JP3337697B2 - Image processing method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing method for converting a color space of color image data according to a color reproduction range of a color image output device.

[0002]

2. Description of the Related Art Conventionally, there is an apparatus which inputs color image data (multi-valued data) from a color scanner and performs color recording by a color printer based on the color image data. In this case, the color characteristics (reading color characteristics) of the color scanner and the color reproduction characteristics (color reproduction range) of the printer may not match. Even in such a case,
For example, an image conforming to a color space range that can be handled by the printer side by itself without deciding which range in the color space the possible value (0 to 255) of the multi-valued image data (8 bits) from the scanner indicates. It was interpreted as data and recorded based on the multi-valued image data.

[0003]

Therefore, conventionally, when performing hard copy or soft copy of an original image, the original original or the color of the original image data is not faithfully reproduced on the printer side, and the original original is not reproduced. There was a problem that the image was reproduced in a different color from the image.

The present invention has been made in view of the above conventional example, and provides an image processing method for satisfactorily reproducing input color data including colors outside the color reproduction range of a color image output device by the color image output device. The purpose is to do.

[0005]

In order to achieve the above object, an image processing method according to the present invention comprises the following steps.
That is, an image processing method for converting input color data according to a color reproduction range of a color image output device, wherein a non-linear line that changes a hue on a uniform perceived color space according to saturation is converted to the uniform perceived color. For each of a plurality of different hues on the space, prepare the input color data within the color reproduction range of the color image output device using the nonlinear line corresponding to the hue on the uniform perceived color space of the input color data. It is characterized by being converted into.

[0006]

[0007]

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below in detail with reference to the accompanying drawings. <Description of Color Space Conversion Example (FIGS. 2 and 3)> FIG. 2 is a diagram showing the range of the color space.

In FIG. 2, a closed curve 201 including a point F is shown.
Indicates the outer edge of the full color space, and indicates the closed curve 2 including the point E.
02 indicates the outer edge of the color space within a range in which the printer can reproduce colors. In the present embodiment, the closed curve 201 including the point F
In other words, the color reproducibility of the printer is improved by spatially compressing the entire color space (closed curve 201) into the color space of the closed curve 202 including the point E.

In this embodiment, FIG. 2A shows L, a,
Although the uniform perceived color space such as b, Y, I, and Q is shown, another color space may be used. FIG. 2B shows how the color space is compressed in the cross section when the color space in FIG. 2A is cut along the cutting curve 203. In FIG. 2B, F indicates the outer edge of the entire color space, E indicates the outer edge of a range in which color reproduction by a printer is possible, and D indicates
Indicates a reproduction range in which color shift does not occur by the printer, and C indicates the center of the color space.

The color information contained in the color space surrounded by the closed curve 204 passing through the point D in the entire color space indicated by the closed curve 201 (corresponding to F in FIG. 2B) is reproduced by the printer as it is. . When a color in the color space surrounded by the closed curves 204 and 201 is reproduced by the printer, the colors are compressed in the color space surrounded by the closed curves 202 and reproduced by the printer. This is shown in FIG. 2B, in which the line segment DF on the cutting line is compressed into a line segment DE, and each point on the line segment DF is compressed on the line segment DE at a uniform rate. The point C shown in FIG. 2 indicates the center position of the uniform color space according to the present embodiment, and its color is white.

FIG. 3 is a diagram showing an example of color space conversion. Here, each color component of a given full color space is represented by 8 bits (25 bits).
The case of 5) is indicated by S1. The range of possible values of each color component when this is spatially compressed as shown in FIG. 2B under the same conditions is indicated by S2. Further, S3 denotes a case utilizing a range capable of color reproduction of the printer to 8 bit full, to a final D _i terms of total color space to the possible color reproduction of the printer of D _f, the total color space Point F
_i only needs to be converted to E _f .

Further, as shown in FIG.
The colors on the line cut by 03 indicate colors of the same hue.

