JP2001169097A - Picture processor and picture processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To execute resolution conversion without color slurring, and the blotting and blurring of a color and to execute an interpolation processing corresponding to the partial feature of a picture. SOLUTION: Multivalue color picture data shown by first color space (RGB) are converted into second color space (L*a*b*) in a first color conversion processing part 4. A peripheral pixel selection part 5 selects plural peripheral pixels for using them for the operation of the pixel value of the pixel being the object of interpolation from the existed pixel at the periphery of the pixel being the object of interpolation. In a luminance relation operation part 6 and a color relation ration part 7, the color relation operation part 7 decides an operation method for operating the pixel value of the pixel being the object of interpolation based on luminance information and/or color information on the peripheral pixel. A pixel interpolation part 8 interpolates the pixel being the object of interpolation by using the decided interpolation operation.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus and an image processing method capable of realizing resolution conversion without color shift.

[0002]

2. Description of the Related Art Conventionally, in resolution conversion, Nearest Neighbor (NN), Bi-l
inear (hereinafter BL), Cubic convolution (hereinafter CC)
And the like. 15 is NN, FIG. 16 is BL, FIG.
Shows examples of CC.

NN is a pixel value of an existing pixel closest to the pixel to be interpolated, which is the pixel value of the interpolated pixel. The operation is simplest and the processing amount is small, so that high-speed processing can be performed.
However, the image quality after conversion deteriorates due to generation of a step-like shaggy in an oblique line or the like. In FIG. 15, when there are four existing pixels obliquely around the pixel to be interpolated, the pixel value of the pixel at the upper left position is always interpolated, and the existing pixel is vertically shifted around the pixel to be interpolated. Alternatively, when there are two pixels on the left and right, the pixel value of the pixel at the upper or left position is interpolated.

[0004] BL is an average of weighted values of four existing pixels surrounding a pixel to be interpolated (when four existing pixels are obliquely present around the pixel to be interpolated) in a form proportional to the distance from the interpolated pixel. And obtains the pixel value of the interpolation pixel. This interpolation method has a small amount of calculation and is excellent in smooth interpolation. On the other hand, since smooth interpolation is executed in an edge portion or the like, the edge is dull and the image is blurred.
In FIG. 16, when two existing pixels are present vertically or horizontally around the pixel to be interpolated, an average of these two pixels is determined and the pixel value of the interpolated pixel is used.

In the CC, an interpolation operation is performed by using a total of 16 pixels including 4 pixels surrounding pixels to be interpolated and 12 pixels surrounding them. Although the specific calculation method is omitted,
Of the three methods, it is possible to perform the most complicated operation and prevent smooth interpolation and edge dullness.
Since the calculation is complicated, it is costly to implement the hardware.

[0006] In these conventional resolution conversion methods, if the input image is an RGB color image, interpolation calculation is usually performed using the RGB values of each pixel.

[0007]

In the above-described conventional resolution conversion method, there is no particular problem in the case of converting a grayscale image. Two approaches (ie,
BL and CC) may cause the following problems.

For example, assuming that the upper two points are red and the lower two points are cyan among the four surrounding points in BL, these two colors have a complementary color relationship. You should be able to choose the color. However, in the case of BL, since the interpolation point is obtained linearly, the color becomes gray having RGB values between red and cyan. As a result, an extra color is generated at the boundary between red and cyan. However, it is difficult to determine, from the RGB values, how much the colors of the peripheral pixels referred to during the interpolation have a complementary color relationship.

In image processing, if it is difficult to perform processing in the RGB color space, color conversion may be performed to a uniform color space, and image processing may be performed in the uniform color space. In the case of a complementary color relationship, when linear interpolation is performed on the peripheral pixels, a gray scale is obtained as an interpolated pixel value, and correct interpolation may not be performed. For this reason, there is a method using BL for luminance and NN for color, but color deviation may still occur.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and an object of the present invention is to make it possible to suppress the generation of unnecessary colors when converting the resolution of input image data. An object of the present invention is to provide an image processing apparatus and an image processing method that can suppress color shift, color blur, and the like.

[0011]

In order to solve the above-mentioned problems, an image processing apparatus of the present invention converts multi-valued color image data represented by a first color space into the first color space and the first color space. Is first color conversion means for converting into data composed of luminance information and color information expressed in a different second color space, and calculation of the pixel value of the interpolation target pixel from existing pixels around the interpolation target pixel Selecting means for selecting a plurality of calculation pixels used for the calculation, and calculating a calculation method for calculating a pixel value of the interpolation target pixel based on luminance information and / or color information of the calculation pixel. It is characterized by including a method determining means and an interpolating means for performing an interpolation process on the interpolation target pixel by using the arithmetic method determined by the arithmetic method determining means.

Further, the image processing apparatus further comprises a second color conversion means for converting the image data represented in the second color space into a third color space different from the second color space. May be provided.

Preferably, the first color conversion means converts the multi-valued color image data represented in the first color space into a uniform color space as a second color space. .

For example, when an input device such as a scanner
When multi-valued color image data expressed in a color space or the like is subjected to resolution conversion, simply performing an interpolation operation for each color may result in extra colors, color shift and color bleeding. Can occur.

On the other hand, according to the above configuration, the first
By converting image data expressed in a color space (for example, an RGB color space) into a second color space in which the relationship between luminance and color can be easily obtained, an interpolation operation is performed. Different interpolation calculations can be performed, and resolution conversion without color shift, color bleeding, or blurring can be performed.

At this time, a calculation pixel is selected from existing pixels around the interpolation target pixel, and the interpolation calculation method for the interpolation target pixel is changed based on luminance information and / or color information of the calculation pixel. By doing so, the interpolation processing can be performed by an appropriate interpolation calculation method according to the partial features of the image.

When the interpolation processing for all the pixels to be interpolated is completed and the resolution-converted image data is output or displayed on an output device (or display device) such as a printer or a display, the luminance information and the color information are output. Output processing cannot be performed in the data format represented by the second color space including information, and the data needs to be converted to a third color space such as a CMY color space. Here, the process of converting the image data represented in the second color space into the third color space can be performed on the output device side, but the conversion process is performed on the image processing device side. Thus, the output processing of the image data can be performed even in an output device having no such conversion function.

Further, in the image processing apparatus, the uniform color space is used as the second color space, so that the selected operation pixels have a complementary color relationship or a gray scale and a color relationship. This can be easily obtained from the angle and distance of each calculation pixel from the center point of the uniform color space, and an appropriate interpolation calculation method can be obtained.

In the above-mentioned image processing apparatus, the calculating method determining means determines, in advance, a luminance difference between the plurality of calculating pixels selected by the selecting means and a calculating pixel closest to the interpolation target pixel. When there is an operation pixel having a value less than the set first threshold value, a linear interpolation method is selected as an interpolation operation method of the luminance value of the interpolation target pixel, and in this case, the interpolation unit is selected by the selection unit. Among the calculated plurality of calculation pixels, the luminance value of the calculation pixel closest to the interpolation target pixel and the luminance difference between the calculation pixel closest to the interpolation target pixel are smaller than a preset first threshold value. It is possible to adopt a configuration in which linear interpolation is performed using the luminance value of the calculation pixel to determine the luminance value of the interpolation target pixel.

According to the above configuration, the linear interpolation is performed using the luminance value of the calculation pixel having a small luminance difference (the luminance difference is less than the first threshold value). This indicates that there is no edge between pixels, and a smooth change can be expressed by linear interpolation.

In the above-mentioned image processing apparatus, the calculating method determining means may determine, among the plurality of calculating pixels selected by the selecting means, a luminance difference with the closest calculating pixel to the interpolation target pixel. When there is no calculation pixel that is less than the first threshold, an interpolation calculation method is selected in which the luminance value of the calculation pixel closest to the interpolation target pixel is substituted as the luminance value of the interpolation target pixel.

That is, when the luminance difference between the selected plurality of calculation pixels is large, an edge exists between the calculation pixels, and the luminance value of the calculation pixel closest to the interpolation target pixel is calculated by the interpolation target pixel. By directly substituting as the luminance value of, the interpolation processing that preserves the edge can be performed.

