JP3713348B2 - Image reading device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image reading apparatus in which a plurality of image sensors such as R, G, and B are separated in the sub-scanning direction.
[0002]
[Prior art]
As this type of image reading apparatus, there is known a line scanning type image reading apparatus in which a plurality of R, G, and B image sensors are separated in the sub-scanning direction and arranged in parallel. In such an image reading apparatus, there is a time lag in the image data at the same position of the document read by each sensor, and therefore correction is not performed so that the image data at the same position of the document can be obtained from each sensor. When reading a color image, color misregistration occurs and the color cannot be read correctly, so correction according to the interval between sensors is performed.
[0003]
For example, as shown in FIG. 4, when using a color image sensor arranged in the sub-scanning direction at regular intervals (every N lines) in the order of RGB, when reading in the sub-scanning direction of the arrow, the RGB reading position is every N lines. Since the G data at the time of reading the B data is the read data before the Nth line and the R data is the read data before the 2Nth line, the read data of G and R are set to N respectively on the basis of the read data of B. RGB data at the same position can be obtained by delaying by 2N lines.
[0004]
[Problems to be solved by the invention]
However, in such an image reading apparatus, the deviation of the read data between the R, G, and B sensors is determined according to the interval between the sensors and the sub-scanning speed. However, there is a problem that unevenness in the sub-scanning speed causes color misregistration and the black reproducibility deteriorates.
[0005]
In view of the above-described conventional problems, the present invention corrects the positional deviation of each image data according to the variation of the sub-scanning speed in the configuration in which the positions of the respective sensors such as R, G, and B are shifted in the sub-scanning direction. An object of the present invention is to provide an image reading apparatus capable of preventing color misregistration and improving black reproducibility.
[0006]
[Means for Solving the Problems]
In order to achieve the above object, the first means, in the image reading apparatus in which the R, G, B image sensors are separated in the sub-scanning direction, the R, G, B data read by the image sensors is obtained. The first delay means for delaying the data of the other two colors with reference to the data of one color according to the interval of each image sensor so that the reading lines are the same, and the delay by the first delay means A second delay unit that delays the other two colors of data for each line and outputs a plurality of lines of data; and each image sensor reads a predetermined pattern serving as a reference, and the first and second delay units multiple resulting R, G, based on the set of consisting of B data, consisting of further R obtained by moving only the sub-scanning direction following distance point between one line, G, B data by Ask for a pair of before From the set and the plurality of sets of groups, R, G, the difference in the data B by searching for sets is minimized, the movement amount calculating means for calculating R, G, the movement amount of each data in the B And each image sensor reads a document image on the basis of the movement amount calculated by the movement amount calculation means, and the positional deviation of each read data of R, G, B obtained by the first delay means. And a misalignment correcting means for correcting .
[0007]
A second means is characterized in that, in the first means, the predetermined pattern is a hatched pattern.
[0008]
The third means is the first and second means , wherein the movement amount calculating means is obtained by the first and second delay means, each image sensor reading a predetermined pattern as a reference. The other two colors of data are moved with reference to one point of one line of one color of R, G, B data.
[0009]
Fourth means, in the third means, the movement amount calculating means, characterized in that the concentration of the predetermined pattern to calculate the amount of movement of the other two colors of data as a reference point equal to or higher than a predetermined value .
[0010]
According to a fifth means, in the third means, the movement amount calculation means determines a reference point based on a change amount of density of a predetermined pattern and calculates a movement amount of the other two colors of data. Features.
