JPH0451672A - Color reader - Google Patents

Abstract

PURPOSE:To correct the reading error of each color component by correcting the read result of each corrected object area by calculating a filter to cancel the deterioration of a read-out picture based on a point spread function showing the deterioration of the read-out picture by a reading means in respect of each of plural corrected object areas. CONSTITUTION:A line on an original 101 is divided into plural corrected object areas, and the filter to cancel the deterioration of the read-out picture by the reading means 110 is calculated in accordance with each of these corrected object areas. Then, the read result of each color component is corrected by executing the convolution integral of the read result of each color component by the reading means 110 and the corresponding filter. Information related to the error due to all factors caused by reading the picture of the corresponding corrected object area by the reading means 110 is included in the point spread function corresponding to each corrected objec area. Thus, the reading error of each color component can be corrected as considering various factors of the error.

Description

[Detailed Description of the Invention] [Table of Contents] Overview Field of Application Prior Art Problems to be Solved by the Invention Means for Solving the Problems Action Examples Effects of the Invention [Summary] Correction of a read image The purpose of this color reading device is to correct reading errors of each color separation caused by various factors. A reading means separates and reads a color image into multiple color components, and a line on the document is divided into multiple correction target areas, and based on the reading result of each color component by the reading means, these correction target areas are read. Corresponding to each, a point spread function calculation means calculates a point spread function representing deterioration of an image read by the reading means for each color component, and a point spread function calculating means for calculating a point spread function representing the deterioration of an image read by the reading means, and a point spread function calculation means for calculating a point spread function representing the deterioration of an image read by the reading means, and a point spread function calculating means for calculating a point spread function representing the deterioration of an image read by the reading means, and a point spread function calculating means for calculating a point spread function representing the deterioration of an image read by the reading means, A filter calculation means for calculating a filter coefficient representing a filter that cancels out the deterioration of the read image of and a correction means for selecting a filter for the corresponding correction target area from the filter and convolving and integrating the reading result for each introduced color component with the filter for each selected color component and outputting the result as the reading result for each color component. do.

[Industrial application field]

The present invention relates to a color reading device, and particularly to a color reading device that corrects a read image.

Color reading devices are used as image input devices for facsimile machines and computers.
It is known to measure and read a color image as the intensity distribution of R (red), G (green), and B (blue) components.

In addition, such color reading devices are required to read color images with high precision, and if the read color image is degraded, it is necessary to correct the image.
There is a demand for outputting high quality read images.

[Conventional technology]

FIG. 6 shows the configuration of a conventional filter switching type color reading device.

A linear portion (hereinafter referred to as a reading line) 613 on the original 611 is illuminated by an illumination device 612 such as a fluorescent lamp. The reflected light from this reading line 613 is imaged onto a sensor (for example, a CCD line image sensor) 615 by a lens 614.

In addition, filters 621r and 621g that transmit red (R) component, green (G) component, and blue (B) component, respectively,
621b is selectively inserted into the optical path from the document 611 to the sensor 615 (FIG. 6 shows the filter 621g inserted into the optical path). In such a color reading device, the color components of the light reaching the sensor 615 are changed by changing the filter inserted in the optical path, and the intensity distribution of each color component of the light reflected from the reading line 613 is measured in a time-sharing manner. The color image on the reading line 613 is read.

Thereafter, the document 611 is moved by a predetermined length in the sub-scanning direction (indicated by arrow A in the figure) perpendicular to the reading line 613, and the same process is repeated to read a two-dimensional color image.

As another example of a color reading device, FIG. 7 shows the structure of a contact type color reading device.

A reading line 713 on the original 711 is illuminated by an illumination device 712, and the reflected light from this reading line 713 has a diameter of 1 m! CCD
The image is formed on a sensor 715 such as a line image sensor.

Here, the length of each element of the sensor 715 described above in the arrangement direction is equal to the width of the document 711. Furthermore, as shown in FIG. 8, each pixel on the reading line 713 has
Three sets of elements on the sensor 715 correspond to each other, and each of these three elements is covered with a filter that separates the R, G, and B components, respectively. Further, the output of the set of these three elements is the reading line 7 of the original 711.
It is output as a reading result of each component of each pixel on 13.

(Problems to be Solved by the Invention) By the way, in the conventional method described above, a color image is first decomposed into R, G, and B components and the intensity distribution of each component is measured, and then these measurements are performed. By combining the results, a color image is obtained.

Here, the measurement results for each color separation include errors due to aberrations of the imaging optical system such as the lens 614 and the rod lens array 714, and these errors generally differ for each color separation. ing.

For example, in the filter switching type color reading device shown in FIG.
3 is focused on the sensor 615. By the way, in such a reduction optical system, the imaging position differs for each separated color, especially due to the chromatic aberration of the lens 614. Further, since reading is performed for each separated color with a time difference, when reading is performed while moving the document 611 in the sub-scanning direction, the reading position itself is different for each separated color.

In the contact type color reading device shown in FIG. 7, each pixel on the reading line 713 is
Three elements correspond to each other, and the reading positions corresponding to each of these three elements differ by 1/3 pixel.

For this reason, there is a problem in that a color shift occurs in a color image obtained by combining the measurement results of each separated color, and the quality of the read image deteriorates.

The method to prevent color shift as described above is to investigate the error factors for each color separation (chromatic aberration, etc.), and based on these factors, analytically evaluate the error for each color separation. There are ways to correct the measurement results. However, since a color reading device is composed of a large number of components, there are a large number of factors, and it is difficult to analytically determine the value by considering all the factors.