In the above image processing apparatus, the calculating method determining means indicates a chromaticity value for the calculating pixel closest to the interpolation target pixel among the plurality of calculating pixels selected by the selecting means. The first parameter is equal to or greater than a second threshold set in advance, and the first parameter is equal to or greater than the second threshold set in advance. When there is an operation pixel in which the second parameter indicating the color difference is equal to or greater than a preset third threshold value, a linear interpolation method is selected as an interpolation operation method of the chromaticity value of the interpolation target pixel, and the interpolation means In this case, of the plurality of calculation pixels selected by the selection means, the chromaticity value of the calculation pixel closest to the interpolation target pixel and the first parameter are equal to or greater than a predetermined second threshold value. And Third to second parameter indicating a color difference between the closest basis pixel to the interpolation target pixel is set in advance
It is possible to adopt a configuration in which linear interpolation is performed using the chromaticity values of the calculation pixels that are equal to or larger than the threshold value of, and the chromaticity values of the interpolation target pixels are determined.

According to the above configuration, the chromaticity of the calculation pixel closest to the pixel to be interpolated is large (the first parameter is equal to or larger than the second threshold value). When there are calculation pixels having colors similar to the close calculation pixels, a smooth change can be expressed by obtaining the chromaticity of the interpolation target pixel by linear interpolation using only these calculation pixels. Here, the first parameter can be represented by, for example, a distance between a center point on the chromaticity diagram and a point indicating the position of the calculation pixel on the chromaticity diagram.

In the image processing apparatus, the calculating method determining means indicates a chromaticity value for a calculating pixel closest to the interpolation target pixel among a plurality of calculating pixels selected by the selecting means. The first parameter is equal to or greater than a second threshold set in advance, and the first parameter is equal to or greater than the second threshold set in advance. If there is no calculation pixel for which the second parameter indicating the color difference is equal to or greater than the third threshold value set in advance, the chromaticity value of the calculation pixel closest to the interpolation target pixel is set as the chromaticity value of the interpolation target pixel. Select the interpolation calculation method to be substituted.

That is, when the chromaticity difference between the selected plurality of calculation pixels is large, an edge exists between the calculation pixels, and the chromaticity value of the calculation pixel closest to the interpolation target pixel is calculated by the interpolation. By substituting the chromaticity value of the target pixel as it is, interpolation processing that preserves edges can be performed.

In the above image processing apparatus, the calculation method determining means indicates a chromaticity value for a calculation pixel closest to the interpolation target pixel among the plurality of calculation pixels selected by the selection means. When the first parameter is less than the second threshold set in advance and there is another calculation pixel in which the first parameter is less than the second threshold set in advance, the chromaticity of the pixel to be interpolated is A linear interpolation method is selected as a value interpolation calculation method. In this case, the interpolation means selects the chromaticity of the calculation pixel closest to the interpolation target pixel among the plurality of calculation pixels selected by the selection means. Performing linear interpolation using the value and the chromaticity value of the calculation pixel in which the first parameter is less than a preset second threshold value;
The chromaticity value of the interpolation target pixel may be determined.

According to the above configuration, when the chromaticity of the calculation pixel closest to the pixel to be interpolated is small (the first parameter is smaller than the second threshold value) and there is another calculation pixel with a small chromaticity By obtaining the chromaticity of the pixel to be interpolated by linear interpolation using only the calculation pixels having a small chromaticity, a smooth change can be expressed.

In the image processing apparatus, the calculating method determining means indicates a chromaticity value for the calculating pixel closest to the interpolation target pixel among the plurality of calculating pixels selected by the selecting means. If the first parameter is less than the preset second threshold and there is no other calculation pixel where the first parameter is less than the preset second threshold, the pixel closest to the interpolation target pixel is An interpolation calculation method in which the chromaticity value of the calculation pixel is substituted as the chromaticity value of the interpolation target pixel is selected.

That is, when the chromaticity difference between the selected plurality of calculation pixels is large, an edge exists between the calculation pixels, and the chromaticity value of the calculation pixel closest to the interpolation target pixel is calculated by the interpolation. By substituting the chromaticity value of the target pixel as it is, interpolation processing that preserves edges can be performed.

In order to solve the above-mentioned problems, the image processing method of the present invention converts multi-valued color image data represented in a first color space into a second color image data different from the first color space. A first step of converting into data composed of luminance information and color information expressed in space, and a plurality of operations for calculating pixel values of the interpolation target pixel from existing pixels around the interpolation target pixel A second step of selecting a pixel; a third step of determining a calculation method when calculating a pixel value of the interpolation target pixel based on luminance information and / or color information of the calculation pixel; And a fourth step of calculating the interpolation target pixel using the calculation method determined in the third step. By repeating the first to fourth steps, An image whose resolution has been converted by interpolation It is characterized by obtaining the data.

The image processing method may further include converting the resolution-converted image data represented in the second color space into
It may be configured to include a fifth step of converting to a third color space different from the second color space.

In the first step, it is preferable to use a uniform color space as the second color space.

For example, if an input device such as a scanner
When multi-valued color image data expressed in a color space or the like is subjected to resolution conversion, simply performing an interpolation operation for each color may result in extra colors, color shift and color bleeding. Can occur.

On the other hand, according to the above configuration, the first
By converting image data expressed in a color space (for example, an RGB color space) into a second color space in which the relationship between luminance and color can be easily obtained, an interpolation operation is performed. Different interpolation calculations can be performed, and resolution conversion without color shift, color bleeding, or blurring can be performed.

At this time, a calculation pixel is selected from the existing pixels around the interpolation target pixel, and the interpolation calculation method for the interpolation target pixel is changed based on luminance information and / or color information of the calculation pixel. By doing so, the interpolation processing can be performed by an appropriate interpolation calculation method according to the partial features of the image.

When the interpolation processing for all the pixels to be interpolated is completed and the resolution-converted image data is output or displayed on an output device (or display device) such as a printer or a display, the luminance information and the color information are output. Output processing cannot be performed in the data format represented by the second color space including information, and the data needs to be converted to a third color space such as a CMY color space. Here, the image processing method includes a fifth step of converting the image data represented in the second color space into the third color space, thereby enabling output processing.

Further, in the above-described image processing method, by using a uniform color space as the second color space, the selected operation pixels may have a complementary color relationship or a gray scale and a color relationship. This can be easily obtained from the angle and distance between two points of each peripheral pixel from the center point of the uniform color space, and an appropriate interpolation calculation method can be obtained.

[0039]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to FIGS.

FIG. 1 shows a schematic configuration of the image processing apparatus according to the present embodiment.

The image processing apparatus 1 performs resolution conversion on original image data (multi-valued color image data) input from the input device 2 and outputs the converted image data as an output image by the output device 3. It is to let.
Although the image processing device 1 shown in FIG. 1 does not include the input device 1 and the output device 3, the image processing device of the present invention is provided as an image forming device having an input unit and an output unit. It may be.

The image processing apparatus 1 includes a first color conversion processing section (first color conversion means) 4, a peripheral pixel selection section (selection means).
5. Luminance relation calculation unit (calculation method determination unit) 6, Color relation calculation unit (calculation method determination unit) 7, Pixel interpolation unit (interpolation unit)
8 and a second color conversion processing unit (second color conversion means) 9.

The first color conversion processing unit 4 performs a color space conversion process on image data input from the input device 2 and, for example, an RGB color space (first color space).
Is converted into color data in a uniform color space. The peripheral pixel selection unit 5 selects a pixel to be used as information for obtaining an interpolated pixel value from existing pixels adjacent to a pixel position to be interpolated.

The luminance relationship calculating section 6 obtains a positional relationship in the color space of the selected existing pixel in the color space, and determines an appropriate interpolation calculating method for the luminance for each interpolation pixel based on the positional relationship. . Further, the color relation calculation unit 7
Calculates the color positional relationship of the selected peripheral pixels in the color space, and determines an appropriate interpolation calculation method for the color for each interpolation pixel based on the positional relationship.

The pixel interpolation unit 8 performs an interpolation process using the interpolation calculation method determined by the luminance relationship calculation unit 6 and the color relationship calculation unit 7.

The second color conversion processing section 9 repeats the processing from the peripheral pixel selection section 5 to the pixel interpolation section 8 for all the interpolation pixels of the input image data, The data is converted back to the original color space and output to the output device as an output image.

Next, an example of a processing method in the image processing apparatus will be described with reference to a flowchart of FIG.
Here, an example is described in which 8-bit RGB color data for each color is used as an input image, and an L ^* a ^* b ^* color space (second color space) is used as a color space to be converted by the first color conversion processing unit 4. .

First, when original image data composed of RGB color data is input from the input device 2 to the image processing device 1 (S1), the first color conversion processing section 4 converts the color space of the original image data into an RGB color space. Is converted to an L ^* a ^* b ^* color space (first step: S2).

