JP7169231B2 - Image processing device, image processing method, and image processing program - Google Patents

Description

The technology disclosed herein relates to an image processing device, an image processing method, and an image processing program.

2. Description of the Related Art Some image reading devices such as scanners and copiers use a contact image sensor (CIS). In an image reading apparatus using CIS, LEDs (Light Emitting Diodes) of a light source are arranged, for example, by B light (blue light)→R light (red light)→G light (green light)→B light→R light→G light→B The image is read by lighting in a predetermined lighting order of light → .

JP 2012-060690 A JP 2016-092474 A

In an image reading apparatus using CIS, when an image is read by lighting the LEDs of the light source in the predetermined lighting order as described above, the image read by CIS (hereinafter sometimes referred to as "read image") has " "Color deviation" may occur. For example, when the read image is an image of black characters (hereinafter sometimes referred to as a "black character image"), chromatic color shift may occur at the upper and lower ends of the black character image.

The disclosed technique has been made in view of the above, and aims to reduce color misregistration that occurs in a read image.

An image processing apparatus according to an aspect of the disclosure performs image processing on a read image read by an image sensor. The image sensor has a plurality of imaging elements arranged in a row in the main scanning direction, and a light source capable of irradiating a medium to be read with blue, red, and green lights in a predetermined lighting order. The image processing apparatus has a storage section, a detection section, and a correction section. The storage unit stores the read image. The detection unit detects a hue switching portion in the sub-scanning direction of the image sensor in the read image. The correcting unit corrects the color shift occurring in the switching portion according to the mode of hue switching in the switching portion and the predetermined lighting order.

According to the disclosed aspect, it is possible to reduce color misregistration that occurs in a read image.

FIG. 1 is a diagram illustrating a configuration example of an image reading apparatus according to a first embodiment; FIG. 2 is a diagram showing an example of the lighting order of the light sources according to the first embodiment. FIG. 3 is a diagram illustrating an example of color misregistration according to the first embodiment. FIG. 4 is a diagram illustrating an example of color misregistration according to the first embodiment. FIG. 5 is a diagram illustrating an example of color misregistration according to the first embodiment. FIG. 6 is a flowchart illustrating an example of processing of the image processing apparatus according to the first embodiment; FIG. 7 is a flowchart illustrating an example of processing of the image processing apparatus according to the first embodiment; FIG. 8 is a diagram illustrating an operation example of the image processing apparatus according to the first embodiment; FIG. 9 is a diagram illustrating an operation example of the image processing apparatus according to the first embodiment; FIG. 10 is a diagram illustrating an operation example of the image processing apparatus according to the first embodiment; FIG. 11 is a diagram illustrating an example of the lighting order of the light sources according to the second embodiment. FIG. 12 is a diagram illustrating an example of color misregistration according to the second embodiment. FIG. 13 is a diagram illustrating an operation example of the image processing apparatus according to the second embodiment; FIG. 14 is a diagram illustrating an operation example of the image processing apparatus according to the second embodiment;

Embodiments of an image processing apparatus, an image processing method, and an image processing program disclosed in the present application will be described below with reference to the drawings. Note that the image processing apparatus, the image processing method, and the image processing program disclosed in the present application are not limited to these embodiments. Also, in the embodiments, the same symbols are attached to the configurations having the same functions and the steps performing the same processing.

A black character image will be described below as an example of a content image. However, content images to which the technology disclosed herein can be applied are not limited to black character images. For example, the technology disclosed herein can also be applied to content images such as pattern images and graphic images.

[Example 1]
<Configuration of image reading device>
FIG. 1 is a diagram illustrating a configuration example of an image reading apparatus according to a first embodiment; In FIG. 1 , the image reading device 1 has a CIS 20 , a lighting control section 31 and an image processing device 10 . Examples of the image reading device 1 include a scanner device, a copier, and the like.

The CIS 20 has a light source 21 formed by, for example, an LED array, and a plurality of imaging elements 22 arranged in a row in the main scanning direction of the CIS 20 . The light source 21 can irradiate the medium to be read with B light (blue light), R light (red light), and G light (green light) in a predetermined lighting order. The lighting control unit 31 controls the lighting order of the B light, the R light, and the G light, and the lighting time of each of the B light, the R light, and the G light. An example of a medium to be read is white paper on which black characters are printed.

The image processing device 10 has a storage unit 11 , a smoothing unit 12 , a detection unit 13 and a correction unit 14 .

Images read by the CIS 20 (that is, “read images”) are stored in the storage unit 11 .

The smoothing unit 12, the detection unit 13, and the correction unit 14 perform image processing on the read image. A smoothing unit 12 smoothes the saturation of the read image. The detection unit 13 detects a hue switching portion (hereinafter referred to as a “hue switching portion”) in the sub-scanning direction of the CIS 20 in the read image after the saturation is smoothed (hereinafter sometimes referred to as a “smoothed image”). ) is detected. The correcting unit 14 corrects color misregistration that occurs in hue switching portions. Details of the smoothing unit 12, the detecting unit 13, and the correcting unit 14 will be described later.

The storage unit 11 is implemented as hardware, for example, by a memory. Examples of memory include RAM (Random Access Memory) such as SDRAM (Synchronous Dynamic Random Access Memory), ROM (Read Only Memory), and flash memory.