SUMMARY OF THE INVENTION An object of the present invention is to provide a color reading device that corrects reading errors of each separated color caused by various factors.

[Means to solve the problem] FIG. 1 is a block diagram of the principle of the present invention.

(In the day map of j-n 11, the reading means 110 in the invention of claim 1 has a line image sensor 111 in which a plurality of light receiving elements are arranged in a row, and a plurality of light receiving elements corresponding to each light receiving element of the line image sensor 111 The color image on the original 101 is separated into a plurality of color components and read for each line made up of pixels on the original 101.

The fish bed gap function calculating means 121 divides the line on the original 101 into a plurality of correction target areas, and performs reading corresponding to each of these correction target areas based on the reading result of each color component by the reading means 110. A gradation function representing the deterioration of the image read by the means 110 for each color component is calculated.

The filter calculation means 131 is the fish bed density function calculation means 12.
1, a filter coefficient representing each of the filters for canceling out the deterioration of the read image of each color component in each of the plurality of correction target regions is calculated.

The correction means 140 selects a filter corresponding to the corresponding correction target area from among the filters of each color component introduced from the filter calculation means 131 according to the reading result of each color component of the original 101 by the introduced reading means 110. Then, the result of convolving and integrating the reading result of each introduced color component with the filter of each selected color component is output as the reading result of each color component.

In the invention of claim 2, the reference image 122 includes a line image sensor 111 in each of the plurality of correction target areas.
At least one pattern corresponding to one light receiving element is arranged.

Overall, in the color reading device according to the invention of claim 1, the fishbed density function calculation means 121 is configured to
Based on the reading result of the reference image 122 obtained by , the fish bed density function of each color component corresponding to each of the plurality of correction target regions is calculated.

In the invention of claim 3, the correction means 140 includes a plurality of correction processing means 141 corresponding to each of the plurality of color components.
It is configured with

Further, each of the plurality of correction processing means 141 corresponds to each of the plurality of correction target regions, and includes a plurality of filter storage means 142 that stores filter coefficients representing filters of the corresponding color components, and a plurality of filter storage means 142 corresponding to the plurality of lines. An output means 143 to which results are input continuously and outputs the reading results corresponding to each pixel of a plurality of lines in parallel; and a holding means into which the output of the output means 143 is input and holds a predetermined amount of the reading results of the plurality of lines. 144, a determining means 145 for determining a correction target area corresponding to the reading result according to the reading result held in the holding means 144; selection means 146 for selecting the filter storage means 142 to be used;
and a multiplication means 14 for multiplying the filter coefficient held in the filter storage means 142 selected by the selection means 146 by the corresponding reading result held in the holding means 144.
7, and an arithmetic means 148 for calculating the sum of the multiplication results by the multiplication means 147.

Overall, the output of the calculation means 148 is corrected by the correction processing means 14.
1 output.

-〇LL过郎J ロコ 膜子所 In the invention of claim 4, the synthesis means 151 has a filter for each color component corresponding to each of the plurality of correction target regions calculated by the filter calculation means 131 and an image processing function. Generate a synthetic filter by convolving and integrating with other filters.

Overall, the composition filters generated by the composition means 151 are introduced into the correction means 140 of the color reading device according to the first aspect of the invention as filters for each color component corresponding to each correction target area.

[For production]

In the invention of claim 1, the color image on the original 101 is divided into a plurality of color components for each line consisting of each pixel corresponding to each light receiving element of the line image sensor 111 of the reading means 110. It is decomposed and read.

The point spread function calculation means 121 calculates the above-mentioned original 10.
The line on 1 is divided into a plurality of correction target areas, and the point spread function calculating means 121 calculates a value corresponding to each of the above correction target areas based on the reading result of each color component by the above-mentioned reading means 110. Therefore, the reading means 1
A point spread function representing the deterioration of the read image of each color component by 10 is calculated. Based on this point spread function, the filter calculation means 131 calculates filter coefficients representing filters that cancel out the deterioration of the read image of each color component in each of the plurality of correction target areas described above.

According to the reading result of each color component of the original 101 by the introduced reading means 110, the correction means 140 selects each color corresponding to one of the plurality of correction target areas from among the filters introduced from the filter calculation means 131 described above. A component filter is selected. Further, the reading result of each introduced color component and the filter of each selected color component are convoluted and integrated by the correction means 140, and outputted as the reading result of each color component.

In the invention of claim 1, a line on the original 101 is divided into a plurality of correction target areas, and a filter for canceling the deterioration of the image read by the reading means 110 is calculated corresponding to each of these correction target areas. be done. Furthermore, the reading results of each color component are corrected by performing convolution integration between the reading results of each color component by the reading means 110 and the corresponding filter. The point spread function corresponding to each correction target area described above includes information regarding errors caused by all factors that occur when the reading means 110 reads the image of the corresponding correction target area. Therefore, by correcting the reading results using the filter described above,
It becomes possible to correct the reading error of each color component by considering various error factors.

In the invention of claim 2, at least one pattern corresponding to one light receiving element of the line image sensor 111 is arranged in each correction target area of the reference image 122. Further, this reference image 122 is read by the reading means 110, and based on the reading result of each color component at this time, the point spread function calculation means 121 calculates the point spread of each color component corresponding to each of the plurality of correction target areas. The function is calculated.