Here, in order to perform interpolation processing on the image data converted into the L ^* a ^* b ^* color space, the peripheral pixel selection unit 5 determines a pixel to be interpolated, and An existing pixel existing around the pixel to be interpolated (hereinafter referred to as a peripheral pixel) for determining a pixel value is extracted (S
3). As an example, as shown in FIG. 3A, four existing pixels (indicated by oblique hatching in the figure) surrounding the interpolation target pixel (indicated by black hatching in the figure) are set as peripheral pixels. However, this is a case where the interpolation target pixel does not exist in the same pixel column as the existing pixel, and as shown in FIG. 3B or FIG. 3C, the interpolation target pixel exists in the same pixel column as the existing pixel. In such a case, two existing pixels that exist so as to sandwich the interpolation target pixel may be set as peripheral pixels. Further, the peripheral pixel selection unit 5 selects a pixel that is close to the target pixel from the extracted peripheral pixels (S4). In this example, S3 and S4 correspond to a second step described in the claims.

Next, in step S5, the interpolation processing is performed based on the luminance information and the color information of the pixel selected in step S4 (hereinafter referred to as “selected pixel”) by the processing in the luminance relation calculation section 6 and the color relation calculation section 7. An appropriate interpolation calculation method for the target pixel is determined (third step: S5). The pixel interpolation unit 8 performs pixel interpolation using the interpolation calculation method determined by the processing of S5 (fourth step: S6). The process for determining the interpolation calculation method will be described later.

The interpolation processing from S3 to S6 is processing for a certain interpolation target pixel. The interpolation target pixel sequentially moves in a predetermined order, and the interpolation processing is executed for all the interpolation pixels. Is done. For this reason, when the above S6 is completed, it is confirmed whether or not the interpolation processing has been completed for all the interpolated pixels (S7). If not, the next interpolated pixel is set as the interpolation target pixel (S8). S3
Move to

In step S7, if the interpolation processing has been completed for all the interpolated pixels, the image data after the interpolation processing represented in the L ^* a ^* b ^* color space is again converted into the RGB color space (the (The third color space) is performed in the second color conversion processing section 9 (fifth step: S9). The converted image thus returned to the RGB color space is output to the output device 3 again.
(S10) to obtain an output image whose resolution has been converted.

Next, the processing in each section of the image processing apparatus 1 will be described in more detail.

First, in the first color conversion processing unit 4, the RGB color
The input image data expressed in space is represented by L ^*a^*b^*Color space
Convert to data. Here, the above L^*a^*b^*Color space
Is one of the uniform color spaces and is perceived as equal
Color differences are intended to correspond to equal distances in space.
Color space. This allows any color
However, the extent to which the colors differ can be calculated from the coordinates of the color space.
Etc. can be compared.

In order to convert data in the RGB color space into data in the L ^* a ^* b ^* color space, RGB and L ^* a
^The color space can be converted by tabulating the relationship with ^* b ^* for some representative colors and performing an interpolation operation using table data of the representative colors close to the input pixel values. For this reason, in the RGB color space, the R value, the G value,
A representative color is selected so that each of the values is arranged at regular intervals, and the color is measured as L ^* a ^* b ^* as a color patch, which is stored in a memory as a look-up table as shown in FIG. Store it.

More specifically, for example, data of an input pixel (existing pixel) whose coordinates in the RGB color space are represented by (r, g, b) is converted into data of the L ^* a ^* b ^* color space. In this case, the representative color closest to the input pixel value is selected. The representative colors are colors corresponding to the vertices of a cube obtained by dividing the three-dimensional space around each of the RGB colors at equal intervals, and as shown in FIG.
When represented by (r, g, b), the vertices p _{000 to} p _{111 of the} cube including the point p therein are selected as the representative colors. Here, (r, g, b) = (150, 104, 3)
Assuming that 2), based on the table data in FIG. 4, at the coordinates of the points p _{000 to} p ₁₁₁ , r ₀ = 12
8, r ₁ = 192, g ₀ = 64, g ₁ = 128, b ₀ =
0, b ₁ = 64.

The data of each representative color selected in this way is
L based on the table data of FIG. ^*a^*b^*Color space
Converted to data. For example, the point p₀₀₀(R₀, G₀,
b₀) = (128,64,0) is (L^*, A^*,
b^*) = (61.54, 13.78, 61.72)
Is replaced. Similarly, the color of other selected representative colors will change.
Is replaced.

The converted data of the input pixel value is
It is expressed by the following equation (1). Here, q in the following equation indicates a corresponding point in the L ^* a ^* b ^* color space after the point p is subjected to color conversion, and is represented by coordinates (L ^* , a ^* , b ^* ). You.

[0059] ^{q (L *, a *,} b *) = c 0 + c 1 Δr + c 2 Δg + c 3 Δb + c 4 ΔrΔg + c 5 ΔrΔb + c 6 ΔgΔb + c 7 ΔrΔgΔb ... (1) _{where, Δr = r-r 0 =} 150- 128 = 22 Δg = g−g ₀ = 104−64 = 40 Δb = b−b ₀ = 32−0 = 32 c ₀ = p ₀₀₀ c ₁ = (p ₁₀₀ −p ₀₀₀ ) / (r ₁ −r ₀ ) c ₂ = (p ₀₁₀ -p ₀₀₀ ) / (g ₁ -g ₀ ) c ₃ = (p ₀₀₁ -p ₀₀₀ ) / (b ₁ -b ₀ ) c ₄ = (p ₁₁₀ -p ₀₁₀ -p ₁₀₀ + p ₀₀₀₎ ) / {(R ₁ −r ₀ ) · (g ₁ −g ₀ )} c ₅ = (p ₁₀₁ −p ₀₀₁ −p ₁₀₀ + p ₀₀₀ ) / ｛(r ₁ −r ₀ ) · (b ₁ −b ₀₎ )｝ C ₆ = (p ₀₁₁ -p ₀₀₁ -p ₀₁₀ + p ₀₀₀ ) / {(g ₁ -g ₀ ) · (b ₁ -b ₀ )} c ₇ = (p ₁₁₁ -p ₀₁₁ -p ₁₀₁ -p _{110 + p 100 + p 001 + p} 010 -p 000) / { _{r 1 -r 0) · (g} 1 -g 0) is a _{· (b 1 -b 0)}} . Thereby, (150, 10) on the RGB color space
The input pixel value of (4, 32) is represented by coordinates (72.85, 8.44, 69.70) in the L ^* a ^* b ^* color space. Such processing is performed for all input pixels of the input image data.

The peripheral pixel selecting section 5 first extracts four peripheral pixels surrounding the pixel to be interpolated from the existing pixels.
As a method for extracting the peripheral pixels, as shown in FIG.
When the position of the pixel to be interpolated is based on the position information of the existing pixel, and the position of the pixel to be interpolated is (x, y),
([X], [y]), ([x] +1, [y]),
([X], [y] +1), ([x] +1, [y] +1)
Becomes an extraction pixel. Here, [x] and [y] represent integer parts of x and y.

Next, the peripheral pixel selecting section 5 selects two pixels close to the position of the pixel to be interpolated from the four peripheral pixels as selected pixels. Therefore, the distance between the four neighboring pixels and the pixel to be interpolated is calculated using the equation (nx−x) ² + (ny−y) ² . Here, the x and y coordinates of the peripheral pixels for which the distance from the interpolation target pixel is obtained are substituted for nx and ny in the above equation. Then, based on the result of this equation, two pixels having small values are selected as selected pixels. That is, here, the selected pixel corresponds to the calculation pixel described in the claims.

For example, when the position of the pixel to be interpolated is set to (3.4, 1
0.6), the extracted peripheral pixels are (3,1)
0), (4,10), (3,11) and (4,11). Then, when the distance between these peripheral pixels and the pixel to be interpolated is calculated using the above equation,
0.52, 0.72, 0.32, and 0.52. The smaller this value is, the closer it is to the position of the pixel to be interpolated. Therefore, first, the peripheral pixels (3, 11) are selected. Further, the value of the pixel that is the next selection candidate is 0.52, and such peripheral pixels are 2 (3, 10) and (4, 11).
Pixel exists. Here, priorities are assigned in the order of upper left, upper right, lower left, and lower right, and (3, 10) is selected. Thereby, (3,10) and (3,1)
This means that the peripheral pixel in 1) is selected as the selected pixel.

Next, the brightness-related calculation unit 6 determines an appropriate interpolation calculation method regarding brightness for the two selected pixels selected by the peripheral pixel selection unit 5 using the values of L ^* . Here, the processing in the luminance relation calculation unit 6 will be described with reference to the flowchart in FIG.
Note that the flowchart of FIG. 7 shows the processing related to the luminance information in S5 of the flowchart of FIG. 2 in detail.