The lighting control unit 31, the smoothing unit 12, the detecting unit 13, and the correcting unit 14 are implemented as hardware by, for example, a processor. Examples of processors include CPUs (Central Processing Units), DSPs (Digital Signal Processors), and FPGAs (Field Programmable Gate Arrays). Also, the lighting control unit 31, the smoothing unit 12, the detection unit 13, and the correction unit 14 may be realized by an LSI (Large Scale Integrated circuit) including a processor and peripheral circuits. Furthermore, the lighting control unit 31, the smoothing unit 12, the detection unit 13, and the correction unit 14 may be implemented using a GPU (Graphics Processing Unit), an ASIC (Application Specific Integrated Circuit), or the like.

<Relationship between lighting order of light sources and color shift>
FIG. 2 is a diagram showing an example of the lighting order of the light sources according to the first embodiment. 3 to 5 are diagrams showing an example of color misregistration according to the first embodiment.

As shown in FIG. 2, as the CIS 20 moves in the sub-scanning direction, the lighting control unit 31 performs B light→R light→G light→B light→R light→G light→B at a constant period LT. The light sources 21 are turned on in a predetermined lighting order OR1 of light → . The cycle LT corresponds to the reading cycle of one line (one row) of the CIS 20 . That is, in the lighting order OR1, the lighting control unit 31 starts lighting the B light at the beginning of the cycle LT, and after lighting the B light for a predetermined lighting time TB1, switches the B light to the R light, and turns the R light on. After lighting for a predetermined lighting time TR1, the R light is switched to G light, and the G light is lighted for a predetermined lighting time TG1 until the end of the period LT. Further, the lighting control unit 31 sets the relationship between the lighting times TB1, TR1, and TG1 of the B light, the R light, and the G light in one period of the period LT to be "TB1=TG1" and "TB1, TG1<TR1."

When the light sources 21 are lit in the lighting order OR1 and the lighting times TB1, TR1, and TG1 as described above, for example, when the read image includes four black character images "TEST" as shown in FIG. , color shift may occur between the upper end and the lower end of the image of each character. Hereinafter, the upper end portion of the black character image may be referred to as the "black character image upper end portion", and the lower end portion of the black character image may be referred to as the "black character image lower end portion". As shown in FIG. 3, the black character image of "T" has, for example, one black character image top portion and three black character image bottom portions, and the black character image of "E" has, for example, three black character image top portions. and three black character image bottoms, and the black character image of "S" has, for example, four black character image tops and four black character image bottoms. Then, a color shift occurs at the black character image upper end portion and the black character image lower end portion of the black character image "TEST".

For example, focusing on the black character image "T" in the four black character images "TEST", as shown in FIG. The read image RI includes a black content image and a white image other than the content image (hereinafter sometimes referred to as a “background image”). It is formed from a circle, a background pixel Δ, and a background pixel ▽. Of the background pixels ◯, Δ, and ▽, the background pixel △ is the background pixel adjacent to the content pixel ● in the upper edge of the black character image (hereinafter sometimes referred to as the “upper edge adjacent pixel”), and the background pixel ▽ , and background pixels adjacent to the content pixel ● at the lower edge of the black character image (hereinafter sometimes referred to as “lower edge adjacent pixels”). When the light source 21 is turned on in the lighting order OR1 and the lighting times TB1, TR1, and TG1 as described above, B light and R light are applied to the upper-end adjacent pixel Δ and the lower-end adjacent pixel ▽ as shown in FIG. , a color shift of a specific hue corresponding to each lighting timing of the G light. In FIG. 5, for the sake of simplification, the B light, R light, and G light are shown to have the same length of lighting time. The relationship between the lighting times TB1, TR1 and TG1 of the R light and the G light is "TB1=TG1" and "TB1, TG1<TR1".

In FIG. 5, state A1 shows a state in which the upper end of the black character image arrives at the timing when the B light is turned on, and the lower end of the black character image arrives at the timing when the G light is extinguished. In the state A1, no color shift occurs in the upper edge adjacent pixel Δ or the lower edge adjacent pixel ▽.

State B1 shows a state in which the upper end of the black character image is reached while the B light is on. In state B1, a blue color shift occurs in the upper edge adjacent pixel Δ.

State C1 indicates a state in which the upper end of the black character image is reached while the R light or the G light is being lit. In state C1, a magenta color shift occurs in the upper edge adjacent pixel Δ.

State D1 indicates a state in which the lower end of the black character image is reached while the G light is on. In state D1, a green color shift occurs in the pixels ▽ adjacent to the lower edge.

State E1 indicates a state in which the lower end of the black character image is reached while the B light is being lit or the R light is being lit. In the state E1, a yellow color shift occurs in the pixels ▽ adjacent to the lower edge.

Thus, if the light sources 21 are lit in the lighting order OR1 and the lighting times TB1, TR1, and TG1, there is a possibility that color misregistration having blue or magenta may occur in the pixels Δ adjacent to the upper edge. Pixel ▽ may have a color shift with a green or yellow color. The combination of the blue and magenta color shifts in the top adjacent pixel Δ and the green and yellow color shifts in the bottom adjacent pixel ▽ corresponds to the “first type of color shift”. As for the color shift that occurs in the upper-end adjacent pixel Δ, since the occurrence rate of magenta color shift is higher than that of blue color shift, the upper-end adjacent pixel Δ has a hue closer to magenta. There is a high possibility that color misalignment will occur. On the other hand, regarding the color shift that occurs in the bottom-portion adjacent pixel ▽, yellow color shift is more likely to occur than green color shift. There is a high possibility that color misregistration will occur.