Here, the point spread function indicates the spread of a read image when a point light source placed at the position of the document 101 is read by the reading means 110. Further, the pattern of the reference image described above corresponds to a point light source placed at the position of the document 101.

In the invention of claim 2, the point spread function can be easily determined based on the reading result when the reference image 122 in which the pattern corresponding to the point light source is placed in each correction target area is read. . Furthermore, by correcting the reading results using a filter calculated based on this point spread function, it becomes possible to correct the reading errors of each color component by taking various error factors into consideration.

iii) -;'13 - In the invention of claim 3, the correction means 140 is composed of a plurality of correction processing means 141 corresponding to each of a plurality of color components.

A plurality of filter storage means 142 of each correction processing means 141
The filter coefficients of the corresponding color components of the respective corresponding correction target regions are stored.

Further, the reading results of a plurality of lines of corresponding color components are continuously input to the output means 143, and the reading results corresponding to each pixel of these lines are outputted and held in parallel by the output means 143. The holding means 144 holds a predetermined amount of the reading results of the plurality of lines described above.

According to the reading result held in the holding means 144, the determination means 145 determines the correction target area corresponding to the above-mentioned reading result, and the selection means 146 selects the above-mentioned plurality of filters according to the determination result. Storage means 142
The corresponding filter storage means 142 is selected from the list.

Furthermore, the multiplication means 147 allows the above-mentioned selection means 146
The filter coefficient stored in the filter storage means 142 selected by is multiplied by the corresponding reading result held in the holding means 144, and the calculation means 148 calculates the sum of the multiplication results by the multiplication means 147. , the output of this calculation means 148 is taken as the output of the correction processing means 141.

In the invention of claim 3, the reading result of each line by the reading means 110 is continuously supplied to the correction processing means 141, so that the reading result and the filter are convolved and integrated. In parallel with the reading operation, it is possible to perform correction processing on the reading results, and it is possible to speed up the correction processing.

In the invention of claim 4, the synthesis means 151 convolves and integrates the filter of each color component corresponding to each correction target area calculated by the filter calculation means 131 with another filter having an image processing function. is performed to generate a synthesis filter. Further, this synthesis filter is introduced into the correction means 140 as a filter for each color component corresponding to each correction target area described above.

Here, the synthesis filter generated as described above has both the function of canceling out the deterioration of the read image of each color component and the image processing function.

Therefore, in the invention of claim 4, the correction means 140 corrects the reading results of each color component using the synthesis filter generated by the synthesis means 151 described above, thereby correcting the reading error of each color component. At the same time, it becomes possible to perform image processing using other combined filters.

〔Example〕

Hereinafter, embodiments of the present invention will be described in detail based on the drawings.

FIG. 2 shows the configuration of a color reading device according to an embodiment of the present invention.

FIG. 5 shows the configuration of a correction processing section according to another embodiment of the present invention.

Here, the correspondence between FIG. 1 and the embodiment will be explained.

The original 101 corresponds to the original 201.

The reading means 110 corresponds to the image reading section 210.

The line image sensor 111 corresponds to a CCD line image sensor (CCD sensor) 212.

The point spread function calculation means 121 corresponds to the PSF calculation section 251.

Reference image 122 corresponds to reference image 202.

The filter calculation means 131 corresponds to the filter calculation section 252.

The correction means 140 includes correction circuits 253r, 253g, 25
Corresponds to 3b.

The correction processing means 141 includes correction circuits 253r, 253g,
Corresponds to 253b.

The filter storage means 142 includes a filter coefficient register 27
Corresponds to 2.

The output means 143 includes the line memory 271. ．． 2712 and memory controller 274.

The holding means 144 includes latches 273A, . . . , 273,
corresponds to

The determining means 145 corresponds to the selector control section 276.

The selection means 146 corresponds to the selector 275.

Multiplying means 147 corresponds to multiplier 281.

The calculation means 148 includes a determination circuit 282 and an adder 283.2.
84 and ROM285.

The synthesizing means 151 corresponds to a filter storage section 551 and a filter synthesizing section 552.

The configuration and operation of the embodiment will be described below assuming that the above-mentioned correspondence exists.

In FIG. 2, 201 represents a document, 210 represents an image reading section, and 211 represents a color reading mechanism. As this color reading mechanism 211, for example, the above-mentioned filter switching type color reading mechanism (see FIG. 6) or contact type color reading mechanism (see FIG. 7) may be used.

In this image reading section 210, a color reading mechanism 211
CCD line image sensor (hereinafter referred to as CC) equipped with
A D sensor (referred to as D sensor) 212 measures the intensity distribution of each of the R, G, and B components of the color image in the reading line on the original 201. Further, the analog output of this CCD sensor 212 is transmitted to an analog-to-digital (A/D) converter 21.
3, and the measurement results of the intensity distributions of the R, G, and B components are converted into digital values and output as the reading results by the image reading section 210.

Thereafter, the image reading unit 210 scans the original document in a direction (sub-scanning direction) perpendicular to the above-mentioned reading line direction (main-scanning direction).
01 by a predetermined distance and repeating the above-described reading operation, a two-dimensional color image on the document 201 is read.

After performing shading correction processing, color conversion processing, and correction processing on the images of each component read in this way, the images of each component are combined and displayed on a display device (not shown).
It is designed to output to etc.