First, as shown in FIG. 8, the luminance relation calculation unit 6 calculates the L ^* values of these two selected pixels as L
^* ₁ , L ^* ₂ (L ^* ₁ > L ^* ₂ ), and ₂ on the L ^* axis
The distance L ^* ₁₋ L ^* ₂ between pixels is obtained (S11). Here, the distance L ^* ₁₋ L ^* ₂ indicates the difference in luminance between these two selected pixels. If the difference in luminance is large, it can be said that an edge exists in that portion. Therefore, the interpolation processing method is changed depending on whether or not L ^* ₁₋ L ^* ₂ is equal to or greater than a certain threshold.

That is, in S12, L^* ₁-L^*
_TwoIs a predetermined threshold (first threshold: here, threshold 40)
But the threshold value is not limited), and L^* ₁−
L^* _TwoIs smaller than the threshold value (when the brightness difference is small)
Is a linear expression using the luminance value of the selected pixel as shown in FIG.
By performing interpolation (S13), a smooth luminance change can be obtained.
Can be expressed. Conversely, L^* ₁-L^* _TwoIs greater than or equal to the threshold
Does not blur the edge, as shown in FIG.
Assign the luminance value of the selected pixel closest to the interpolation pixel to the interpolation pixel
(S14).

For example, the L of the selected pixel of (3, 10)
^When the value of ^* a ^* b ^* is (72.85, 8.44, 69.7)
0), when the value of L ^* a ^* b ^* of the selected pixel of (3, 11) is (52.85, 10.44, 39.70), L ^* ₁₋ L ^* ₂ = 72.85. -52.85 = 20. Then, since the value of L ^* ₁₋ L ^* ₂ is less than the threshold value 40, it is determined that the linear interpolation processing should be performed on the luminance.

Further, the color relation calculation section 7 determines an appropriate color interpolation calculation method for the two selected pixels selected by the peripheral pixel selection section 5 by using their a ^* and b ^* values. I do. Here, the processing in the color relation calculation unit 7 will be described with reference to the flowchart in FIG.
It should be noted that the flowchart of FIG. 11 shows in detail the processing relating to the color information in S5 of the flowchart of FIG.

First, the color relation calculation section 7 shown in FIG.
As shown in the chromaticity diagram (a^*-B^*Two choices on the plane)
As the positional relationship of pixels, a^*-B^*The center point of the plane and the point q
₁, Q_TwoAngle consisting of two straight lines formed by connecting
(Inner angle) θ and a ^*-B^*Center point of plane
(A^*= 0, b^*= 0) corresponds to two selected pixels
Point q₁, Q_TwoDistance l of a straight line connecting each₁, L_Two
(S21, S22). And the distance l₁,
l_TwoAnd the angle θ with a certain threshold,
Change the interpolation method according to the result.

Here, first, in S23, the distances l ₁ and l ₂ are set to thresholds (second threshold: here, threshold 30
, But the threshold value is not limited). If both the distances l ₁ and l ₂ are smaller than the threshold value, linear interpolation using the chromaticity value of the selected pixel is performed as shown in FIG. (S24). That is, it can be said that a pixel having a short distance from the center point of the a ^* -b ^* plane has a color close to gray with almost no vividness. Therefore, the distances l ₁ and l ₂
Are smaller than the threshold value, it indicates that the two selected pixels are close to each other. In this case, the change in color should be expressed smoothly.
A smooth color change can be expressed by linear interpolation.

If NO in S23, at least one of the distances l ₁ and l ₂ is greater than or equal to the threshold value, and S25
Move to. Here, if _one of the distances l ₁ and l ₂ is greater than or equal to the threshold and the other is less than the threshold, as shown in S10, the interpolation calculation method of substituting the chromaticity value of the selected pixel closest to the interpolation pixel into the interpolation pixel is performed. It shall be used (S26). That is, when only _{one of} the distances l ₁ and l ₂ is smaller than the threshold value, only that color becomes a color close to gray. At this time, if linear interpolation is performed between a certain color and a color close to gray, the boundary line of the color may be blurred or the color may be blurred. This is because it is preferable.

Further, in S25, the distances l ₁ and l
If ₂ are both equal to or greater than the threshold, the angle θ is a threshold value in S27 (in this case, is a threshold 90 °, the value of the threshold is not limited) is compared with, if the angle θ is smaller than the threshold value,
It is assumed that linear interpolation using the chromaticity value of the selected pixel is performed (S24). At this time, if the angle θ is equal to or greater than 90 ° (NO in S27), an interpolation calculation method of substituting the chromaticity value of the selected pixel closest to the interpolation pixel into the interpolation pixel is used (S26). That is, if the angle θ takes a large value, it means that the relationship between the two colors is closer to the complementary color. In this case, it is better to interpolate between the two selected pixels than to use either color. This is because, when the interpolation is performed, the color bleeding and the color shift are less likely to occur.

In step S27, the angle θ is 9
If it is smaller than 0 °, linear interpolation using the chromaticity value of the selected pixel is performed (S24). This is because, when the angle θ is smaller than 90, the two colors do not have a complementary color relationship, and an interpolation method that expresses a smooth color change between colors close to each other is preferable.

For example, similarly to the example of the luminance, (3, 1
0) Pixel L^*a^*b^*Is (72.85, 8.4)
4,69.70), the L of the pixel of (3,11)^*a^*b
^*Is (52.85, 10.44, 39.70)
First, each a^*b ^*Extract only values. Soshi
And a^*-B^*Center point for each pixel on the plane
Euclidean distance l₁, L_TwoAsk for. a formula
Is ｛(a^{* 2}+ B^{* 2})｝ ^1/2, And calculate each
Then, l₁= 70.21 and l_Two= 41.05 and
Become.

Next, the inner angle θ is calculated. First, a straight line formula that is perpendicular to the straight line formula with the longer Euclidean distance and that passes through the point with the shorter Euclidean distance is obtained. In this example, b ^* = 0.12a ^* + 38.45. Find the point of the straight line with the longer Euclidean distance that intersects this equation vertically. Then, a distance from the center point to the vertical intersection is obtained. In this example, the vertical intersection is (4.7
2, 39.1), and the distance from the center point is 3
9.38. Further, when cos θ is obtained from these values, it becomes 0.96. When cos θ> 0, the interior angle is less than 90 degrees, and it can be determined that the two selected pixels have close colors. The distance is also 3 for both pixels
Since there is 0 or more, it is determined that the linear interpolation process is also performed on the color.

In the pixel interpolating section 8, the luminance relation calculating section 6
In addition, according to the interpolation method determined by the color relation calculation unit 7, the luminance and color of the interpolation target pixel are interpolated. The interpolation calculation method used here is set in two ways as shown in FIGS. 9 and 10. One is an existing peripheral pixel closest to the position of the pixel to be interpolated, as in the conventional NN method. The value is directly used as the interpolation target pixel value (see FIG. 10). The other is to use a linear interpolation value of two peripheral pixel values as an interpolation target pixel value as in the conventional BL method (see FIG. 9).

For example, according to the above example, the luminance value of the pixel to be interpolated is obtained by linear interpolation from the L ^* value of the peripheral pixel (72.85−52.85) /2+52.85=62.
It is obtained as 85. Further, the color value of the pixel to be interpolated is calculated by a ^* = (69.70−39.70) /2+39.7 by linear interpolation from the a ^* value and b ^* value of the peripheral pixels.
0 = 54.70 and b ^* = (10.44-8.4
4) It is obtained as + 8.44 = 9.44. Therefore,
The value of L ^* a ^* b ^* of the interpolated pixel is (62.85, 54.
70, 9.44).

Finally, the second color conversion processing section 9 converts the converted L ^* a ^* b ^* data back to the original RGB data. Here, contrary to the processing of the first color conversion processing unit 4,
The relation between RGB and L ^* a ^* b ^* is ^{tabulated for} a representative color of L ^* a ^* b ^* . For this reason, in the L ^* a ^* b ^* color space, a representative color is selected such that the L ^* value, the a ^* value, and the b ^* value are arranged at regular intervals, and a color close to that is used as a color patch in RGB. Figure 1
3 is stored in a memory as a look-up table. Then, the color space is converted by performing an interpolation operation from a representative color close to the converted pixel value using the lookup table, and output as an output image.

When performing color conversion on L ^* a ^* b ^* data, first, a representative color closest to the above data is selected. The representative color is a color corresponding to a vertex of a cube obtained by dividing a three-dimensional space around each color of L ^* a ^* b ^* at equal intervals, and as shown in FIG. 14, the L ^* a ^* b ^* data point ^{q (L *, a *,} b *) when expressed in, each vertex q ₀₀₀ to q ₁₁₁ cube containing the point q therein is selected as the representative color. In these representative colors, L ^* value, a ^* value,
The maximum value of each b ^* value is L ^* ₁ , a ^* ₁ , b ^* ₁ ,
Where the minimum values are L ^* ₀ , a ^* ₀ , b ^* ₀ .