<Processing/Operation of Image Processing Apparatus>
6 and 7 are flowcharts illustrating an example of processing of the image processing apparatus according to the first embodiment. 8 to 10 are diagrams showing an operation example of the image processing apparatus according to the first embodiment.

In FIG. 6, in step S101, the smoothing unit 12 acquires the read image RI from the storage unit 11, targets all the pixels ●, ◯, Δ, and ▽ forming the acquired read image RI, and calculates the saturation of each pixel. is averaged to smooth the saturation of the read image RI. The smoothing unit 12 uses, for example, a smoothing filter having a size of 5 pixels×5 pixels, and smoothes all the pixels ●, ◯, Δ, and ▽ forming the read image RI. While moving sequentially, the saturation components of a total of 25 pixels, two pixels on the left, right, top and bottom centered on the smoothing target pixel, are averaged, and the saturation components after averaging are Update the saturation component of the smoothing target pixel. In this way, by smoothing the saturation using a smoothing filter having a size of, for example, 5 pixels×5 pixels, it is possible to reduce the color shift to about one-fifth.

Next, in step S103, the detection unit 13 detects a hue switching portion in the smoothed image. FIG. 7 shows a processing flow for detecting a hue change portion performed in step S103.

In FIG. 7, in step S201, the detection unit 13 resets the row counter n to "1".

Next, in step S203, the detection unit 13 resets the column counter m to "1".

Next, in step S205, the detection unit 13 selects the n-th row (that is, the n-th line) in the sub-scanning direction and m YUV data (Y: luminance component, UV: chroma component) is converted from RGB data of the pixel in the column (that is, the n×m-th pixel) (hereinafter sometimes referred to as the “target pixel”) according to a known conversion formula. , and the pixel of interest is decomposed into a Y component (luminance component) and a UV component (chroma component).

Next, in step S207, the detection unit 13 determines whether or not the UV component of the pixel of interest (that is, the saturation of the pixel of interest) is less than the threshold TH1.

If the UV component of the pixel of interest is less than the threshold TH1 (step S207: Yes), the detection unit 13 determines that the hue of the pixel of interest is achromatic (white, black, or gray), and the process proceeds to step S211. move on.

On the other hand, when the UV component of the pixel of interest is equal to or greater than the threshold TH1 (step S207: No), in step S209, the detection unit 13 determines that the hue of the pixel of interest is chromatic, and assigns the "label C” is given. After the process of step S209, the process proceeds to step S217.

In step S211, the detection unit 13 determines whether or not the Y component of the pixel of interest (that is, the luminance of the pixel of interest) is equal to or greater than the threshold TH2.

If the Y component of the pixel of interest is equal to or greater than the threshold TH2 (step S211: Yes), in step S213, the detection unit 13 determines that the luminance of the pixel of interest is "highlight" and assigns the pixel of interest "label H ” is given. After the process of step S213, the process proceeds to step S217.

On the other hand, when the Y component of the pixel of interest is less than the threshold TH2 (step S211: No), in step S215, the detection unit 13 determines that the luminance of the pixel of interest is "shadow" and assigns " Label S” is given. After the process of step S215, the process proceeds to step S217.

In step S217, the detection unit 13 determines whether or not the column counter m has reached the number of columns M in all pixels of the smoothed image. If the column counter m has not reached the number of columns M (step S217: No), the process proceeds to step S219. On the other hand, if the column counter m has reached the number of columns M (step S217: Yes), the process proceeds to step S221.

In step S219, the detection unit 13 increments the column counter m. After the process of step S219, the process returns to step S205.

On the other hand, in step S221, the detection unit 13 determines whether or not the row counter n has reached the number of rows N in all pixels of the smoothed image. If the row counter n has not reached the number of rows N (step S221: No), the process proceeds to step S223. On the other hand, if the row counter n has reached the number of rows N (step S221: Yes), the process proceeds to step S225.

In step S223, the detection unit 13 increments the row counter n. After the process of step S223, the process returns to step S203.

Through the processing of steps S201 to S223 described above, each of all pixels in the smoothed image is labeled with either label C, label H, or label S.

Then, in step S225, the detection unit 13 examines the labels of all pixels in the smoothed image, and detects pixels labeled H (hereinafter referred to as "H pixels") in the sub-scanning direction (that is, column direction). ) to the pixel labeled S (hereinafter sometimes referred to as “S pixel”) (hereinafter sometimes referred to as “HS switching portion”), and from the S pixel to A switching portion to H pixels (hereinafter sometimes referred to as “SH switching portion”) is detected. In other words, of the modes of hue switching in the hue switching portion in the sub-scanning direction of the smoothed image, modes to be corrected by the correction unit 14 include switching from highlight to shadow and switching from shadow to high. There are two modes of switching to light.

Therefore, for example, in FIG. 4, the boundary in the sub-scanning direction between the upper-end adjacent pixel Δ that is the H pixel and the content pixel that is the S pixel, that is, the portion where the background pixel is switched to the content pixel in the sub-scanning direction is detected. It is detected by the unit 13 as the HS switching portion. In FIG. 4 above, the boundary in the sub-scanning direction between the content pixel that is the S pixel and the lower-end adjacent pixel ▽ that is the H pixel, that is, the portion where the content pixel is switched to the background pixel in the sub-scanning direction is the detection unit. 13 as the SH switching portion.