[0011]
The sixth means has an image sensor extended in the main scanning direction for each of the three colors, and scans the document in the sub-scanning direction with respect to the document by scanning each scanner in the sub-scanning direction. In the image reading apparatus, the read image data is delayed for each line in the main scanning direction in the first color which is one of the read colors, and the first position at the same position in the sub-scanning direction is delayed. And a plurality of virtual points of the first color at the same position in the sub-scanning direction based on the first delay means capable of obtaining a plurality of image data of a plurality of colors and the plurality of first color image data. The first computing means for interpolating and outputting the image data and the second color different from the first color delay the read image data for each line in the main scanning direction, and the same in the sub scanning direction Position second A second delay unit configured to obtain a plurality of color image data, and a plurality of virtual point images of the second color at the same position in the sub-scanning direction based on the plurality of second color image data A second computing means for interpolating and outputting data; a diagonal pattern having a plurality of diagonal images provided outside the area for reading the document image and inclined with respect to the main scanning direction and the sub-scanning direction; In a third color different from the first and second colors, detection means for detecting that the edge of the hatched image has been read, and image data of the third color corresponding to the detected edge On the other hand, a plurality of image data of the first color, image data of a plurality of virtual points of the first color, a plurality of image data of the second color, and a plurality of virtual points of the second color. The difference in image data for each color from the image data is minimized. First selection means for selecting each of the first and second color image data and an interpolation calculation formula for obtaining the selected first color image data are set so that the original image is First line correction means for calculating a plurality of image data of the first color output from the first delay means when read by the interpolation equation and outputting image data of the first color; A plurality of second color image data output from the second delay means when an interpolation calculation formula for obtaining the selected second color image data is set and a document image is read. And a second line correction means for outputting image data of the second color by calculating with the interpolation calculation formula.
[0012]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a copying machine equipped with an image reading device according to the present invention, FIG. 2 is a block diagram showing in detail the main part and image processing unit of the image reading device of FIG. 1, and FIG. 3 is a document of FIG. FIG. 4 is an explanatory diagram showing the configuration of the RGB sensor of the image reading apparatus in FIG. 1, FIG. 5 is a block diagram for explaining the concept of the interline correction unit in FIG. 2, and FIG. Is an explanatory diagram showing a hatched pattern, FIG. 7 is a block diagram showing in detail the interline correction unit of FIG. 2, FIG. 8 is an explanatory diagram showing a cubic function convolution method, and FIG. 9 is a line correction coefficient calculation unit of FIG. FIG. 10 is a flowchart for explaining the processing of the line correction coefficient calculator in FIG. 7, and FIG. 11 is an explanatory diagram showing the reference point calculation method of the line correction coefficient calculator in FIG. .
[0013]
The copying machine shown in FIG. 1 reads an original and outputs an image reading unit 1 as RGB digital data, an image processing unit 2 as shown in detail in FIG. 2, and a color image processed by the image processing unit 2. Is output on the paper. The image reading unit 1 includes a scanner having R, G, and B sensors as shown in FIG. 4 and a reading data shift between the R, G, and B sensors as shown in FIG. An inter-line correction unit 10 that performs correction according to sub-scanning speed unevenness is provided.
[0014]
The RGB gamma correction unit 11 corrects the gray balance of the RGB data from the document reading unit 1 and converts it into a density, and outputs it to the delay unit 12 and the document recognition unit 13. The delay unit 12 delays the RGB data and outputs it to the RGB filter unit 14 in order to synchronize with the output result of the document recognition unit 13.
[0015]
As shown in FIG. 3, the document recognition unit 13 includes a line drawing recognition unit 13a and a color determination unit 13b. The line drawing recognition unit 13a determines whether the region is a character region or a design region based on RGB data from the RGBγ correction unit 11. , A character / picture determination signal C / P (“H” is a character and “L” is a picture) is output. In this case, it becomes “H” at the edge of the line drawing. The color determination unit 13b determines whether the region is a chromatic region or an achromatic region based on the RGB data from the RGBγ correction unit 11, and the chromatic / achromatic determination signal B / C (“H” is achromatic, “L”). Is chromatic). In this case, 4 × 4 pixels correspond to one pixel with one block.