The data of each representative color thus selected is
The data is converted into data in the RGB color space based on the table data in FIG. Then, the converted data of the L ^* a ^* b ^* data is represented by the following equation (2).

P (r, g, b) = c ₀ + c ₁ ΔL ^* + c ₂ Δa ^* + c ₃ Δb ^* + c ₄ ΔL ^* Δa ^* + c ₅ ΔL ^* Δb ^* + c ₆ Δa ^* Δb ^* + c ₇ ΔL ^* Δa ^* Δb ^* (2) where ΔL ^* = L ^* −L ^* ₀ Δa ^* = a ^* −a ^* ₀ Δb ^* = b ^* −b ^* ₀ , and the counts c ₀ to c ₇ ｛P ₀₀₀ , p
₁₀₀ , p ₀₁₀ , p ₀₀₁ , p ₁₁₀ , p ₁₀₁ , p ₀₁₁ , p
₁₁₁ , r ₁ , r ₀ , g ₁ , g ₀ , b ₁ , b ₀ } instead of {q ₀₀₀ , q ₁₀₀ , q ₀₁₀ , q ₀₀₁ , q ₁₁₀ ,
_{q 101, q 011, q 111} , L * 1, L * 0, a * 1, a
^* ₀ , b ^* ₁ , b ^* ₀ } are used. Such processing is performed for all the pixels of the image data.

As described above, each processing is performed, and an output image whose resolution is converted is finally obtained. In the present embodiment, L ^* a ^{* is used} as the color space to be converted ^.
Although the b ^* color space was used, if it is a uniform color space, L ^* a ^* b
^* Needless to say, other than the color space may be used. Further, in the above description, the third color space is the same as the first color space in RGB.
Although the color space is used, the third color space and the first color space need not be the same, and the CMY color space can be used as the third color space.

As a method of converting the color space, a conversion by a matrix operation may be used instead of the conversion by the table data. Further, as a method of obtaining an interpolation value at the time of the table conversion, in the present embodiment, the color space is converted. Is divided into cubes, but may be divided into other shapes such as triangular prisms and pyramids.

Further, in the example described above, the peripheral pixel selecting unit 5 selects four peripheral pixels from the existing pixels around the pixel to be interpolated, and further selects the interpolation pixel from the four peripheral pixels. The pixel closest to the target pixel is selected as a selected pixel, and interpolation processing is performed using the selected pixel. However, in the present invention, it is not always necessary to consider the selected pixel, and the interpolation process may be executed after the peripheral pixels are extracted. That is, in this case, the peripheral pixels correspond to the calculation pixels described in the claims. In this case, the number of peripheral pixels to be extracted is not limited to four. The range of the peripheral pixels is determined in advance based on calculation accuracy, manufacturing cost, and the like. Hereinafter, a method of determining the interpolation processing in the luminance relationship calculation unit 6 and the color relationship calculation unit 7 when only the peripheral pixels are defined will be described.

First, regarding the luminance (L ^* ), the luminance value of the peripheral pixel closest to the pixel to be interpolated is L ^* _1, and the luminance values of the other peripheral pixels are L ^* _{2 to} L ^* _n . As it follows, obtaining the difference between the L ^* ₁ and L ^* ₂ ~L ^* _n.

ΔL ^* ₁ = L ^* ₁₋ L ^* ₂ ΔL ^* ₂ = L ^* ₁₋ L ^* ₃ : ΔL ^* _n−1 = L ^* ₁₋ L ^* _n Then, the obtained difference value ΔL ^* _m ( m: 1 to n-1)
Is compared with a threshold value (first threshold value: for example, 40), and among these difference values ΔL ^* _m , only peripheral pixels satisfying ΔL ^* _m <40 are extracted, and the extracted pixels including L ^* ₁ are extracted. A weighted average based on the ratio of the spatial distance between the peripheral pixel and the target pixel is calculated, and the weighted average is used as the L ^* value of the interpolation target pixel.

For example, in the above difference value, ΔL^* _Twoonly
Is less than the threshold, L^* ₁, L^* _ThreeIs extracted. This
, L of the interpolation target pixel^*The value is L^*= L^* ₁× (1- | xx_L1|) × (1− | y−y
_L1|) + L^* _Three× (1- | xx_L3|) × (1- | y−
y_L3|). If there is no extracted pixel (the difference
If none of the values are below the threshold)
L^* ₁To L of the pixel to be interpolated^*Value.

Next, the chromaticity ^{(C: a *, b *} ) for the chromaticity closest surrounding pixels in the interpolation target pixel and C _1, and the chromaticity of other peripheral pixel C ₂ -C _n I do. And
The C ₁ and C ₂ -C interior angle a color difference between _n the second parameter _{θ m (m: 1~n-1} ) obtained as. here,
theta _m is chromaticity diagram (a ^* -b ^* plane) on the center point of the (a ^*
= B ^* = 0) and C ₁ and C _{m + 1} . Also, C ₁ -C
The respective distances from the center point on the chromaticity diagram at the position _n are obtained as D _{1 to} D _n that are the first parameters.

[0088] Here, the D ₁ threshold (second threshold value: e.g., 30) as compared to, if D ₁ <30, C ₁ is determined to be a gray scale, then, D ₂ to D Similarly, _n is compared with a threshold value. And D _m <30
(M: 2 to n) are extracted, a weighted average is calculated based on a ratio of a spatial distance between the extracted peripheral pixel and the interpolation target pixel, and this is set as the chromaticity of the interpolation target pixel.

If D ₁ ≧ 30, it is determined that C ₁ is a color, and then D _m ≧ 30 (m: 2
~n), and theta _m extracts made with <90 ° (third threshold), extracted peripheral pixel and obtains a weighted average according to the ratio of the spatial distance between the interpolation target pixel, the interpolation target pixel so Chromaticity.

[0090] Further, when the peripheral pixel C ₁ is gray-scale, both for color, if satisfying D _m are extracted, the C ₁ and chromaticity of the interpolation target pixel.

As described above, the image processing apparatus according to the present embodiment converts multi-valued color image data represented by the first color space (RGB) into the second color space (L ^* a ^* b). ^* ) The first color conversion processing unit 4 that converts the data into the data represented by ⁽ ), and a plurality of peripheral pixels (or selected pixels) used for calculating the pixel value of the interpolation target pixel from the existing pixels around the interpolation target pixel
And luminance information (L ^* a ^* b ^* ) and / or color information (a ^* b ^* ) of the peripheral pixels.
And a luminance relation calculation unit 6 and a color relation calculation unit 7 that determine a calculation method when calculating the pixel value of the interpolation target pixel based on the luminance relation calculation unit 6 and the color relation calculation unit 7. A pixel interpolating unit 8 that performs an interpolation process on the interpolation target pixel by using a calculation method.

Further, the image processing apparatus further comprises a second color conversion processing unit for converting the image data represented in the second color space into a third color space different from the second color space. 9 is provided.

It is preferable that the first color conversion processing unit 4 converts multi-valued color image data expressed in the first color space into a uniform color space as a second color space. .

According to the above arrangement, the image data expressed in the first color space is converted into the second color space in which the relationship between luminance and color can be easily obtained, and then interpolation is performed.
Different interpolation calculations can be performed for each relationship between luminance and color, and resolution conversion without color shift, color bleeding, or blurring can be performed.

At this time, a peripheral pixel is selected from existing pixels around the pixel to be interpolated, and the interpolation calculation method for the pixel to be interpolated is changed based on luminance information and / or color information of the peripheral pixel. Accordingly, the interpolation processing can be performed by an appropriate interpolation calculation method according to the partial feature of the image.

When the interpolation processing for all the pixels to be interpolated is completed and the resolution-converted image data is output or displayed on an output device (or display device) such as a printer or a display, the luminance information and the color information are obtained. Output processing cannot be performed in the data format represented by the second color space including information, and the data needs to be converted to a third color space such as a CMY color space. Here, the process of converting the image data represented in the second color space into the third color space can be performed on the output device side, but the conversion process is performed on the image processing device side. Thus, the output processing of the image data can be performed even in an output device having no such conversion function.

Further, in the above image processing apparatus, by using a uniform color space as the second color space, when the selected peripheral pixels have a complementary color relationship or a gray scale and color relationship, This can be easily obtained from the angle and distance of each peripheral pixel from the center point of the uniform color space, and an appropriate interpolation calculation method can be obtained.