In this way, the detection unit 13 detects a hue switching portion in the sub-scanning direction of the CIS 20 in the smoothed image.

After the process of step S225, the process proceeds to step S105 (FIG. 6).

Returning to FIG. 6, in step S105, the correction unit 14 corrects color shift occurring in the HS switching portion (hereinafter sometimes referred to as "HS partial color shift") and SH switching portion. The color shift that occurs (hereinafter sometimes referred to as "SH partial color shift") is corrected.

As described above, when the light source 21 is lit in the lighting order OR1 and the lighting times TB1, TR1, and TG1, the pixels Δ forming the HS switching portion are colored blue or magenta as the HS partial color shift. There is a possibility that a color shift having a color may occur, and a color shift having green or yellow may occur as an SH partial color shift at the lower edge adjacent pixel ▽ forming the SH switching portion. As for the color shift (that is, the HS partial color shift) that occurs in the upper-end adjacent pixel Δ, the occurrence rate of magenta color shift is higher than that of blue color shift, and it occurs in the lower-end adjacent pixel ▽. As for the color shift (that is, SH partial color shift), yellow color shift has a higher occurrence rate than green color shift.

Therefore, the correction unit 14 first converts the RGB data of the upper-end adjacent pixel Δ and the lower-end adjacent pixel ▽ into XYZ data according to a known conversion formula, and further converts the XYZ data into Lab data according to a known conversion formula. . In the Lab data, "L" is the luminance component and "ab" is the saturation component. When the saturation component ab is used, as shown in FIGS. 8 and 9, the hue of each pixel is represented by the two-dimensional coordinates of the a-axis (±a) and the b-axis (±b). In FIGS. 8 and 9, as a and b approach 0 (zero), the hue approaches achromatic color.

In FIG. 9, ten predetermined ranges set by dividing the hue range represented by the two-dimensional coordinates of the a-axis (±a) and the b-axis (±b) by 36° into 10 A case where each hue is assigned to each is shown as an example. In the example shown in FIG. 9, for example, the range of 108° or more and less than 144° corresponds to magenta, the range of 144° or more and less than 180° corresponds to bluish purple, and the range of 180° or more and less than 216° corresponds to blue. Equivalent to. For example, in the example shown in FIG. 9, the range of 324° or more and less than 360° corresponds to yellow, the range of 288° or more and less than 324° corresponds to yellowish green, and the range of 252° or more and less than 288° corresponds to green. Equivalent to.

Next, as shown in FIG. 10, the correction unit 14 corrects the color misregistration of the upper-end adjacent pixel Δ and the lower-end adjacent pixel ▽. In FIG. 10, "L-in" is the luminance component before correction, "a_in" is a before correction, "b_in" is b before correction, "L_out" is the luminance component after correction, and "a_out" is a after correction, and "b_out" indicates b after correction.

That is, as shown in FIG. 10, the correction unit 14 does not correct the luminance component L of both the upper-end adjacent pixel Δ and the lower-end adjacent pixel ▽.

Further, the correction unit 14 determines that, for the upper-edge adjacent pixel Δ, the hue before correction of the upper-edge adjacent pixel Δ (that is, the hue of the color shift occurring in the upper-edge adjacent pixel Δ) is (a_in, b_in)= If (20, -20), CS11 corrects (a_in, b_in)=(20, -20) to (a_out, b_out)=(2.0, -2.0). The hue specified by (a_in, b_in)=(20,-20) corresponds to magenta, as shown in FIG.

In addition, for the upper-end adjacent pixel Δ, for example, when the hue before correction of the upper-end adjacent pixel Δ is (a_in, b_in)=(5, −30), the correction unit 14 outputs (a_in, b_in ) = (5, -30) to (a_out, b_out) = (2.5, -15.0). The hue specified by (a_in, b_in)=(5,-30) corresponds to bluish purple, as shown in FIG.

For the upper-end adjacent pixel Δ, for example, when the hue before correction of the upper-end adjacent pixel Δ is (a_in, b_in)=(−10, −30), the correction unit 14 outputs (a_in, b_in) = (-10, -30) is corrected to (a_out, b_out) = (-9.0, -27.0). The hue specified by (a_in, b_in)=(-10,-30) corresponds to blue, as shown in FIG.

Thus, the hue component reduction rate in case CS11 is greater than the hue component reduction rate in case CS12, and the hue component reduction rate in case CS12 is greater than the hue component reduction rate in case CS13. , CS11, CS12, and CS13 in the order of large, medium, and small hue component reduction ratios for the upper-end adjacent pixel Δ (FIGS. 9 and 10).

Further, the correction unit 14 determines that, for the lower-end adjacent pixel ▽, for example, the hue before correction of the lower-end adjacent pixel ▽ (that is, the hue of the color shift occurring in the lower-end adjacent pixel ▽) is (a_in, b_in)= If (-5, 30), CS21 corrects (a_in, b_in)=(-5, 30) to (a_out, b_out)=(-0.5, 3.0). The hue specified by (a_in, b_in)=(-5, 30) corresponds to yellow, as shown in FIG.

For example, when the hue before correction of the lower-end adjacent pixel ▽ is (a_in, b_in)=(−20, 20), the correction unit 14 sets (a_in, b_in ) = (-20, 20) is corrected to (a_out, b_out) = (-10.0, 10.0). The hue specified by (a_in, b_in)=(-20, 20) corresponds to yellow-green, as shown in FIG.