Here, shading refers to the appearance of shading in an image, and refers to the characteristics of the imaging optical system (see FIGS. 6 and 7) provided in the color reading mechanism 211 (reduction in the amount of peripheral light due to the so-called cos ' law). ) etc. The shading correction unit 231 performs shading correction processing by normalizing the reading result of the original 201 based on the reading result when a white plate or the like is read by the image reading unit 210.

Further, the color conversion unit 241 performs color conversion processing to convert the reading results of each color component so that the colors of the image on the original 201 are reproduced in the signal system (for example, NTSC signal system) in the display device.

The correction processing unit 250 that performs image correction processing will be described below.

Here, the R, G, and B components of the read image output by the color reading device y ], (x + (i is R, G).

B) is the R of the image on the manuscript (hereinafter referred to as the original image),
G, B component as xi(x+y), R of color reading device,
Point spread function of G and B components (Po1nt 5pread
Function: P S F ) as hi(x, y
), it is expressed by equation (1).

2, the errors included in the reading results of the R, G, and B components are expressed by the point spread functions of the R, G, and B components.

Further, if fi (x, y) is a correction filter that cancels image deterioration caused by the reading process in the image reading unit 210, the restored image Ri (x, y) is expressed as shown in equation (2).
It is expressed as a convolution integral between the correction filter fi(x,) and the read image yt(x, y).

9 i(x, y) = f i(x, i* y i(x
, I...(2) Therefore, the above formula (1),
2 Based on (2), a correction filter fi (x, y) is determined, and by correcting the read image yi (x, i) using this correction filter fi (x, y), all the The reading results of each color component can be corrected in consideration of error factors.

Here, since the aberration of the coupling optical system used in the color reading mechanism 211 generally changes continuously depending on the position of the object point, the PSFhi ( x, y) also change continuously.

In this way, PSF h corresponding to each pixel of one line
When i(x, y) changes continuously, PSFhi(x
, y) is P of pixels in a predetermined range including the above-mentioned pixels.
It is considered that S F hi(x,

Therefore, one line is divided into a plurality of correction target areas, and P S F h i (x, y) corresponding to each pixel selected from each of these correction target areas is calculated,
Using these P SF h i (x, y), PSFhi (xy
) may be approximated.

The configuration and operation of the correction processing section 250 will be specifically described below.

Prior to reading the original 201, the correction processing unit 250 includes a PSE calculation unit 251 that calculates P S F h i (x, y) of each color component corresponding to a plurality of correction target areas (described later), and Prior to reading the PSF
P of each correction target area calculated by the calculation unit 251
A filter calculation unit 252 that calculates a correction filter fi (x, i) for each color component corresponding to S F h i (x, y), and a correction filter fi (x, y) for each color component calculated by the filter calculation unit 252. Based on the image reading unit 2
The image forming apparatus 10 is configured to include correction circuits 253r, 253g, and 253b for correcting the reading results of the R2O and B color components of the original 201 according to No. 10.

First, the calculation process of PSFhi(xy) by the PSF calculation unit 251 will be explained. Here, using a reference image corresponding to the reference image 122 of claim 2, P S F
The case of calculating hi(x,i) will be explained.

For example, as a reference image 202 corresponding to the reference image 112 of claim 2, one line is divided into three correction target areas (indicated by dotted lines in the figure) as shown in FIG. At the center of each target area, a CCD
A reference image 202 is prepared in which a black pattern (in the case of a white background) of a size corresponding to one pixel of the sensor 212 is arranged. Note that there are no limitations on the number and position of the black patterns described above, and a reference image having a white pattern on a black background may be used.

For example, when starting up the color reading device, the operator may set the above-mentioned reference image 202 at the position of the document 201 and drive the image reading unit 210 to read it. At this time, the result of processing the reading result of each reading line by the image reading unit 210 by the shading correction unit 231 and the color conversion unit 241 is obtained as the reading data of each R, G, and B component corresponding to each reading line. The signal is supplied to the PSF calculation unit 251 described above.

Here, each of the patterns of the reference image 202 described above corresponds to an ideal point light source placed on the reading line. Therefore, the read image yi (x, y) when the reference image 202 described above is read by the image reading unit 210
The part corresponding to the image of each pattern in is expressed by the above formula (
This corresponds to the case where the original image xi(X+y) is set as a δ function in 1).

Therefore, the PSF calculation unit 251 calculates the PSF h i (x, y) of the corresponding correction target area based on the read data of the portion corresponding to the image of the pattern corresponding to the point light source described above.
All you have to do is calculate.

For example, the PSF calculation unit 251 first calculates, as read data of the image portion corresponding to each of the patterns corresponding to the point light sources described above, an area consisting of the corresponding pattern and each pixel around it (for example, as shown in FIG. 3). 3× shown in b)
The read data di(
j, k). Here, di(j, k) is the coordinate (Lk) (Lk=-1), with the position of the pattern corresponding to the point light source as the origin of the coordinates.

The reading data of each color component corresponding to a pixel of 0.1) is shown.

The PSF calculation unit 251 extracts d as described above.
Applying the normalization process shown in equation (3) to i(j, k),
P S F h i (x, y
) can be calculated by calculating the value h iJl.

ΣΣd i(x, y) = 1... (3)
In addition, 3 rows 3 consisting of the values calculated as described above
The column matrix is P S F h corresponding to each correction target area.
It is called PSF coefficient Hi representing i(x,y).