Further, in the image processing apparatus, the luminance-related operation unit 6 and the color-related operation unit 7 include, among the plurality of peripheral pixels selected by the peripheral pixel selection unit 5,
When there is a peripheral pixel whose luminance difference from the peripheral pixel closest to the interpolation target pixel is smaller than a preset first threshold value,
A linear interpolation method is selected as an interpolation calculation method for the luminance value of the pixel to be interpolated. In this case, the pixel interpolating unit 8 selects the pixel to be interpolated among the plurality of peripheral pixels selected by the peripheral pixel selecting unit 5. Linear interpolation using the luminance value of the peripheral pixel closest to the pixel and the luminance value of the peripheral pixel whose luminance difference between the peripheral pixel closest to the interpolation target pixel is less than a preset first threshold. A configuration for determining the luminance value of the pixel to be interpolated may be employed.

In this case, the luminance difference is small (the luminance difference is
Linear interpolation is performed using the luminance values of the peripheral pixels. A small luminance difference indicates that there is no edge between these peripheral pixels, and a smooth change is expressed by the linear interpolation. be able to.

Further, in the image processing apparatus, the luminance-related operation unit 6 and the color-related operation unit 7 include, among the plurality of peripheral pixels selected by the peripheral pixel selection unit 5,
The luminance difference between the pixel closest to the interpolation target pixel and the peripheral pixel is the first pixel.
When there is no peripheral pixel having a value less than the threshold value, the interpolation calculation method of substituting the luminance value of the peripheral pixel closest to the interpolation target pixel as the luminance value of the interpolation target pixel is selected.

That is, when the luminance difference between the selected plurality of peripheral pixels is large, an edge exists between the peripheral pixels, and the luminance value of the peripheral pixel closest to the interpolation target pixel is calculated as the luminance value of the interpolation target pixel. By substituting as it is, interpolation processing that preserves edges can be performed.

Further, the peripheral pixel selecting unit 5 further selects two selected pixels near the pixel to be interpolated from the peripheral pixels, and the luminance relation computing unit 6 performs an interpolation operation method relating to the luminance of the pixel to be interpolated. Can be determined using the distance between the two selected pixels on the luminance axis in the uniform color space.

At this time, if the distance between the two selected pixels on the luminance axis of the uniform color space is more than a certain distance, the luminance of the selected pixel closest to the pixel to be interpolated is Let it be luminance. If the distance between the two selected pixels on the luminance axis of the uniform color space is less than a certain value, the luminance of the interpolation target pixel is obtained from the luminance of the two selected pixels by linear interpolation. Shall be.

Although it seems that the luminance information does not directly affect the color shift, even if the color information is close, if the luminance is significantly different, an edge exists between them, and it is linearly interpolated. Then, color bleeding or blurring may occur. For this reason, by using the distance parameter for the luminance, since the difference in brightness increases as the distance between the two selected pixels increases, an edge exists, and any brightness is used as the luminance to be interpolated. It will be easier to understand which is best.

That is, the fact that there is a difference in luminance over a certain level means that there is an edge between them.
Performing the same processing as N can prevent color bleeding and blurring.

Further, in the image processing apparatus, the luminance-related operation unit 6 and the color-related operation unit 7 include, among the plurality of peripheral pixels selected by the peripheral pixel selection unit 5,
A first parameter (for example, a distance between a center point on the chromaticity diagram and a point indicating the position of the peripheral pixel on the chromaticity diagram) indicating the chromaticity value of the peripheral pixel closest to the interpolation target pixel is preset. The second parameter that is equal to or greater than the second threshold that has been set and the first parameter is equal to or greater than the second threshold that is set in advance and that indicates the color difference with the nearest pixel closest to the pixel to be interpolated is When there is a peripheral pixel that is equal to or greater than the third threshold value set in advance, the linear interpolation method is selected as the interpolation calculation method of the chromaticity value of the pixel to be interpolated. The chromaticity value of the peripheral pixel closest to the interpolation target pixel among the plurality of peripheral pixels selected by the pixel selection unit 5 and the first parameter are equal to or greater than a second threshold set in advance, and the interpolation is performed. A second pattern indicating a color difference from a peripheral pixel closest to the target pixel. Performs linear interpolation by using the chromaticity value of the surrounding pixel meter is the third threshold value or more set in advance, it may be configured to determine the luminance value of the interpolation target pixel.

In this case, the chromaticity of the peripheral pixel closest to the pixel to be interpolated is large (the first parameter is equal to or more than the second threshold), and the color of the chromaticity that is large and close to the peripheral pixel closest to the pixel to be interpolated is When there are peripheral pixels, the smooth change can be expressed by obtaining the chromaticity of the interpolation target pixel by linear interpolation using only these peripheral pixels.

In the image processing apparatus, the luminance-related operation unit 6 and the color-relation operation unit 7 include a plurality of peripheral pixels selected by the peripheral pixel selection unit 5.
The first parameter indicating the chromaticity value of the peripheral pixel closest to the interpolation target pixel is equal to or greater than a preset second threshold, and the first parameter is equal to or greater than the preset second threshold. And if there is no peripheral pixel whose second parameter indicating the color difference from the peripheral pixel closest to the interpolation target pixel is equal to or greater than a preset third threshold, the chromaticity of the peripheral pixel closest to the interpolation target pixel An interpolation calculation method for substituting the value as the chromaticity value of the interpolation target pixel is selected.

That is, when the chromaticity difference between the selected plurality of peripheral pixels is large, an edge exists between the peripheral pixels, and the chromaticity value of the peripheral pixel closest to the interpolation target pixel is calculated as By directly substituting the chromaticity value, an interpolation process that preserves edges can be performed.

Further, in the image processing apparatus, the luminance-related operation unit 6 and the color-related operation unit 7 include, among the plurality of peripheral pixels selected by the peripheral pixel selection unit 5,
The first parameter indicating the chromaticity value of the peripheral pixel closest to the interpolation target pixel is less than a preset second threshold, and the first parameter is also less than a preset second threshold. If there is a peripheral pixel, a linear interpolation method is selected as an interpolation calculation method of the chromaticity value of the interpolation target pixel. In this case, the pixel interpolation unit 8 selects the plurality of peripheral pixels selected by the peripheral pixel selection unit 5. Among the pixels, linear interpolation is performed using a chromaticity value of a peripheral pixel closest to the interpolation target pixel and a chromaticity value of a peripheral pixel whose first parameter is less than a preset second threshold value, It is possible to adopt a configuration for determining the luminance value of the interpolation target pixel.

In this case, when the chromaticity of the peripheral pixel closest to the pixel to be interpolated is small (the first parameter is smaller than the second threshold value), and there are other peripheral pixels with small chromaticity, these chromaticities are reduced. By obtaining the chromaticity of the pixel to be interpolated by linear interpolation using only small peripheral pixels, a smooth change can be expressed.

Further, in the image processing apparatus, the luminance-related operation unit 6 and the color-related operation unit 7 include the plurality of peripheral pixels selected by the peripheral pixel selecting unit 5.
The first parameter indicating the chromaticity value of the peripheral pixel closest to the interpolation target pixel is less than a preset second threshold, and the first parameter is also less than a preset second threshold. If there is no peripheral pixel, an interpolation calculation method of substituting the chromaticity value of the peripheral pixel closest to the interpolation target pixel as the chromaticity value of the interpolation target pixel is selected.

That is, if the chromaticity difference between the selected plurality of peripheral pixels is large, an edge exists between the peripheral pixels, and the chromaticity value of the peripheral pixel closest to the interpolation target pixel is calculated as By directly substituting the chromaticity value, an interpolation process that preserves edges can be performed.

Further, the peripheral pixel selecting section 5 further selects two selected pixels close to the interpolation target pixel from the peripheral pixels, and the color relation calculating section 7 performs an interpolation operation on the chromaticity of the interpolation target pixel. When deciding on a method,
The distance (first parameter) of each selected pixel from the center point from the chromaticity diagram plane of the uniform color space, and the angle between straight lines connecting the center point of the chromaticity diagram plane and these selected pixels ( (Second parameter: interior angle).

At this time, if both the first parameters are less than a certain value, or if both the first parameters are more than a certain value and the second parameter is less than a certain angle, , The chromaticity of the interpolation target pixel is 2
It is determined by linear interpolation from the chromaticities of the selected pixels. If both the first parameters are equal to or more than a certain value and the second parameters are equal to or more than a certain angle, or one of the first parameters is equal to or more than a certain value and the other is less than a certain value. In this case, the chromaticity of the selected pixel closest to the interpolation target pixel is set as the chromaticity of the interpolation target pixel.