In addition, for the lower-end adjacent pixel ▽, for example, when the hue before correction of the lower-end adjacent pixel ▽ is (a_in, b_in)=(−30, 0), the correction unit 14 outputs (a_in, b_in ) = (-30, 0) to (a_out, b_out) = (-27.0, 0). The hue specified by (a_in, b_in)=(-30, 0) corresponds to green, as shown in FIG.

Thus, the hue component reduction rate in case CS21 is greater than the hue component reduction rate in case CS22, and the hue component reduction rate in case CS22 is greater than the hue component reduction rate in case CS23. , and lower-end adjacent pixels ▽ are arranged in the order of large, medium, and small, in the order CS21, CS22, and CS23 (FIGS. 9 and 10).

The correction unit 14 corrects the chroma component as shown in FIG. This is done using table TB1. The hue conversion table TB1 is stored in the storage unit 11 in advance.

Then, the correction unit 14 inversely transforms the Lab data of the upper-end adjacent pixel Δ and the lower-end adjacent pixel ▽ after the saturation component correction into XYZ data according to a known conversion formula, and converts the XYZ data to XYZ data according to a known conversion formula. Convert back to RGB data.

As described above, the correction unit 14 reduces the saturation of magenta, bluish purple, blue, yellow, yellowish green, and green, which are specific hues that the first type of color shift may have. This corrects the first type of color misregistration.

In addition, the correction unit 14 performs a reduction rate according to the saturation of magenta, bluish purple, blue, yellow, yellowish green, and green, which are specific hues that the first type of color shift may have. Reduce saturation.

The first embodiment has been described above.

[Example 2]
In the second embodiment, the lighting order of the B light, the R light, and the G light in the light source 21 is different from that in the first embodiment. Differences from the first embodiment will be described below.

<Relationship between lighting order of light sources and color shift>
FIG. 11 is a diagram illustrating an example of the lighting order of the light sources according to the second embodiment. FIG. 12 is a diagram illustrating an example of color misregistration according to the second embodiment.

As shown in FIG. 11, as the CIS 20 moves in the sub-scanning direction, the lighting control unit 31 performs R light→B light→G light→R light→B light→G light→R at a constant period LT. The light sources 21 are turned on in a predetermined lighting order OR2 of light → . The period LT corresponds to the one-line read period of the CIS 20 . That is, in the lighting order OR2, the lighting control unit 31 starts lighting the R light at the beginning of the period LT, and after lighting the R light for the predetermined lighting time TR2, switches the R light to the B light, and turns the B light on. After lighting for the predetermined lighting time TB2, the B light is switched to the G light, and the G light is lit for the predetermined lighting time TG2 until the end of the cycle LT. Further, the lighting control unit 31 sets the relationship between the lighting times TR2, TB2, and TG2 of the R light, B light, and G light in one period of the period LT to be "TR2=TG2" and "TR2, TG2<TB2."

When the light sources 21 are lit in the lighting order OR2 and the lighting times TR2, TB2, and TG2 as described above, color shift may occur at both the upper end of the black character image and the lower end of the black character image, as in the first embodiment. .

Further, when the light source 21 is lit in the lighting order OR2 and the lighting times TR2, TB2, TG2 as described above, R light and B light are applied to the upper edge adjacent pixel Δ and the lower edge adjacent pixel ▽ as shown in FIG. , a color shift of a specific hue corresponding to each lighting timing of the G light. In FIG. 12, for the sake of simplification, the R light, B light, and G light are shown to have the same length of lighting time. The relationship between the lighting times TR2, TB2 and TG2 of the B light and the G light is "TR2=TG2" and "TR2, TG2<TB2".

In FIG. 12, state A2 shows a state in which the upper end of the black character image arrives at the timing when the R light is turned on, and the lower end of the black character image arrives at the timing when the G light is extinguished. In the state A2, no color shift occurs in the upper edge adjacent pixel Δ or the lower edge adjacent pixel ▽.

State B2 shows a state in which the upper end of the black character image is reached while the R light is on. In state B2, a red color shift occurs in the upper edge adjacent pixel Δ.

State C2 indicates a state in which the upper end of the black character image is reached while the B light is on or the G light is on. In state C2, a magenta color shift occurs in the upper edge adjacent pixel Δ.

State D2 indicates a state in which the lower end of the black character image is reached while the G light is on. In the state D2, a green color shift occurs in the pixels ▽ adjacent to the lower edge.

State E2 indicates a state in which the lower end of the black character image is reached during the lighting of the R light or the lighting of the B light. In the state E2, a cyan color shift occurs in the pixels ▽ adjacent to the lower edge.

In this way, if the light source 21 is lit in the lighting order OR2 and the lighting times TR2, TB2, and TG2, there is a possibility that the pixels Δ adjacent to the upper edge will have a color shift of red or magenta. Pixel ▽ may have a color shift with a green or cyan color. The combination of the red and magenta color shifts in the upper-end adjacent pixel Δ and the green and cyan color shifts in the lower-end adjacent pixel ▽ corresponds to the “second type of color shift”. In addition, regarding the color shift that occurs in the upper edge adjacent pixel Δ, since the occurrence rate of magenta color shift is higher than that of red color shift, the upper edge adjacent pixel Δ has a hue closer to magenta. There is a high possibility that color misalignment will occur. On the other hand, regarding the color shift that occurs in the bottom-portion adjacent pixel ▽, since the rate of occurrence of cyan color shift is higher than that of green color shift, the bottom-portion adjacent pixel ▽ is closer to cyan. There is a high possibility that color shift in hue occurs.