[0016]
The character / design determination signal C / P is an image obtained by cascade connection to the RGB filter unit 14, the color correction unit 15, the UCR unit 16, the scaling unit 17, the CMYBk filter unit 18, the CMYBkγ correction unit 19, and the gradation processing unit 20. The chromatic / achromatic determination signal B / C is applied in synchronization with the image data by cascade connection to the RGB filter unit 14, the color correction unit 15, and the UCR unit 16. Each of the blocks 14 to 20 performs character processing and pattern processing based on the character / picture determination signal C / P and the chromatic / achromatic determination signal B / C. Further, the document recognition unit 13 also determines whether the document is a color document or a monochrome document, and switches the sequence and processing based on the determination result.
[0017]
The RGB filter unit 14 is an N × N pixel filter that performs MTF correction on the RGB data from the delay unit 12. When the character / picture determination signal C / P is “H” (character), sharpening processing is performed, and “L "(Picture), smoothing processing is performed. The color correction unit 15 converts the RGB data processed by the RGB filter unit 14 into C, M, and Y data using a primary masking method or the like, and the UCR unit 16 uses the color correction unit 15 to improve the color reproducibility of the image data. Bk data is generated by performing UCR (additional color removal) processing on the common part of the C, M, and Y data converted by the above.
[0018]
The UCR unit 16 also generates skeleton black Bk data when the character / picture determination signal C / P is “L” (picture), and generates full black Bk data when it is “H” (character). The UCR unit 16 also converts a black character into a black component only when the character / picture determination signal C / P is “H” (character) and the chromatic / achromatic determination signal B / C is “H” (achromatic). The C, M, and Y data are erased for the purpose of expression.
[0019]
The UCR unit 16 provides the C, M, Y, and Bk to the magnification unit 17 in order for the image reading unit 1 to read one original document four times and to output the original to the image recording unit 2 in a surface sequential manner. 1 signal IMG is output. In this case, when the document is pre-scanned and the document type is determined (ACS), Bk data is generated during the first scan, and the document recognition unit 13 determines whether the document is a color document or a monochrome document based on the result of determination. In the case of a document, scanning is completed once, and in the case of a document other than a monochrome document (color document), scanning is performed four times and a large amount of Bk data is generated.
[0020]
The scaling unit 17 performs enlargement, reduction, and equal magnification processing in the main scanning direction, and the CMYBk filter unit 18 is a filter of N × N pixels, and the frequency characteristics of the image recording unit 3 and the character / picture determination signal C / P A sharpening process or a smoothing process is performed based on the above. The CMYBkγ correction unit 19 changes the γ curve based on the frequency characteristics of the image recording unit 3 and the character / picture determination signal C / P. Specifically, the character / picture determination signal C / P is “L” (picture). The contrast is emphasized by raising γ to faithfully reproduce the image sometimes, and γ for “H” (character).
[0021]
The gradation processing unit 20 performs quantization such as dither processing based on the gradation characteristics of the image recording unit 3 and the character / picture determination signal C / P. Specifically, the character / picture determination signal C / P is “L”. ”(Pattern), gradation-oriented processing is performed, and“ H ”(character), resolution-oriented processing is performed.
[0022]
Next, the interline correction will be described in detail. As an example, the RGB sensors of the scanner constituting the image reading unit 1 are arranged at equal intervals (every N lines) in the sub-scanning direction in the order of RGB as shown in FIG. Therefore, when reading in the direction of the arrow, the RGB reading position is shifted every N lines. When reading B, G is read data before the N line, and R is read data before the 2N line.
[0023]
The concept of the correction method when there is a variation in the sub-scanning speed in the configuration in which the positions of the R, G, and B sensors are shifted in the sub-scanning direction will be described with reference to FIG. First, the B data is output after being delayed by one line by the one-line memory 101B so that it becomes a reference for the R and G data. The G data is delayed by N lines, N + 1 lines, and N + 2 lines by the N line delay memory 101G and the 1 line memories 102G and 103G, respectively, and applied to the line correction unit 104G, and the R data is applied to the 2N line delay memory 101R and 1 line. The memories 102R and 103R are delayed by 2N lines, 2N + 1 lines, and 2N + 2 lines, respectively, and applied to the line correction unit 104R.