With the image processing method implemented in the image processing apparatus as described above, it is possible to perform resolution conversion while suppressing color misregistration, color bleeding and blurring, and to obtain a higher quality output image.

[0117]

As described above, the image processing apparatus according to the present invention converts multi-valued color image data represented in the first color space into a second color space different from the first color space. First, which is converted into data represented by luminance information and color information
Color conversion means and existing pixels around the interpolation target pixel,
Selecting means for selecting a plurality of calculation pixels used for calculation of the pixel value of the interpolation target pixel; and calculating the pixel value of the interpolation target pixel based on luminance information and / or color information of the calculation pixel. It is configured to include a calculation method determining means for determining a calculation method at the time, and an interpolation means for performing an interpolation process on the interpolation target pixel using the calculation method determined by the calculation method determination means.

Further, the image processing apparatus further includes a second color conversion means for converting the image data represented in the second color space into a third color space different from the second color space. May be provided.

It is preferable that the first color conversion means converts the multi-valued color image data represented in the first color space into a uniform color space as a second color space. .

Therefore, the first color space (for example, RGB
By converting the image data expressed in (color space) into a second color space in which the relationship between luminance and color can be easily obtained, an interpolation operation is performed, so that a different interpolation operation can be performed for each relationship between luminance and color. This enables resolution conversion without color shift, color bleeding, or blurring.

At this time, the interpolation calculation method is changed based on the luminance information and / or the color information of the calculation pixels selected from the existing pixels around the interpolation target pixel, so that the partial characteristics of the image can be obtained. There is an effect that the interpolation processing can be performed by an appropriate interpolation calculation method.

Further, by using a uniform color space as the second color space, if the selected calculation pixels have a complementary color relationship or a gray scale and color relationship, this is easily obtained. This makes it possible to obtain an appropriate interpolation calculation method.

Further, in the above-mentioned image processing apparatus, the calculating method determining means determines in advance the luminance difference between the calculating pixel closest to the interpolation target pixel among the plurality of calculating pixels selected by the selecting means. When there is an operation pixel having a value less than the set first threshold value, a linear interpolation method is selected as an interpolation operation method of the luminance value of the interpolation target pixel, and in this case, the interpolation unit is selected by the selection unit. Among the calculated plurality of calculation pixels, the luminance value of the calculation pixel closest to the interpolation target pixel and the luminance difference between the calculation pixel closest to the interpolation target pixel are smaller than a preset first threshold value. It is possible to adopt a configuration in which linear interpolation is performed using the luminance value of the calculation pixel to determine the luminance value of the interpolation target pixel.

In this case, linear interpolation is performed using the luminance values of the operation pixels having a small difference in luminance, and when no edge exists between these operation pixels, a smooth change by the linear interpolation can be expressed. .

In the above-mentioned image processing apparatus, the calculating method determining means determines that the luminance difference between the calculating pixel closest to the interpolation target pixel among the plurality of calculating pixels selected by the selecting means is the same. When there is no calculation pixel that is less than the first threshold value, an interpolation calculation method of substituting the luminance value of the calculation pixel closest to the interpolation target pixel as the luminance value of the interpolation target pixel can be selected.

As described above, when the luminance difference between the selected plurality of calculation pixels is large and there is an edge between the calculation pixels, the luminance value of the calculation pixel closest to the interpolation target pixel is determined by the luminance value of the interpolation target pixel. By directly substituting it as a luminance value, an interpolation process that preserves edges can be performed.

In the image processing apparatus, the calculating method determining means indicates the chromaticity value of the calculating pixel closest to the interpolation target pixel among the plurality of calculating pixels selected by the selecting means. The first parameter is equal to or greater than a second threshold set in advance, and the first parameter is equal to or greater than the second threshold set in advance. When there is an operation pixel in which the second parameter indicating the color difference is equal to or greater than a preset third threshold value, a linear interpolation method is selected as an interpolation operation method of the chromaticity value of the interpolation target pixel, and the interpolation means In this case, of the plurality of calculation pixels selected by the selection means, the chromaticity value of the calculation pixel closest to the interpolation target pixel and the first parameter are equal to or greater than a predetermined second threshold value. And Third to second parameter indicating a color difference between the closest basis pixel to the interpolation target pixel is set in advance
It is possible to adopt a configuration in which linear interpolation is performed using the chromaticity values of the calculation pixels that are equal to or larger than the threshold value of, and the chromaticity values of the interpolation target pixels are determined.

In this case, a smooth change can be expressed by obtaining the chromaticity of the pixel to be interpolated by linear interpolation using only the calculation pixels having a large chromaticity and close to each other in color relation.

In the above image processing apparatus, the calculating method determining means indicates a chromaticity value for a calculating pixel closest to the interpolation target pixel among a plurality of calculating pixels selected by the selecting means. The first parameter is equal to or greater than a second threshold set in advance, and the first parameter is equal to or greater than the second threshold set in advance. If there is no calculation pixel for which the second parameter indicating the color difference is equal to or greater than the third threshold value set in advance, the chromaticity value of the calculation pixel closest to the interpolation target pixel is set as the chromaticity value of the interpolation target pixel. A configuration for selecting an interpolation calculation method to be substituted can be adopted.

As described above, when the chromaticity difference between the selected plurality of calculation pixels is large and an edge exists between the calculation pixels, the chromaticity value of the calculation pixel closest to the interpolation target pixel is calculated by the interpolation target pixel. By directly substituting as the chromaticity value of the pixel, interpolation processing that preserves edges can be performed.

In the image processing apparatus, the calculation method determining means indicates a chromaticity value for a calculation pixel closest to the interpolation target pixel among the plurality of calculation pixels selected by the selection means. When the first parameter is less than the second threshold set in advance and there is another calculation pixel in which the first parameter is less than the second threshold set in advance, the chromaticity of the pixel to be interpolated is A linear interpolation method is selected as a value interpolation calculation method. In this case, the interpolation means selects the chromaticity of the calculation pixel closest to the interpolation target pixel among the plurality of calculation pixels selected by the selection means. Performing linear interpolation using the value and the chromaticity value of the calculation pixel in which the first parameter is less than a preset second threshold value;
The chromaticity value of the interpolation target pixel may be determined.

In this case, a smooth change can be expressed by obtaining the chromaticity of the pixel to be interpolated by linear interpolation using only the calculation pixels having a small chromaticity.

In the image processing apparatus, the calculating method determining means indicates a chromaticity value for the calculating pixel closest to the interpolation target pixel among the plurality of calculating pixels selected by the selecting means. If the first parameter is less than the preset second threshold and there is no other calculation pixel where the first parameter is less than the preset second threshold, the pixel closest to the interpolation target pixel is It is possible to adopt a configuration in which an interpolation calculation method for substituting the chromaticity value of the calculation pixel as the chromaticity value of the interpolation target pixel is selected.

As described above, when the chromaticity difference between the selected plurality of calculation pixels is large and there is an edge between the calculation pixels, the chromaticity value of the calculation pixel closest to the interpolation target pixel is determined by the interpolation target pixel. By directly substituting as the chromaticity value of the pixel, interpolation processing that preserves edges can be performed.

In order to solve the above-mentioned problems, the image processing method of the present invention converts multi-valued color image data represented in a first color space into a second color image data different from the first color space. A first step of converting into data composed of luminance information and color information expressed in space, and a plurality of operations for calculating pixel values of the interpolation target pixel from existing pixels around the interpolation target pixel A second step of selecting a pixel; a third step of determining a calculation method when calculating a pixel value of the interpolation target pixel based on luminance information and / or color information of the calculation pixel; And a fourth step of calculating the interpolation target pixel using the calculation method determined in the third step. By repeating the first to fourth steps, An image whose resolution has been converted by interpolation It is configured to obtain data.

Further, the image processing method further includes converting the resolution-converted image data represented in the second color space into
It may be configured to include a fifth step of converting to a third color space different from the second color space.

In the first step, it is preferable to use a uniform color space as the second color space.

Therefore, similar to the above-described image processing apparatus,
Different interpolation calculations can be performed for each relationship between luminance and color, and resolution conversion without color shift, color bleeding, or blurring can be performed. Further, the interpolation processing can be performed by an appropriate interpolation calculation method according to the partial feature of the image.

[Brief description of the drawings]

FIG. 1 illustrates a position embodiment of the present invention, and is a block diagram illustrating a configuration of an image processing apparatus.

FIG. 2 is a flowchart illustrating a main flow of an image processing method in the image processing apparatus.