<Processing/Operation of Image Processing Apparatus>
13 and 14 are diagrams illustrating an operation example of the image processing apparatus according to the second embodiment.

In step S105 of FIG. 6, the correction unit 14 corrects the HS partial color shift and the SH partial color shift as follows.

As described above, when the light source 21 is lit in the lighting order OR2 and the lighting times TR2, TB2, and TG2, red or magenta colors are generated as the HS partial color shift in the upper edge adjacent pixels Δ forming the HS switching portion. There is a possibility that a color shift having color occurs, and a color shift having green or cyan color may occur as an SH partial color shift at the lower edge adjacent pixel ▽ forming the SH switching portion. As for the color shift (that is, the HS partial color shift) that occurs in the upper edge adjacent pixel Δ, the occurrence rate of magenta color shift is higher than that of red color shift, and it occurs in the lower edge adjacent pixel ▽. As for the color shift (that is, SH partial color shift), the occurrence rate of cyan color shift is higher than that of green color shift.

Therefore, as in the first embodiment, the correction unit 14 first converts the RGB data of the upper-end adjacent pixel Δ and the lower-end adjacent pixel ▽ into XYZ data according to a known conversion formula, and further converts XYZ data according to a known conversion formula. Convert the data to Lab data.

Next, as shown in FIG. 13, the correction unit 14 corrects the color misregistration of the upper-end adjacent pixel Δ and the lower-end adjacent pixel ▽.

That is, as shown in FIG. 13, the correction unit 14 does not correct the luminance component L of both the upper-end adjacent pixel Δ and the lower-end adjacent pixel ▽.

Further, the correction unit 14 determines that, for the upper-edge adjacent pixel Δ, the hue before correction of the upper-edge adjacent pixel Δ (that is, the hue of the color shift occurring in the upper-edge adjacent pixel Δ) is (a_in, b_in)= If (20, -20), CS31 corrects (a_in, b_in)=(20, -20) to (a_out, b_out)=(2.0, -2.0). The hue specified by (a_in, b_in)=(20, -20) corresponds to magenta, as shown in FIG.

For the upper-end adjacent pixel Δ, for example, when the hue before correction of the upper-end adjacent pixel Δ is (a_in, b_in)=(30, 0), the correction unit 14 adds (a_in, b_in) to CS32. = (30, 0) is corrected to (a_out, b_out) = (15.0, 0). The hue specified by (a_in, b_in)=(30, 0) corresponds to reddish purple, as shown in FIG.

For the upper-end adjacent pixel Δ, for example, when the hue before correction of the upper-end adjacent pixel Δ is (a_in, b_in)=(20, 20), the correction unit 14 adds (a_in, b_in) to CS 33 . = (20, 20) is corrected to (a_out, b_out) = (18.0, 18.0). The hue specified by (a_in, b_in)=(20, 20) corresponds to red, as shown in FIG.

Thus, the hue component reduction rate in case CS31 is greater than the hue component reduction rate in case CS32, and the hue component reduction rate in case CS32 is greater than the hue component reduction rate in case CS33. , the magnitude of the hue component reduction rate for the upper edge adjacent pixel Δ is arranged in the order of large, medium, and small, in the case CS31, in the case CS32, and in the case CS33 (FIGS. 13 and 14).

Further, the correction unit 14 determines that, for the lower-end adjacent pixel ▽, for example, the hue before correction of the lower-end adjacent pixel ▽ (that is, the hue of the color shift occurring in the lower-end adjacent pixel ▽) is (a_in, b_in)= If (-20, -20), CS41 corrects (a_in, b_in) = (-20, -20) to (a_out, b_out) = (-2.0, -2.0). The hue specified by (a_in, b_in)=(-20, -20) corresponds to cyan, as shown in FIG.

For example, when the hue before correction of the lower-end adjacent pixel ▽ is (a_in, b_in)=(-25,-10), the correction unit 14 outputs (a_in, b_in) = (-25, -10) is corrected to (a_out, b_out) = (-12.5, -5.0). The hue specified by (a_in, b_in)=(-25,-10) corresponds to cyan green, as shown in FIG.

In addition, for the lower-end adjacent pixel ▽, for example, when the hue before correction of the lower-end adjacent pixel ▽ is (a_in, b_in)=(−30, 0), the correction unit 14 outputs (a_in, b_in ) = (-30, 0) to (a_out, b_out) = (-27.0, 0). The hue specified by (a_in, b_in)=(-30, 0) corresponds to green, as shown in FIG.

Thus, the hue component reduction rate in case CS41 is greater than the hue component reduction rate in case CS42, and the hue component reduction rate in case CS42 is greater than the hue component reduction rate in case CS43. , and lower-end adjacent pixels ▽ are arranged in the order of large, medium, and small, in the order CS41, CS42, and CS43 (FIGS. 13 and 14).

In FIG. 14, as in FIG. 9, the hue range represented by the two-dimensional coordinates of the a-axis (±a) and the b-axis (±b) is divided into 10 by 36°. A case where each hue is assigned to each of the predetermined ranges is shown as an example. In the example shown in FIG. 14, similarly to FIG. 9, for example, the range of 108° or more and less than 144° corresponds to magenta, the range of 144° or more and less than 180° corresponds to bluish purple, and the range of 180° or more and 216° corresponds to magenta. The range below corresponds to blue. Further, for example, in the example shown in FIG. 14, similar to FIG. 9, the range of 324° or more and less than 360° corresponds to yellow, the range of 288° or more and less than 324° corresponds to yellowish green, and the range of 252° or more and 288° corresponds to yellow. The range below corresponds to green.