In addition, in an ideal case where the color reading mechanism 211 has no factors that degrade images, the CC for each R, G, and B component is
The output of the D sensor 212 also becomes a δ function. Assuming this, h'x(x+y) obtained by convolution-integrating the reading result di(x, y) and the δ function and the reading result di(x, y) described above satisfy equation (4). To satisfy this, PSFhi(x, i) which minimizes the mean square error ε2 of equation (5) may be estimated by the least squares method.

di (x, y) = hi (x, y) * h'i (x, y)
+++ (4) ε”=E((di(x, y)
-h'i (x, y))21 - (5) Here, the formula (
In 3), operator * indicates a convolution integral, and in 2(4), operator E indicates an expected value operator.

In this way, the PSF calculation unit 251 calculates the PSF coefficient Hi representing the PSFhi (x+y) of each color component for each correction target area.

Next, the calculation process of the correction filter correction filter fi (x, y) by the filter calculation unit 252 will be explained.

Here, a Wiener filter is known as a correction filter for restoring a degraded read image.

In the filter calculation unit 252, the correction filter fi(x
, y), the original image xi(
Define the mean square error ε2 between the restored image 9 i(x, y) and the restored image 9 i(x, y), and find the correction filter fi(x, y) to minimize this mean square error ε2. .

E2=E ((xi(x,y)−fatheri(x,y))”)
-(6) Note that the correction filter fi(x, y) is not limited to a Wiener filter, and the filter calculation unit 252 may calculate an inverse filter or the like.

The filter calculation unit 252 uses the above-mentioned formula 2 (1) 1 (2)
P S F h i (x, y) corresponding to each correction target area described above was calculated based on equation (6).
Corresponding to each of ×3 pixels, the correction filter fi(x
, y) value fiJk may be calculated. Further, as shown in FIG. 4, the filter calculation unit 252 converts the correction filter values fiJll calculated as described above into a 3-by-3 matrix, and converts this matrix into a 3-by-3 matrix for each color component corresponding to each correction target area. It is designed to be output as a filter coefficient Fi.

In addition, in the PSF calculation unit 251, the above-mentioned P S
There is no limit to the range in which the value hiJk of F h i (x, y) is calculated, and read data corresponding to nXn pixels is extracted as read data corresponding to an image corresponding to a point light source, and based on these read data, , for a range of nXn pixels,
What is necessary is to find the value h iJl of P S F h i (x, y). In this case, the filter calculation unit 252 calculates filter coefficients Fi for n rows and n columns.

As described above, prior to reading the original 201, p S F h of each color component corresponding to each correction target area is determined.
Based on i(x,y), the filter coefficients Fr, Fg, and Fb of each RGB component representing the correction filter fi(x,y) of each color component corresponding to each correction target area are calculated, and the corresponding correction circuits are calculated. It has been introduced in 253r, 253g, and 253b. However, in the figure, the filter coefficient Fg
, Fb are introduced into the correction circuits 253g and 253b, and the lines are omitted.

The configurations of the correction circuits 253r, 253g, and 253b and the correction processing of the reading results of the original 201 by these correction circuits 253r, 253g, and 253b will be described below.

The correction circuit 253r includes two line memories 271 . ．． 271□, and three filter coefficient registers 2 that hold R component filter coefficients Fr corresponding to each of the three correction target regions described above.
72a, 272b, 272c, and nine latches 273A, . . . , 2, each holding one pixel worth of read data.
73. and the above-mentioned line memory 271. ．． 271□, and each component r of the filter coefficient Fr held in each of the three filter coefficient registers 272a, 272b, and 272c mentioned above.
a selector 275 that outputs either jh, and a selector control unit 2 that controls the selection operation by this selector 275.
76 and the nine latches 273A, . . . , 27 described above.
31 and an arithmetic circuit 277 that performs an arithmetic operation equivalent to a convolution integral of the read data held in the filter 31 and each component of the filter coefficient Fi supplied from the selector 275.

The correction circuits 253g and 253b are the correction circuit 25 described above.
It has the same configuration as the 3r, and has three filter coefficient registers corresponding to each correction target area (
(not shown) hold the corresponding G component filter coefficient Fg and B component filter coefficient Fb, respectively.

The two line memories 271 mentioned above. ．． 271□ is composed of a shift register, etc., and is configured to hold one line of read data and sequentially output the held read data in response to the input of the next line's read data. ing.

In addition, the latches 273A, 273. ．． 273c are connected in series, the output of latch 273C is connected to latch 2738, and the output of latch 273. The outputs of are sequentially input to the latch 273A. Latch 213D, 273t, 2
73. and latch 273r, 273. ．． 273□
are also connected in series.

Further, the read data of the R component of each line is continuously supplied from the color conversion unit 241, and the above-mentioned line memory 271.
It is also input to the latch 273. The output of this line memory 2711 is the line memory 2
71□ and the latch 2731, and the output of the line memory 2712 is input to the latch 273°.

The memory controller 274 is configured to output an output timing signal S7 synchronized with the timing at which read data is input from the color conversion section 241, and the line memories 2711.271□ output timing signals S7 in accordance with this output timing signal SA. , the stored read data is output. Therefore, the first two lines of read data are stored in the line memory 271. ．． 271□, and when the input of the third line read data is started, the latch 2
In synchronization with the timing at which the read data corresponding to each pixel on the third line is input to the latch 273F,
The first line is input to the latch 273°. In this way, three lines of read data that are continuously supplied are transmitted in parallel to the latches 273c, 273F, 273 . is input.

As described above, nine latches 273. .