FIGS. 3A to 3C are explanatory diagrams illustrating an arrangement relationship between an interpolation target pixel and peripheral pixels.

FIG. 4 is an explanatory diagram showing table data used when converting representative color data in an RGB color space into an L ^* a ^* b ^* color space.

FIG. 5 is a diagram illustrating RGB of a certain pixel in input image data.
FIG. 4 is an explanatory diagram showing a relationship between data and representative colors extracted from the RGB data.

FIG. 6 is an explanatory diagram illustrating a method of extracting four peripheral pixels surrounding an interpolation target pixel.

FIG. 7 is a flowchart showing a procedure for selecting an interpolation calculation method for luminance information.

FIG. 8 is an explanatory diagram showing a positional relationship between L ^* of two selected pixels.

FIG. 9 is an explanatory diagram showing an interpolation calculation method using linear interpolation.

FIG. 10 is an explanatory diagram illustrating a method of performing interpolation processing by substituting data of a peripheral pixel closest to the interpolation target pixel into the interpolation target pixel.

FIG. 11 is a flowchart showing a procedure for selecting an interpolation calculation method for color information.

FIG. 12 is an explanatory diagram showing a distance and an angle obtained based on a positional relationship between two selected pixels on an a ^* -b ^* chromaticity diagram.

FIG. 13 shows representative color data of the L ^* a ^* b ^* color space as RGB.
FIG. 4 is an explanatory diagram showing table data used when converting to a color space.

FIG. 14^*a^*b^*Of image data expressed in color space
L for a certain pixel in^*a^*b ^*Data and this L^*
a^*b^*Indicates the relationship between the data and the representative colors extracted
FIG.

FIG. 15 is an explanatory diagram showing a processing example of a NN which is one of the conventional methods.

FIG. 16 is an explanatory diagram showing an example of processing of BL which is one of the conventional methods.

FIG. 17 is an explanatory diagram showing an example of CC processing which is one of the conventional methods.

[Explanation of symbols]

 Reference Signs List 1 image processing device 4 first color conversion processing unit (first color conversion unit) 5 peripheral pixel selection unit (selection unit) 6 luminance relation calculation unit (calculation method determination unit) 7 color relation calculation unit (calculation method determination unit) 8 pixel interpolation unit (interpolation means) 9 second color conversion processing unit (second color conversion means)

Continued on the front page F term (reference) 5B057 CA01 CA08 CA12 CA16 CB01 CB08 CB12 CB16 CC02 CD06 CE18 CH07 CH08 DB02 DB06 DB09 DC25 5C076 AA26 AA36 BA06 BA07 BB04 5C077 LL19 MP08 PP20 PP32 PP33 PP36 PP47 PP68 PQ12 PQ01 H085 LA37 LB02 MA05 MA11 NA03 PA03 PA05

Claims (12)

[Claims] 1. Multi-valued color image data represented in a first color space is converted into data comprising luminance information and color information represented in a second color space different from the first color space. First color conversion means for converting the pixel data into a plurality of calculation pixels used for calculating the pixel value of the interpolation target pixel from existing pixels around the interpolation target pixel; An arithmetic method determining means for determining an arithmetic method for calculating the pixel value of the interpolation target pixel based on the luminance information and / or the color information; and an arithmetic method determined by the arithmetic method determining means. An image processing apparatus comprising: an interpolation unit that performs an interpolation process on an interpolation target pixel. 2. The image processing apparatus according to claim 1, further comprising a second color conversion means for converting the image data represented in the second color space into a third color space different from the second color space. The image processing apparatus according to claim 1, wherein: 3. The apparatus according to claim 2, wherein said first color conversion means converts the multi-valued color image data represented in the first color space into a uniform color space as a second color space. 3. The image processing device according to 1 or 2. 4. The method according to claim 1, wherein the calculating method determining means includes a first threshold value having a preset brightness difference from a calculating pixel closest to the interpolation target pixel among the plurality of calculating pixels selected by the selecting means. If there is a calculation pixel that is less than the pixel value, the linear interpolation method is selected as the interpolation calculation method for the luminance value of the pixel to be interpolated. In this case, the interpolation means selects the plurality of calculation pixels selected by the selection means. Among the brightness values of the calculation pixels closest to the interpolation target pixel and the brightness value of the calculation pixels whose brightness difference between the calculation pixel closest to the interpolation target pixel is less than a first threshold value set in advance 4. The image processing apparatus according to claim 1, wherein linear interpolation is performed by using (1) and (2) to determine a luminance value of the pixel to be interpolated. 5. The computing method determining means, wherein, among the plurality of computing pixels selected by the selecting means, a luminance difference from a computing pixel closest to the interpolation target pixel is less than the first threshold. 5. The image according to claim 4, wherein when there is no operation pixel, an interpolation operation method of substituting the luminance value of the operation pixel closest to the interpolation target pixel as the luminance value of the interpolation target pixel is selected. Processing equipment. 6. The calculation method determining means sets a first parameter indicating a chromaticity value for a calculation pixel closest to an interpolation target pixel among a plurality of calculation pixels selected by the selection means. A second parameter that is equal to or greater than the second threshold that has been set, and that has a first parameter that is equal to or greater than the second threshold that is set in advance and that indicates a color difference between the pixel for calculation and the pixel closest to the interpolation target pixel. If there is a calculation pixel whose value is equal to or larger than a third threshold value set in advance, a linear interpolation method is selected as an interpolation calculation method for the chromaticity value of the interpolation target pixel. In this case, the interpolation means The chromaticity value of the calculation pixel closest to the interpolation target pixel among the plurality of calculation pixels selected by the means, the first parameter is equal to or greater than a second threshold value set in advance, and Calculation pixel closest to pixel Linear interpolation using a chromaticity value of a pixel for calculation in which a second parameter indicating a color difference between the second pixel and the third pixel is equal to or greater than a third threshold value set in advance to determine a chromaticity value of the pixel to be interpolated. The image processing device according to claim 1, wherein: 7. The calculation method determining means sets a first parameter indicating a chromaticity value for a calculation pixel closest to an interpolation target pixel among a plurality of calculation pixels selected by the selection means. A second parameter that is equal to or greater than the second threshold that has been set, and that has a first parameter that is equal to or greater than the second threshold that is set in advance and that indicates a color difference between the pixel for calculation and the pixel closest to the interpolation target pixel. If there is no calculation pixel whose value is equal to or larger than the third threshold value set in advance, an interpolation calculation method of substituting the chromaticity value of the calculation pixel closest to the interpolation target pixel as the chromaticity value of the interpolation target pixel is selected. The image processing apparatus according to claim 6, wherein: 8. The calculation method determining means sets a first parameter indicating a chromaticity value for a calculation pixel closest to an interpolation target pixel among a plurality of calculation pixels selected by the selection means. If there is an operation pixel whose first parameter is less than the second threshold that is less than the second threshold that has been set in advance, the linear interpolation method for the chromaticity value of the pixel to be interpolated is present. In this case, the interpolation means selects the interpolation method, and in this case, among the plurality of calculation pixels selected by the selection means, the chromaticity value of the calculation pixel closest to the interpolation target pixel and the first parameter 8. A linear interpolation is performed using a chromaticity value of a pixel for calculation that is smaller than a second threshold value set in advance to determine a chromaticity value of the pixel to be interpolated. The image processing device according to any one of the above. 9. The calculation method determining means sets a first parameter indicating a chromaticity value for a calculation pixel closest to an interpolation target pixel among a plurality of calculation pixels selected by the selection means in advance. If there is no other calculation pixel whose first parameter is smaller than the second threshold value set in advance, the chromaticity value of the calculation pixel closest to the interpolation target pixel The image processing apparatus according to claim 8, wherein an interpolation calculation method is selected, in which is substituted as a chromaticity value of the interpolation target pixel. 10. Multi-valued color image data represented in a first color space is converted into data comprising luminance information and color information represented in a second color space different from the first color space. A second step of selecting a plurality of calculation pixels used for calculating a pixel value of the interpolation target pixel from existing pixels around the interpolation target pixel; and A third step of determining a calculation method when calculating the pixel value of the pixel to be interpolated based on the luminance information and / or the color information; and a calculation method determined by the third step.
And a fourth step of calculating the pixel to be interpolated, wherein the first to fourth steps are repeated to perform interpolation processing on all the pixels to be interpolated to obtain resolution-converted image data. An image processing method characterized by the following. 11. A method according to claim 11, further comprising the step of: converting the resolution-converted image data represented in the second color space into a third color space different from the second color space. The image processing method according to claim 10, wherein: 12. The image processing method according to claim 10, wherein in the first step, a uniform color space is used as the second color space.