The correction unit 14 corrects the chroma component as shown in FIG. This is done using table TB2. The hue conversion table TB2 is stored in the storage unit 11 in advance.

Then, the correction unit 14 inversely transforms the Lab data of the upper-end adjacent pixel Δ and the lower-end adjacent pixel ▽ after the saturation component correction into XYZ data according to a known conversion formula, and converts the XYZ data to XYZ data according to a known conversion formula. Convert back to RGB data.

As described above, the correction unit 14 reduces the saturation of magenta, magenta, red, cyan, cyan green, and green, which are specific hues that the second type of color shift may have. The second type of color misregistration is corrected by causing the

In addition, the correction unit 14 sets the reduction rate according to the saturation of magenta, magenta, red, cyan, cyan green, and green, which are specific hues that the second type of color shift may have. to reduce saturation.

The second embodiment has been described above.

In Examples 1 and 2, the smoothing unit 12 is omitted from the image processing apparatus 10, and the detection unit 13 detects the read image stored in the storage unit 11, that is, the read image in which the saturation is not smoothed. The image may be processed in step S103 (FIG. 6). For example, when excessive color misregistration that is difficult to correct by the correction unit 14 occurs in the read image, the smoothing unit 12 smoothes the saturation of the read image as preprocessing for correction by the correction unit 14. It is good to do Further, for example, when it is desired to compress the size of the image file of the read image, it is preferable to smooth the saturation of the read image.

As described above, in the first and second embodiments, the image processing apparatus 10 performs image processing on the read image. The CIS 20 has a plurality of imaging elements 22 arranged in a row in the main scanning direction of the CIS 20, and a light source 21 capable of irradiating a medium to be read with blue, red, and green lights in a predetermined lighting order. The image processing apparatus 10 also has a storage unit 11 , a detection unit 13 , and a correction unit 14 . The storage unit 11 stores read images. The detection unit 13 detects the hue switching portion in the read image. The correction unit 14 corrects the color shift occurring in the hue switching portion, which corresponds to the mode of hue switching in the hue switching portion and the predetermined lighting order of the blue, red, and green lights.

By doing so, it is possible to reduce color misregistration that occurs in the read image. In addition, by correcting the color shift according to the mode of hue switching in the hue switching portion and the predetermined lighting order of the blue, red, and green lights, the color shift whose hue is limited is mainly corrected. Therefore, the effect of reducing color misregistration can be enhanced.

In Embodiments 1 and 2, the detection unit 13 uses, as hue switching portions, an HS switching portion for switching from background pixels to content pixels in the sub-scanning direction and an SH switching portion for switching from content pixels to background pixels in the sub-scanning direction. Detect parts.

In addition, in the first embodiment, the correction unit 14 corrects the HS partial color shift with magenta or blue and the SH partial color shift with yellow or green.

In addition, in the second embodiment, the correction unit 14 corrects the HS partial color shift having magenta or red color and the SH partial color shift having cyan or green color.

By doing so, color misregistration correction is performed only for the HS switching portion and the SH switching portion in the read image, so color misregistration correction can be performed efficiently. In addition, optimum correction can be performed for each of the HS switching portion and the SH switching portion.

Further, in the first and second embodiments, the correction unit 14 corrects color drift by reducing the saturation of a specific hue according to the type of color drift.

By doing so, it is possible to efficiently correct the color misregistration.

In addition, in the first and second embodiments, the correction unit 14 reduces the saturation of a specific hue according to the type of color misregistration at a reduction rate according to the saturation of a specific hue according to the type of color misregistration. Reduce.

By doing so, it is possible to intensively correct the color shift whose hue is limited according to the type of color shift.

Further, in the first and second embodiments, the image processing apparatus 10 has the smoothing unit 12 . A smoothing unit 12 smoothes the saturation of the read image. Then, the detection unit 13 detects the hue switching portion in the smoothed image.

By doing so, even when excessive color misregistration occurs in the read image, the degree of color misregistration can be kept within a range that can be corrected by the correction unit 14 .

Also, the content image in the first and second embodiments is a black character image.

By doing so, the recognition accuracy of OCR (Optical Character Recognition) can be improved.

[Example 3]
All or part of each process in the above description in the image processing apparatus 10 may be realized by causing a processor of the image processing apparatus 10 to execute a program corresponding to each process. For example, a program corresponding to each process described above may be stored in the storage unit 11, and the program may be read from the storage unit 11 and executed by the processor. The program is stored in a program server connected to the image processing apparatus 10 via an arbitrary network, downloaded from the program server to the image processing apparatus 10 and executed, or stored in a record readable by the image processing apparatus 10 . It may be stored in a medium and read and executed from the recording medium. Examples of recording media readable by the image processing apparatus 10 include memory cards, USB memories, SD cards, flexible disks, magneto-optical disks, CD-ROMs, DVDs, and Blu-ray (registered trademark) disks. A portable storage medium is included. A program is a data processing method written in an arbitrary language and an arbitrary description method, and may be in any format such as source code or binary code. In addition, the program is not necessarily configured in a single unit, but may be distributed as multiple modules or multiple libraries, or cooperate with a separate program represented by the OS to achieve its function. Including things.