However, in FIG.
2 is actually a curve as shown in FIG. 2 (A), and therefore, the phase of each point on the cutting line 203 with respect to the horizontal line 205 drawn horizontally to the right from the point C is shown. The angle (this is called the hue of the uniform color space ) is slightly different. Therefore, the angle of the phase shift of the color of the line segment CF at point SS theta _A, if theta _B at point SE, if the color of the position of the point SS in the color space compression is the point of the point SE without changing the phase If compressed, the phase angle at this time is in principle (θ
_It can be seen that _{B- [} theta] _A ) should be increased.

The distance between the straight lines C and SS is r _SS , the distance between the straight lines C and SE is r _SE, and the length of the curved lines C and SS is R
_{If SS} and the length of C and SE of the curve are R _SE , then R _SS and R
_There is no practical problem even if the ratio of _SE is approximated to the ratio of r _SS to r _SE . Therefore, in this embodiment, the coordinates in the color space are determined using the linear distances r _SS and r _SE from the point C to the point SS and the point SE.

FIG. 11 is a diagram showing a data structure of a hue correction table (hue correction table 1).
Data corresponding to the above-described phase angle theta _A and theta _B and the like are written. That is, the angle θ from the horizontal line 205 in FIG.
defined by a linear distance r _n from _n and center point C cutting line 2
Position of the point P on 03 (r _n, θ _n) of, when the difference in phase angle from a straight line CF was _{θ (r n, θ n)} , and the distance r _n for each point, the phase corresponding thereto Data (angle) indicating the difference from the angle θ is written as a table. For example, θ (r ₁ , 0) is an angle 0 ° from the horizontal line 205.
And the position (r ₁ , 0) of the point P on the cutting line defined by the linear distance r ₁ from the center point C, and the angle corresponding to this point is 0
The phase angle difference when moving to a point on the straight line of ° is shown.

Although not particularly shown, a table similar to that shown in FIG. 11 (referred to as a hue correction table 2) is provided. An angle corresponding to the angle θ _B when the SE is returned to the same phase line is written, and the data is θ ′ (r
_n , θ _n ).

Therefore, the procedure of the color space compression is shown in FIG.
From the hue correction table 1 shown in FIG. _A Find and disconnect
Position of point SS 'on straight line CF from point SS on curve 203
Virtual. Next, the same straight line from the point SS 'on the straight line CF
The upper point SE 'is derived by distance compression. Next, point SE '
Hue correction table to determine the point SE of the
Than bull 2_B To find the phase angle of the point SE.
As a result, a point obtained by spatially compressing the point SS on the curve segment 203
SE can be determined.

Further, in order to reduce the data capacity of the hue correction tables 1 and 2, it is easy to generate data interpolated at the time of processing (at the time of retrieval) with data having a coarse step width. I can guess. Further, instead of storing these phase angle differences as data, a function for obtaining the phase angle differences may be prepared. For example, a function indicating the relationship between the distance r and each phase θ as shown in FIG. 12A may be provided for each phase angle.
As shown in (B), a single function indicating the relationship between the distance r and the phase angle θ may be provided for all the phase angles, and these values may be obtained by calculation. FIG. 12 (B)
Is the correction angle (θ) on the vertical axis and r / Fθ on the horizontal axis. Here, since Fθ indicates the outermost edge of the color space, the maximum value is always “1” without depending on the phase angle.

In this embodiment, an example using the above-described color correction tables 1 and 2 will be described. <Description of Image Conversion Circuit (FIG. 1)> FIG. 1 is a block diagram showing a configuration of an image conversion circuit of the present embodiment for realizing the above-described color space compression.