[0024]
Then, the line correction unit 104G corrects the value of each pixel of the N + 1 line based on the N line and the N + 1 line or the N + 1 line and the N + 2 line according to the direction of fluctuation of the sub-scanning speed, and the line correction unit 104R Based on the 2N line and the 2N + 1 line or the 2N + 1 line and the 2N + 2 line, the value of each pixel of the 2N + 1 line is interpolated. As the interpolation calculation method, a smoothing method can be used in addition to the cubic function convolution method.
[0025]
Next, a process for detecting a change in the sub-scanning speed and correcting the line spacing will be specifically described. First, as shown in FIG. 6, on the document reading table 1 a of the scanner constituting the image reading unit 1, a black diagonal 45 ° oblique line pattern P on the white background is placed at a constant pitch in the sub-scanning direction at a position outside the document reading range. The oblique line pattern P is first read by the RGB sensors of the scanner in the main scanning direction, and then the document image is read.
[0026]
Further, the interline correction unit 10 is configured as shown in FIG. The basic configuration is the same as in FIG. 5, but since 5 lines are used, the B data serving as a reference for the R and G data is delayed by 3 lines by the 3 line delay memory 201B and output to the image processing unit 2. And applied to the line correction coefficient calculation unit 208.
[0027]
G data is delayed by N line, N + 1 line, N + 2 line, N + 3 line, N + 4 line and N + 5 line by N line delay memory 201G and 1 line memory 202G, 203G, 204G, 205G and 206G, respectively, and 5 lines N + 1 to N + 5 Are applied to the line correction unit 207G and 6 lines N to N + 5 are applied to the line correction coefficient calculation unit 208.
[0028]
The R data is delayed by 2N line, 2N + 1 line, 2N + 2 line, 2N + 3 line, 2N + 4 line, and 2N + 5 line by 2N line delay memory 201R and 1 line memory 202R, 203R, 204R, 205R, and 206R, respectively, from 2N + 1 to 2N + 5 Five lines are applied to the line correction unit 207R, and six lines 2N to 2N + 5 are applied to the line correction coefficient calculation unit 208.
[0029]
Then, the line correction coefficient calculation unit 208 is configured such that the reference B sensor scans the oblique line pattern in the main scanning direction and the value of the B data read becomes a predetermined value or more (a point where the white data changes from black to black). ) As a reference (correction point), the coefficients in the cubic function convolution method are calculated, and the line correction units 207R and 207G interpolate 2N + 3 lines and N + 3 lines, respectively, based on these coefficients and 5 lines. Further, the system control unit 209 outputs a hatched pattern area signal and a reading end signal to the line correction coefficient calculation unit 208.
[0030]
FIG. 8 shows a general cubic function convolution method, the horizontal axis indicates the sub-scanning direction, and the vertical axis indicates the deviation of the pixel value. When there is no sub-scanning speed unevenness, the pixel positions are equally spaced as indicated by P, and when there is speed unevenness, the distance varies as indicated by Q and deviates from the correct pixel position P. Therefore, first, data within two pixels r0 from the correct pixel row Pn is detected from the position error data. In this case, data strings Q of n, n + 1, n + 2, n + 3, and n + 4 are targeted.
[0031]
Next, an interpolation function h (r) is obtained from the distance r of each data Q from Pn. This becomes a correction coefficient, and correct data Pn is obtained by multiplying the data between the lines. Further, in order to correct density unevenness, the denominator is set as the sum of correction coefficients so that the sum of the correction coefficients becomes “1”. When this process is completed for the data in the main scanning direction of the nth line, the process is shifted to the (n + 1) th line, and this process is repeated until the last line.
[0032]
In this embodiment, after the correction point of B data is determined as shown in FIG. 9, 1/8 on the + side and the − side with respect to 1 dot of R and G data on 2N + 3 line and N + 3 line, respectively. R, G data of each position of 1/8, 2/8, 3/8, 4/8, 5/8, 6/8, 7/8 dots is calculated by shifting the dots, and this R, G data and +3 Find the combination that minimizes the difference in B data in the line. In this case, for the R and G data, the data shifted by 1/8 dot and the data not shifted are combined as described above. Then, the shift amount of the R data and the shift amount of the G data when the difference between the R, G, and B data is minimized are determined, and the R and G data are corrected based on each shift amount.