..., the read data held in 273□ is stored in latch 2.
73E, and these read data correspond to each component rjl+ of the filter coefficient Fr described above. Below, the nine latches 273A,...
The read data held in 2731 is converted to yr(j,k)(
Lk=-1, 0, 1).

The selector control unit 276 also includes a counter 261 that performs a counting operation in accordance with the above-mentioned output timing signal Sr,
The count value of the counter 261 is based on the address register 262 that stores the address indicating the dividing pixel of each correction target area mentioned above, the count value of the counter 261 mentioned above, and the address stored in the address register 262. The determination unit 263 is configured to include a determination unit 263 that determines a filter coefficient Fr corresponding to a pixel.

For example, if one line is divided into three correction target areas ■, ■, and ■ as shown in FIG. 3(a), the correction target area ■ and the correction target area What is necessary is to store an address A indicating a pixel at the break between the two, and an address A2 indicating a pixel at the break between the correction target area (2) and the correction target area (2). In addition, the above-mentioned 3
Filter coefficients Fr+, Fr2 . Fr, are stored in filter coefficient registers 272a, 272b, and 272c, respectively.

Further, the determination unit 263 uses the count value C of the counter 261 and the addresses A+ and A stored in the address register 262.
z, and based on the comparison result, determine the correction target area that includes the pixel indicated by the count value C, and instruct the selector 275 to select the corresponding filter coefficient Fr.

For example, as described above, nine latches 273A,
...When read data corresponding to 3x3 pixels centered on the second pixel of the second line is stored in 2731,
It is assumed that the count value C of the counter 261 is smaller than the address A1. In this case, the determination unit 263 instructs selection of the filter coefficient Fr corresponding to the correction target area (2). In response, the selector 275 selects the filter coefficient Fr corresponding to the correction target region (2), and each component rjk of the filter coefficient Fr is sequentially introduced into the multiplier 281 of the arithmetic circuit 277.

This multiplier 281 includes latches 273A, .

The read data yr (j, k) held in The selected filter coefficient Fr is multiplied by the corresponding component rjk. However, in the figure, latches 273a, 2
Lines indicating the supply of read data from 73c, 273t 273F2738.2731 to the multiplier 281 are omitted.

The multiplication result by this multiplier 281 is supplied to a determination circuit 282, which determines whether the multiplication result described above is a positive value or a negative value, and is determined to be a positive value. If it is a negative value, it is input to an adder 283, and if it is a negative value, it is input to an adder 284.

Each of these adders 283 and 284 is configured to calculate a total sum by sequentially adding up input numerical values. Therefore, the adder 283 calculates the sum of the multiplication results having a positive value, and the adder 284 calculates the sum of the multiplication results having a negative value.

Further, the ROM 285 of the arithmetic circuit 275 stores in advance the results of adding the sum of positive multiplication results and the sum of negative multiplication results, corresponding to each combination, and the adder 283, The output of 284 is used as an address input, and the corresponding addition result is outputted as correction data corresponding to the read data held in latch 273.

In this way, the arithmetic circuit 277 operates the latch 273
The read data yr(j
, k) and the filter coefficient Fr selected by the selector 275, the calculation shown in equation (7) is performed to calculate the correction data r(0,0) corresponding to the center pixel held in the latch 273E. be done.

Thereafter, the read data corresponding to the next pixel of each line is input to the latches 273c, 273F, 27L, and the read data held in the latches 2738, 273E2738 and latches 213c, 273', 2731 are sequentially shifted. Ru. In addition, in response to this, the count value of the counter 261 of the selector control unit 276 is incremented by "rl".

By repeating the above-described operation for each pixel included in the correction target area (2), the read data corresponding to each of these pixels is sequentially corrected based on the filter coefficient Fr.

Thereafter, when the count value C of the counter 261 becomes equal to the address A1 described above, the determination unit 263 instructs selection of the filter coefficient Frz corresponding to the correction target area (■). In response, the selector 275 selects the filter coefficient Fr stored in the filter coefficient register 272b.
z is selected, and in the same way, the read data corresponding to each pixel included in the correction target area
2 is sequentially corrected based on.

Further, after that, when the count MC of the counter 261 becomes equal to the address A2 described above, the determination unit 263 instructs selection of the filter coefficient Frff corresponding to the correction target area (2). In response, the selector 275 selects the filter coefficient Fr3 stored in the filter coefficient register 272C, and similarly, the correction target area
The read data corresponding to each pixel included in the image data is sequentially corrected based on the filter coefficient Fr+.

In this way, the correction process for one line of read data is performed, and by repeating this correction process for each line, the two-dimensional read image of the R component is S F hr(x
, y), the correction filter fr(x, y)
is corrected using

Similarly, the read data of the G component and B component is corrected.

As described above, P S F h i (j, k) calculated corresponding to each correction target area is
1 includes all information regarding errors included in the read data corresponding to each pixel of the corresponding correction target area read by No. 1. Therefore, the correction filter f r(
x, y), f g(x, y), f b(x, y
) by correcting the read data, it becomes possible to correct the color shift of the read data by taking into account various error factors that differ for each color component.

As a result, since the color reading mechanism 211 uses a reduction optical system, even if the aberration changes greatly when each pixel of one line is used as an object point, such changes in aberration can be taken into account. As a result, color shift in read data can be corrected, and a high quality read image can be output.