Reference numeral 101 denotes a color conversion unit which sequentially inputs R,
The G and B image data are converted into Y, I and Q color spaces. Here, it is possible to convert the R, G, and B image data into a Lab space and treat Y and L, I and a, and Q and b in the same manner, but the description is omitted here. Color converter 10
The image data converted into the Y, I, and Q color spaces in 1 is input to the angle calculation unit 102. The angle calculation unit 102 obtains an angle θ indicating how many times the pixel data P of the image data in the I and Q orthogonal coordinate systems is rotated from the Q axis as shown in FIG. This means that, if the coordinates of the pixel data P (I, Q) are (i, q), tan ^-1 i / q (q ≠ 0) or sin ^-1 i / (i ² + q ² ) ^1/2 cos ^The angle θ can be obtained by using any one of ⁻¹ q / (i ² + q ² ) ^1/2 [(i ² + q ² ) ^1/2 ≠ 0].

The output of the color conversion unit 101 is output from the r calculation unit 1
03 is input to the r calculation unit 103.
= (I ² + q ² ) ^1/2 and the distance from the center position C of the color space is calculated. Thereby, the coordinates (r, θ) of the image data P in the polar coordinate system are obtained.
In the orthogonal coordinate system, (i, q) = (rsin θ, rcos
θ).

The output θ of the angle calculation unit 102 is input to the color correction table 1 (121), and the phase angle difference from a point on the cutting curve to the straight line CF is obtained. This difference and the angle θ obtained by the angle calculation unit 102 are added by the adder 122, and the position of a point on the cutting line 203 is converted to a position on the straight line CF. This converted phase difference angle can be obtained from the D table 1
04, input to the accumulation coefficient a table 105 and the day accumulation coefficient b table 106. The output of the adder 122 is input to the hue correction table 2 (123).
From the correction table 2 (123), an angle for converting the coordinates of a point on the straight line into a point on the cutting curve is obtained.
The point on the straight line is returned to the position of the point on the curve by the adder 124. Further, since r3, which is the output of the selector 112 described later, is input to the correction table 2 (123), it is possible to convert a point whose distance has been compressed on a straight line into a curve.

Note that the values F _i , E of the entire color space in FIG.
_i and D _i differ depending on the angle θ, and the color reproduction ranges E _f and D _{f of the} printer also differ depending on the angle θ. FIG. 5 shows the colors in the actual full color space range for each angle θ.
D _i , E _i , and F _i . Here, D with respect to the angle θ
_i , E _i , and F _i are represented by D _i (θ), E _i (θ), F
_i (θ).

In FIG. 5, if the largest one of F (0) to F (359) is 8 bits, it becomes "255". Here, the minimum unit of the angle θ is 1 °, but is not limited thereto, and may be smaller or larger. Further, the difference between the angles θ may not be uniform.

In this embodiment, quantized data within the color reproduction range of the printer is obtained for each angle.
D _f and E _f for each angle θ are determined by measuring the color reproduction characteristics of the printer. Thus, D _f / D _i for the angle θ is written in the b table 106, and (E _f −D _f ) / (F _i −) for the angle θ in the a table 105.
D _i ) has been written.

The output r of the r calculation unit 103 is input to the comparator 108, and the subtraction unit 107 outputs r2 = r−
D _i (θ) is calculated. Here, D _i (θ) is data output from the D table 104 and indicates D _i with respect to the angle θ. This calculation result, r2 (= r−D _i (θ))
) Is input to the multiplier 109 and the product coefficient a table 10
5 multiplied by the output. Then, the output of the multiplier 109 is added to the output of the multiplier 117 by the adder 111.

In the multiplier 117, the output of the D table 104 is output.
Force D_i(θ) and D from product coefficient b table 106_f(θ) /
D_i(θ) is input, resulting in D_f(θ) is output
It is. Thus, the output of the adder 111 is the pixel data P
P (r ′, θ) after color space compression of (r, θ). However
And r of the pixel data P (r, θ) is F_i(θ) ≧ r ≧ D _i 
Is the case.

On the other hand, the output of the product coefficient b table 106 is given to the multiplier 110, where the image data P (r,
θ) is multiplied by r. As a result, r ″ of the image data P (r ″, θ) after color space compression when D _i (θ) ≧ r is obtained. These r 'and r "are input to the selector 112, one of them is selected and output as r3.