[0033]
Next, the processing of the line correction coefficient calculation unit 208 will be described in detail with reference to FIG. First, it is checked whether or not a hatched pattern area signal from the system control unit 209 has been input (step S1). Then, image data of five sampling points related to R data is input from the line memories 201R to 205R (step S2), and image data of five sampling points related to G data is input from the line memories 201G to 205G (step S3). ), 8 image data of 8 virtual points based on the image data of 5 sampling points related to the R data (step S4), and 8 image data of 8 virtual points based on the image data of 5 sampling points related to the G data. Image data is calculated (step S5).
[0034]
Next, based on the image data relating to B, as shown in FIG. 11A, it is checked whether or not a change point of the hatched image has been detected (step S6). If not detected, the process returns to step S2, and on the other hand, Then, it progresses to step S7. In step S7, for the B image data corresponding to the detected edge, a plurality of R image data, a plurality of R virtual point image data, a plurality of B image data, and a plurality of B virtual points. One R, G image data is selected from the image data so that the difference between the image data of each color is minimized.
[0035]
Next, an interpolation calculation expression for obtaining each selected image data is set in the line correction units 207R and 107G (step S8), and then it is checked whether or not a reading end signal from the system control unit 209 is input. (Step S9) If the input is not made, the process returns to Step S2. If the input is made, the process is terminated.
[0036]
Therefore, by using a combination that minimizes the difference in R, G, and B data when the oblique line pattern P is read in this way, it is possible to correct the position error due to the sub-scanning speed unevenness. Further, according to the correction between lines of the present embodiment, correction after the decimal point at the time of zooming is also performed simultaneously. In the circuit shown in FIG. 7, the line correction coefficient calculation unit 208 performs the calculation one line before the processing of the line correction units 207R and 207G, so that the processing can be performed in real time. In this embodiment, the correction point is a time point when the hatched white changes to black. Therefore, when there are a plurality of combinations that minimize the difference in R, G, B data, the most negative side (already A combination having the longest distance on the scanned line side) is selected and used.
[0037]
Here, in this embodiment, the correction point is obtained from the change in the shaded density as shown in FIG. 11A, but instead, it may be obtained from the change amount as shown in FIG. 11B. The oblique line pattern is for line-to-line correction. In an apparatus that outputs an image to paper, such as a copying machine, it is not necessary to print the oblique line pattern, but in an apparatus that outputs an image as digital data to the outside. A diagonal line pattern may be output together to perform subtle line-to-line correction. Further, when the minimum value of the R, G, and B data is larger than the allowable value, a message that cannot be corrected may be notified to the operator.
[0038]
【The invention's effect】
As described above, according to the present invention, since the position error due to the unevenness of the sub-scanning speed is corrected, it is possible to correct the position shift while moving the image sensor during the sub-scanning. It becomes possible to dynamically cope with unevenness in the sub-scanning speed due to the vibration of the apparatus .
[Brief description of the drawings]
FIG. 1 is a block diagram showing a schematic configuration of a copying machine including an image reading apparatus according to the present invention.
2 is a block diagram illustrating in detail a main part and an image processing unit of the image reading apparatus in FIG. 1;
FIG. 3 is a block diagram illustrating in detail a document recognition unit in FIG. 2;
4 is an explanatory diagram showing a configuration of an RGB sensor of the image reading apparatus in FIG. 1. FIG.
FIG. 5 is a block diagram for explaining the concept of the interline correction unit in FIG. 2;
FIG. 6 is an explanatory diagram showing a hatched pattern.
7 is a block diagram showing in detail an interline correction unit in FIG. 2; FIG.
FIG. 8 is an explanatory diagram showing a cubic function convolution method.
