution R and the resolution retrieved from the table memory 195 in the DSP 197 (step S9). At this time, the target resolution R is set in the DSP 197 as the above-mentioned resolution K, and the resolution retrieved from the table memory 195 is set as the above-mentioned resolution J.
Next, the CPU 191 sets the feed speed V in the sub-scanning direction acquired in step S8 in the motor driver 196 (step S5). Thereby, the motor 8 moves the document 1 in the sub-scanning direction at the set speed. After that, the image reading device scans (reads) the document 1 (step S6). At this time, the CPU 191 temporarily stores the image data input from the shading correction circuit 17 in the RAM 193. And DSP1
97 performs line number conversion processing on the image data stored in the RAM 193 according to the above equations (10) to (12), and then outputs it to the post-processing circuit 18. As a result, the image data of the document 1 read at the default resolution is converted into the image data of the target resolution and output without causing color misregistration.

The three primary color line signals whose resolutions have been converted by the RGB position shift correction circuit 19 without color shift are subjected to predetermined post-processing by the post-processing circuit 18, and then the three primary color line signals S _B and S _G. , S _R and stored in the buffer memories M _B , M _G , and M _R of FIG. 11, respectively. FIG. 11 shows, as an example, the stored data when N = 3. In this case, the buffer memory M _G
By reading the line signal with a delay of 3 lines from _B and with a delay of 3 lines from the buffer memory M _G from the buffer memory M _R , read information at the same position can be obtained at the same timing.

[0055]

According to the first aspect of the present invention, the color original is once read at a discrete resolution without color misregistration, and is converted into the image data of the target resolution by the line number conversion processing later. The desired resolution can be continuously changed without causing color shift.

According to the second aspect of the present invention, when the target resolution is input, a resolution that is equal to or higher than the target resolution and is closest to the target resolution is searched from the table memory, and the sub-scanning corresponding to the searched resolution is performed. Since the relative movement speed between each linear image sensor and the color original is adjusted based on the relative movement speed in the direction, the procedure for determining the relative movement speed can be simplified. Therefore, arithmetic processing becomes easy and high-speed processing can be realized.

[Brief description of drawings]

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

FIG. 2 is a block diagram showing a more detailed configuration of an RGB position shift correction circuit 19 in FIG.

FIG. 3 is a diagram showing an optical system of the image reading apparatus shown in FIG.

FIG. 4 is a perspective view showing a detailed configuration of the multi-row image sensor 13 shown in FIG.

FIG. 5 shows a reading state when color misregistration occurs.

FIG. 6 shows a reading state when color misregistration does not occur.

FIG. 7 shows a change state of the resolution when the value of N is changed when the resolution when N = 1 is set to 254 dpi.

8 is a diagram showing an example of data stored in a table memory 195 of FIG.

FIG. 9 is an example of a relationship between an input data string and an output data string when image data read at a resolution J [dpi] is converted into image data having a target resolution K [dpi] by a line number conversion process. FIG.

10 is a flowchart showing the operation of the embodiment of FIG.

11 is a diagram showing the configuration of a buffer memory for adjusting the timing of the three primary color line signals obtained by scanning the original 1. FIG.

FIG. 12 is a diagram showing an example of a configuration of a conventional document reading apparatus.

FIG. 13 is a diagram showing another example of the configuration of a conventional document reading apparatus.

14 is a diagram showing a document reading state of the document reading device shown in FIG.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Original document 2 ... Lens 6 ... Original document table 8 ... Motor 9, 10 ... Illumination light source 11 ... Mirror 13 ... Multi-row image sensor 19 ... RGB misalignment correction circuit 191 ... CPU 192 ... ROM 193 ... RAM 194 ... Input device 195 Table memory 196 Motor driver 197 DSP 131 Light receiving cells 132B, 132G, 132R Linear image sensor 133B, 133G, 133R Color filter

Claims (2)

[Claims] 1. A plurality of linear image sensors extending in the main scanning direction are arranged in parallel in a plurality of rows, and each linear image sensor and a color original are relatively moved in the sub scanning direction while each linear image is being moved. An image reading device that scans each color component of an image of a color original by a scanning line in order, and includes a lens, and magnifies the image of the color original by a predetermined size to form a light-receiving surface on each of the plurality of linear image sensors. Optical system for forming an image, input means for inputting a desired resolution, a value obtained by dividing a reading position shift amount of the plurality of linear image sensors on the color original in the sub-scanning direction by a scanning line pitch on the color original (divisor value) N) is an integer value, and a determining means for determining the relative moving speed in the sub-scanning direction between each linear image sensor and the color original, Based on the relative movement speed determined by the determination means, speed adjusting means for adjusting the relative movement speed in the sub-scanning direction between each of the linear image sensors and the color original, by reading the color original A storage unit that stores the obtained image data, and a line number conversion unit that generates the image data of the target resolution input by the input unit by converting the line number of the image data stored in the storage unit. An image reading device provided. 2. A table memory storing a plurality of sets of relative moving speeds in the sub-scanning direction and corresponding resolutions such that the divisor value N is an integer value while the optical magnification in the optical system is fixed. Further comprising: when the target resolution is input from the input unit, the determining unit searches the table memory for a resolution that is equal to or higher than the target resolution and is closest to the target resolution, The image reading apparatus according to claim 1, wherein a relative moving speed in the corresponding sub-scanning direction is determined as a moving speed given to the speed adjusting unit.