The output r of the r calculator 103 is input to a comparator 108, where it is compared whether or not D _i (θ) ≧ r, and the comparison result is given as a selection signal of a selector 112. Thus, when D _i (θ) ≧ r, r3 = r ″ is selected, and otherwise, r3 = r ′.

The color space compressed data r3 thus obtained, the hue correction table 2 (123) and the adder 1
Since the final phase after spatial compression is determined by the phase obtained in step 24, the combining unit 113 forms one data as P (r3, θ). The color calculation unit 114 calculates P (r3,
θ), I = r3 · sin θ, Q = r3 · cos θ, and further receives Y data from the color conversion unit 101. The Y, I, and Q data prepared in this manner is output to the inverse color conversion unit 1.
15 converts the Y, I, and Q signals into R, G, and B signals.

Here, the conversion between R, G, B, Y, I, and Q in the color conversion unit 101 and the inverse color conversion unit 115 is a linear first-order conversion.

[0033]

(Equation 1) It can be expressed by the above matrix operation. This operation can be realized by a product-sum operation using a multiplier and an adder.
Further, a method of substituting the sum of products of the part for the ROM is generally known.

Also, as shown by the broken line in FIG.
, The I and Q space widths change, so that the D table 104 and the product coefficient a table 1
05, it is more desirable to switch a plurality of tables of the product coefficient b table 106 to change the output value. <Another Embodiment (FIG. 6)> FIG. 6 shows a color space compression in another embodiment of the present invention, in which a closed curve 202 (FIG. 2) including a limit point E of a color gamut of a printer is reproduced as it is. The other color spaces are converted on the closed curve 202 including the limit point E.

FIG. 10 corresponds to FIG. 3 of the first embodiment.
E_iIs represented by the distance from the center point C of the color space.
Then, this is set to 8 bits. And this distance is pudding
Center point C to E in the reproduction range_f Should be converted to
Will be. Thus, E _i ~ E_f The points on the line segment are all
E_f Becomes

FIG. 8 is a block diagram showing an image processing circuit according to a second embodiment for realizing this color space compression. Portions common to those in the first embodiment are designated by the same reference numerals, and description thereof will be omitted. .

In the second embodiment, an E table 402 is used instead of the D table 104. This E table 4
02 is a table storing the data shown in FIG. In the comparator 108, r of P (r, θ) is E
_{If i} (θ) <r, selector 112 selects E table 4
02, the output r5 = E _i (θ) is selected, and the combining unit 113
Is input to

The angle θ of Fθ is obtained from the output θ of the angle calculation unit 102 by the hue correction table 1 (121) and the adder 122, and the angle is given to the product coefficient C table 602, and is multiplied by Y and θ. The coefficients are output. This product coefficient is output as a value of E _f (θ) / E _i (θ) with respect to θ, and these are stored as a table for each angle θ. Further, another table may be provided according to the luminance signal Y.

Therefore, in the multiplier 110, the pixel data P
An operation of r6 = r × E _f (θ) / E _i (θ) is performed on (r, θ) to obtain data after color space compression. Therefore, when E _i (θ) ≧ r, the result of the comparator 108 controls the selector 112 to output the output r6 of the multiplier 110 to r
Output as 3 = r6.

FIG. 7 shows color space compression according to the third embodiment of the present invention, in which compression is performed uniformly within a closed curve including the outermost edge F of the entire color space.

FIG. 9 is a block diagram showing the configuration of an image processing circuit for realizing this color space compression.