FIG. 9 is an explanatory diagram showing processing timing of the line correction coefficient calculation unit in FIG. 7;
10 is a flowchart for explaining processing of a line correction coefficient calculation unit in FIG. 7;
11 is an explanatory diagram illustrating a reference point calculation method of the line correction coefficient calculation unit of FIG. 7;
[Explanation of symbols]
201R 2N line delay memory 201G N line delay memory 201B 3 line delay memories 202R to 206R, 202G to 206G Line memories 207R and 207G Line correction unit 208 Line correction coefficient calculation unit 209 System control unit

Claims (6)

In an image reading apparatus in which R, G, B image sensors are separated in the sub-scanning direction,
The R, G, and B data read by the image sensors are delayed with respect to the data of the other two colors based on the data of one color according to the interval of the image sensors so that the reading lines are the same. 1 delay means;
Second delay means for delaying the other two-color data delayed by the first delay means for each line and outputting a plurality of lines of data;
Reading the predetermined pattern as a reference said each image sensor, the first 1, R obtained by the second delay unit, G, based on the set of consisting of the data B, still following the decimal point between the first line A plurality of sets of R, G, and B data obtained by moving in the sub-scanning direction by a distance is obtained , and each of the R, G, and B data is obtained from the base set and the plurality of sets . A movement amount calculating means for calculating a movement amount of each of the R, G, and B data by searching for a pair having the smallest difference;
Based on the movement amount calculated by the movement amount calculation means, each image sensor reads a document image, and corrects the positional deviation of each read data of R, G, B obtained by the first delay means. Misalignment correction means ;
An image reading apparatus comprising: The image reading apparatus according to claim 1, wherein the predetermined pattern is a hatched pattern. The movement amount calculation means reads a predetermined pattern as a reference by each image sensor, and one point of one line of one color of R, G, B data obtained by the first and second delay means The image reading apparatus according to claim 1, wherein the data of the other two colors are moved with reference to. The movement amount calculating means, the image reading apparatus according to claim 3, wherein the concentration of the predetermined pattern to calculate the amount of movement of the other two colors of data as a reference point for more than a predetermined value. 4. The image reading apparatus according to claim 3, wherein the movement amount calculating means determines a reference point based on an amount of change in density of a predetermined pattern and calculates a movement amount of data of the other two colors. In an image reading apparatus for reading a document image by scanning a document in a sub-scanning direction with a scanner having image sensors extending in the main-scanning direction for each of three colors, each provided separately in the sub-scanning direction ,
In the first color that is one of the read colors, the read image data is delayed for each line in the main scanning direction, and a plurality of image data of the first color at the same position in the sub-scanning direction A first delay means configured to obtain:
First arithmetic means for interpolating and outputting image data of a plurality of virtual points of the first color at the same position in the sub-scanning direction based on the plurality of first color image data;
In the second color different from the first color, the read image data is delayed for each line in the main scanning direction, and a plurality of image data of the second color at the same position in the sub scanning direction is obtained. Second delay means as described above,
Second arithmetic means for interpolating and outputting image data of a plurality of virtual points of the second color at the same position in the sub-scanning direction based on the plurality of second color image data;
An oblique line pattern having a plurality of oblique line images provided outside an area for reading a document image and inclined with respect to the main scanning direction and the sub-scanning direction;
Detecting means for detecting that an edge of the oblique line image is read in a third color different from the first and second colors;
For a third color image data corresponding to the detected edge, a plurality of image data of the first color, a plurality of virtual point image data of the first color, and a second color of the second color. A first selection is made for each of the first and second color image data that minimizes the difference between the image data of each color from the plurality of image data and the image data of the plurality of virtual points of the second color. And means for selecting
An interpolation equation for obtaining the selected first color image data is set, and a plurality of first color image data output from the first delay means when a document image is read. First line correction means for calculating the interpolation equation and outputting image data of the first color;
An interpolation equation for obtaining the selected second color image data is set, and a plurality of second color image data output from the second delay means when the original image is read. Second line correction means for calculating the interpolation equation and outputting image data of the second color;
An image reading apparatus comprising:

Images