In addition, since the correction filters corresponding to multiple pixels included in each correction target area are each approximated by one correction filter, the correction filter The amount of hardware required for storage etc. can be reduced.

Further, as shown in FIG. 3(a), by preparing a reference image 202 in which a pattern corresponding to a point light source is arranged in each correction target area, the PSF calculation unit 251 uses this reference image 202 as an image read. Based on the read data di(j, k) read by the unit 210, P S
It becomes possible to calculate F h i (j, k).

In addition, the line memory 271. ．． 271□ and filter coefficient registers 272a and 272b corresponding to each correction target area
, 272c and nine latches 273A..., 273. and a selector 275 and an arithmetic circuit 277.
Correction circuits 253r and 253 corresponding to each of the B components
correction circuits 253r, 253g, 25 which continuously read the R, G, and B component read data of each line.
3b. As a result, convolution and integration processing is performed between the read data of each color component and the corresponding correction filter, making it possible to perform correction processing on the read image in parallel with the reading operation of the original 201 by the image reading unit 210. High-quality scanned images can be output at high speed.

In the above-mentioned embodiment, the case where the reading line is divided into three correction target areas is explained, but there is no limit to the number of correction target areas, and the reading line is divided into a plurality of correction target areas, and each correction target area is P S F hi(x, y) of each color component corresponding to the target area and filter fi(x, y
) can be applied. For example, if l line is divided into n correction target areas,
Each of the above-mentioned correction circuits 253r, 253g, and 253b is configured to include n filter coefficient registers 272. For example, in the reference image 202, as shown in FIG. 3(C), patterns corresponding to point light sources are densely arranged at both ends of the reading line, and the width of the correction target area at both ends of the reading line is The area may be divided into smaller areas than the correction target area at the center. Furthermore, for example, the third
As shown in Figure (d), a plurality of patterns are arranged in each correction target area, and based on the read data of these image parts, P S F h i(
x, y), and calculate these P S F h i(x, y
), P S F corresponding to each correction target area
h i (x, y) may be calculated.

In addition, the read image 52i (
As shown in (8), the output image zt(x, V) may be obtained by applying filters gx(x, y) having various image processing functions.

zi(x,y)-gi(x,y)*Ri(x,y)
++ (8) Here, the restored image Ri (x, y)
is the correction filter fi(x,
y) and the read image yt(x, y), equation (8) can be transformed as shown in equation (9).

z i (x, y) = g i (x, y) * f i (
x, y) * y i (x, y) - f'i (x, y)
* y i (x, y) ... (9) Therefore, the correction filter fi (x, y) and the above-mentioned filter gt
(x, y) and the synthesis filter white (x, y) is convoluted and integrated.
y), and this synthesis filter f'i (x, y)
The above-mentioned output image zi(x, y) can be obtained by calculating the convolution integral between yi and the read image yi(x+y).

For example, as shown in FIG. 5, the correction processing unit 250 shown in FIG. 2 includes a filter gx(x,
31), a filter storage unit 551 that stores this filter gi(x, and a correction filter fi(x) for each color component corresponding to each correction target area calculated by the filter calculation unit 252.
x, y). In addition, correction circuits 253r, 253g, 253
Filter coefficient register 2 corresponding to each correction target area of b
72, the corresponding synthesis filter f'x (x+y) synthesized by the filter synthesis section 552 described above may be stored.

For example, the filter gi (
If a differential filter such as a Laplacian filter is stored in the filter storage unit 551 as x, y), the filter synthesis unit 552 generates a synthesis filter f'i( x, y
) is generated.

Further, this synthesis filter white (x, y) is stored in the corresponding filter coefficient registers 272 of the correction circuits 253r, 253g, and 253b, and as described above, the arithmetic circuit 2
77, by performing convolution integral processing of this synthesis filter white (x, y) and read image yi (x, y), color shift correction is performed, and contour and edge detection processing of the grayscale image and contour and edge processing are performed. It becomes possible to perform image processing such as edge enhancement.

Furthermore, if a smoothing filter such as a −-weighted linear filter or an edge-preserving smoothing filter is stored in the filter storage unit 551 described above, color shift can be corrected and the read image y
It becomes possible to suppress random noise contained in i(χ, y).

Also, median filter and maximum value (or minimum value)
A filter storage unit 5 stores local statistics filters such as filters.
51, it becomes possible to correct color shift and remove salt-and-pepper noise.

As described above, it is possible to perform various image processing simultaneously with color shift correction, and it is possible to output higher quality images at high speed.

〔Effect of the invention〕

As described above, according to the invention of claim 1, a filter is calculated for each of the plurality of correction target regions based on a point spread function representing the deterioration of the image read by the reading means, and this filter Since the reading results of each correction target area are corrected by , the reading error of each color component is corrected without increasing the hardware, taking into account various error factors including the fact that the reading error by the reading means differs depending on the position. This makes it possible to correct color shift and output high quality scanned images.

Further, according to the invention of claim 2, it is possible to easily obtain a point spread function based on the reading result when reading a reference image in which a pattern corresponding to a point light source is arranged in each correction target area, Using a filter calculated based on this point spread function, it is possible to correct color shift and output a high quality read image.

Further, according to the invention of claim 3, by continuously supplying the reading results of each line by the reading means to the correction processing means, convolution integration between the reading results and the filter corresponding to the corresponding correction target area is executed. Therefore, the reading result can be corrected in parallel with the reading operation of the document by the reading means, and a high quality image with color shift corrected can be outputted at high speed.