The output from the r calculator 103 is output to the multiplier 12
The result is multiplied by 1 and given to the combining unit 113 as data after color space compression. Therefore, the output of the multiplier 121 is the conversion ratio when converting the total color space to 8 bits (0 to 255), which is the color reproduction space by the printer, when the entire color space is 8 bits (0 to 255). This conversion ratio is supplied from the product coefficient d table 603. The contents of the product coefficient d table 603 are obtained by using E _f (θ) and F _i (θ) shown in FIG.
E _f (θ) / F _i (θ). Also, this product coefficient d
The table 603 may be provided for each luminance signal Y, and one table content is read and determined based on the Y information and the θ information. The maximum value of E _f (θ) and F _i (θ) is “255”, and takes a value of “255” or less according to the angle information θ and the luminance information Y.

By the way, in the embodiment of the present invention and other embodiments, all the portions to be compressed in the color space are uniformly compressed using a multiplier in proportion to the distance from the center position of the color space. The present invention is not limited to this. For example, the color difference in the color space compression may be changed depending on the distance r from the center position. Further, the degree of the compression may be varied depending on the angle θ. Furthermore, since the entire color space of the luminance component Y differs depending on Y, the degree of compression may be different.

Further, in FIG.
May be made constant in the ratio between the line segment CE and the line segment CD. The closed curve including the line segment CD is always constant, and the closed curve including D is a perfect circle centered on the point C. May be set.

As described above, according to the present embodiment, most of the color reproduction range of a printer or the like is faithfully reproduced, and a part of the color reproduction range is processed as a compression space of the entire color space. It is possible to record / reproduce while maintaining color continuity with almost satisfying color reproduction and maintaining color continuity.

Further, according to this embodiment, the entire color reproduction range of the printer is faithfully reproduced, other colors are approximated to the outer edge of the color reproduction range of the printer, or the entire color space is reproduced in the printer color reproduction range. Or can be compressed to a range. In addition, since color space compression can be performed on the same hue line, a slight shift in hue does not occur due to the color space compression.

In this embodiment, r indicates the saturation direction, and θ indicates the hue. Therefore, in this embodiment, r
And the θ component is not converted, and only the saturation can be converted without changing the hue by the color space compression.

[0048]

As described above, according to the present invention, the input color data is converted into the color reproduction range of the color image output device by using the non-linear lines prepared for each of a plurality of different hues in the uniform perceived color space. Therefore, even if color data including a color outside the color reproduction range of the color image output device is input, the color image output device can output well.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an image processing circuit according to a first embodiment of the present invention.

FIG. 2A is a diagram showing a range of a color space.

FIG. 2B is a diagram showing an example of color space conversion.

FIG. 2C is a diagram showing a range of a color space.

FIG. 3 is a diagram showing a conversion example of a color space.

FIG. 4 is a diagram showing a range of a color space.

FIG. 5 is a diagram for explaining data contents of various tables corresponding to each angle.

FIG. 6 is a diagram showing a conversion example of a color space.

FIG. 7 is a diagram showing an example of color space conversion.

FIG. 8 is a block diagram illustrating a configuration of an image processing circuit according to a second embodiment.

FIG. 9 is a block diagram illustrating a configuration of an image processing circuit according to a third embodiment.

FIG. 10 is a diagram showing an example of color space conversion.

FIG. 11 is a diagram illustrating a configuration example of a hue correction table.

FIG. 12 is a diagram showing hue correction data as a function.

[Explanation of symbols]

 Reference Signs List 101 color conversion unit 102 angle calculation unit 103 r calculation unit 104 D table 105 product coefficient a table 106 product coefficient b table 107 subtraction unit 108 comparator 109, 110 multiplier 111, 122, 123 adder 112 selector 113 synthesis unit 114 color Calculation unit 115 Inverse color conversion unit 117 Multiplier 121 Color correction table 1 123 Color correction table 2

Claims (1)

(57) [Claims] 1. An image processing method for converting input color data according to a color reproduction range of a color image output device, wherein a hue on a uniform perceived color space is changed according to saturation.
A linear line is represented by a plurality of different colors on the uniform perceived color space.
Prepare for each of the phases , corresponding to the hue in the uniform perceived color space of the input color data
The input color data using the nonlinear line
In the color reproduction range of the color image output device .