The specific form of distribution/integration of the image processing apparatus 10 is not limited to the illustrated one, and all or part of the image processing apparatus 10 can be arranged in arbitrary units according to various additions or functional loads. It can be configured by distributing and integrating functionally or physically.

In Embodiments 1 and 2, the case where the image processing apparatus 10 is installed in the image reading apparatus 1 is taken as an example. However, the image processing apparatus 10 may not be installed in the image reading apparatus, and may exist as a device separate from the image reading apparatus, such as a computer device.

The third embodiment has been described above.

1 image reader 20 CIS
21 light source 22 image sensor 31 lighting control unit 10 image processing device 11 storage unit 12 smoothing unit 13 detection unit 14 correction unit

Claims (5)

Image processing for a read image read by an image sensor having a plurality of image pickup elements arranged in a line in the main scanning direction and a light source capable of irradiating a medium to be read with blue, red, and green light in a predetermined lighting order. An image processing device that performs
a storage unit that stores the read image;
a detection unit that detects a hue switching portion in the sub-scanning direction of the image sensor in the read image;
a correction unit that corrects the color shift that occurs in the switching portion, and that corrects the color shift according to the mode of hue switching in the switching portion and the predetermined lighting order;
and
the predetermined lighting order is the order of blue, red, and green;
the read image includes content pixels forming a content image and background pixels forming a background image other than the content image and adjacent to the content pixels;
The detection unit detects, as the switching portions, a first portion where the background pixels are switched to the content pixels in the sub-scanning direction and a second portion where the content pixels are switched to the background pixels in the sub-scanning direction,
The correcting unit corrects the first color misregistration occurring in the first portion, the first color misregistration having magenta, blue, or bluish purple, and the second color misregistration occurring in the second portion. correcting the second color shift having yellow, green or yellow-green,
In the correction of the first color shift, the reduction rate of the hue component in the magenta color shift is greater than the reduction rate of the hue component in the blue-purple color shift, and the reduction of the hue component in the blue-purple color shift. is greater than the reduction rate of the hue component in the blue color shift, and the reduction rate of the hue component in the yellow color shift is greater than the reduction rate of the hue component in the yellow-green color shift, and the yellow-green The reduction rate of the hue component in the color shift is greater than the reduction rate of the hue component in the green,
Image processing device. a smoothing unit that smoothes the saturation of the read image;
The detection unit detects the switching portion in the read image after smoothing the saturation.
The image processing apparatus according to claim 1. The content image is an image of black characters,
The image processing apparatus according to claim 1 . Image processing for a read image read by an image sensor having a plurality of image pickup elements arranged in a line in the main scanning direction and a light source capable of irradiating a medium to be read with blue, red, and green light in a predetermined lighting order. a method,
detecting a hue switching portion in the sub-scanning direction of the image sensor in the read image;
correcting the color shift that occurs in the switching portion according to the switching mode of the hue in the switching portion and the predetermined lighting order;
the predetermined lighting order is the order of blue, red, and green;
the read image includes content pixels forming a content image and background pixels forming a background image other than the content image and adjacent to the content pixels;
detecting, as the switching portions, a first portion where the background pixels are switched to the content pixels in the sub-scanning direction and a second portion where the content pixels are switched to the background pixels in the sub-scanning direction;
The first color misregistration occurring in the first portion, the first color misregistration having magenta, blue, or bluish purple, and the second color misregistration occurring in the second portion, the yellow , correcting the second color shift having green or yellow-green,
In the correction of the first color shift, the reduction rate of the hue component in the magenta color shift is greater than the reduction rate of the hue component in the blue-purple color shift, and the reduction of the hue component in the blue-purple color shift. is greater than the reduction rate of the hue component in the blue color shift, and the reduction rate of the hue component in the yellow color shift is greater than the reduction rate of the hue component in the yellow-green color shift, and the yellow-green The reduction rate of the hue component in the color shift is greater than the reduction rate of the hue component in the green,
Image processing method. Image processing for a read image read by an image sensor having a plurality of image pickup elements arranged in a line in the main scanning direction and a light source capable of irradiating a medium to be read with blue, red, and green light in a predetermined lighting order. and
detecting a hue switching portion in the sub-scanning direction of the image sensor in the read image;
correcting the color shift that occurs in the switching portion according to the switching mode of the hue in the switching portion and the predetermined lighting order;
An image processing program for causing a processor to execute the image processing ,
the predetermined lighting order is the order of blue, red, and green;
the read image includes content pixels forming a content image and background pixels forming a background image other than the content image and adjacent to the content pixels;
detecting, as the switching portions, a first portion where the background pixels are switched to the content pixels in the sub-scanning direction and a second portion where the content pixels are switched to the background pixels in the sub-scanning direction;
The first color misregistration occurring in the first portion, the first color misregistration having magenta, blue, or bluish purple, and the second color misregistration occurring in the second portion, the yellow , correcting the second color shift having green or yellow-green,
In the correction of the first color shift, the reduction rate of the hue component in the magenta color shift is greater than the reduction rate of the hue component in the blue-purple color shift, and the reduction of the hue component in the blue-purple color shift. is greater than the reduction rate of the hue component in the blue color shift, and the reduction rate of the hue component in the yellow color shift is greater than the reduction rate of the hue component in the yellow-green color shift, and the yellow-green The reduction rate of the hue component in the color shift is greater than the reduction rate of the hue component in the green,
Image processing program .

Images