Further, according to the invention of claim 4, by correcting the reading result of each color component using the synthesis filter generated by the synthesis means, at the same time, the reading error of each color component is corrected.
It becomes possible to perform image processing using other combined filters, and it is possible to output an image of higher quality.

[Brief explanation of drawings]

FIG. 1 is a block diagram of the principle of the present invention; FIG. 2 is a configuration diagram of a color reading device according to an embodiment of the present invention; FIG. 3 is a diagram showing an example of a reference image; FIG. 4 is a configuration diagram of a correction filter; FIG. 5 is a configuration diagram of a correction processing section according to another embodiment, FIG. 6 is a configuration diagram of a conventional color reading device, FIG. 7 is a configuration diagram of a conventional color reading device, and FIG. 8 is an explanatory diagram of an image sensor. It is. In the figure, Ol is an original, 10 is a reading means, 11 is a line image sensor, 21 is a fish bed function calculation means, 22 is a reference image, 31 is a filter calculation means, 40 is a correction means, 41 is a correction processing means, 42 43 is a filter storage means, 43 is an output means, 44 is a holding means, 45 is a discrimination means, 46 is a selection means, 47 is a multiplication means, 148 is a calculation means, 151 is a synthesis means, 201, 611, 711 is a manuscript, 202 is a Reference image, 210 is an image reading unit, 211 is a color reading mechanism, 212 is a CCD line image sensor (CCD sensor), 213 is an analog-digital (A/D) converter, 231
241 is a shading correction unit, 241 is a color conversion unit, 250 is a correction processing unit, 251 is a PSF calculation unit, 252 is a filter calculation unit, 253 is a correction circuit, 261 is a counter, 262 is an address register, 263 is a discrimination unit, 271 is a line memory, 272 is a filter coefficient register, 273 is a latch, 274 is a memory controller, 275 is a selector, 276 is a selector control section, 277 is an arithmetic circuit, 281 is a multiplier, 282 is a judgment circuit, 283.284 is an adder, 285 is ROM. 551 is a filter storage unit, 552 is a filter synthesis unit, 612.712 is an illumination device, 613.713 is a reading line, 614 is a lens, 615.715 is a sensor, 621 is a filter, and 714 is a Rondo lens array. Principle diagram of the present invention Diagram illustrating an example of a reference image Diagram y311 Embodiment diagram of a correction filter Diagram y311 Embodiment (heavy pressure processing @ ■ Diagram diagram Diagram of a conventional color reading device Diagram

Claims (4)

[Claims] (1) It has a line image sensor (111) in which a plurality of light receiving elements are arranged in a row, and the line image sensor (111) has a plurality of light receiving elements arranged in a row.
a reading means (11) for reading the color image on the original (101) after separating it into a plurality of color components for each line consisting of each pixel on the original (101) corresponding to each light-receiving element;
110), dividing the line on the original (101) into a plurality of correction target areas, and corresponding to each of these correction target areas based on the reading result of each color component by the reading means (110), a point spread function calculation means (121) for calculating a point spread function representing deterioration of an image read by the reading means (110) for each color component; and a point spread function for each color component calculated by the point spread function calculation means (121). a filter calculation means (131) for calculating filter coefficients representing each of the filters for canceling the deterioration of the read image of each color component in each of the plurality of correction target regions based on a function; and the reading means (110) introduced. Manuscript by (101
), select a filter corresponding to the corresponding correction target area from among the filters for each color component introduced from the filter calculating means (131);
A color reading device comprising: a correction means (140) for outputting the result of convolution and integration of the reading result of each introduced color component and the filter of each selected color component as the reading result of each color component; Device. (2) A reference image (12) in which at least one pattern corresponding to one light receiving element of the line image sensor (111) is arranged in each of the plurality of correction target areas.
2), wherein the point spread function calculation means (121) calculates each color component corresponding to each of the plurality of correction target areas based on the reading result of the reference image (122) by the reading means (110). The color reading device according to claim 1, wherein the point spread function is calculated. (3) The correction means (140) has a plurality of correction processing means (141) corresponding to each of the plurality of color components, and each of the plurality of correction processing means (141) is configured to perform the plurality of corrections. A plurality of filter storage means (142) corresponding to each target area and storing filter coefficients representing filters of the corresponding color components, and reading results corresponding to a plurality of lines are successively inputted, and each of the plurality of lines is Output means (143) for outputting reading results corresponding to pixels in parallel
), and holding means (143) into which the output of the output means (143) is introduced and holding a predetermined amount of the reading results of the plurality of lines.
144), a determining means (145) for determining a correction target area corresponding to the reading result according to the reading result held in the holding means 144, and according to the determination result by the determining means (145), a selection means (146) for selecting a corresponding filter storage means (142) from the plurality of filter storage means (142); and a filter coefficient held in the filter storage means (142) selected by the selection means (146). and a corresponding reading result held in the holding means (144); and a calculation means (148) for calculating the sum of the multiplication results by the multiplication means (147). 2. The color reading device according to claim 1, wherein the output of the calculation means (148) is used as the output of the correction processing means (141). (4) Synthesis in which a composite filter is generated by convolving and integrating filters for each color component corresponding to each of the plurality of correction target regions calculated by the filter calculating means (131) and another filter having an image processing function. means(
151), characterized in that the composition filter generated by the composition means (151) is introduced into the correction means (140) as a filter for each color component corresponding to each of the correction target regions. Color reader.