JP2527715B2 - Image processing method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Use] The present invention is intended to favorably reproduce a target image in an output means for forming a color image on a recording medium using a recording material of a plurality of colors. The present invention relates to an image processing method for performing color correction.

[Prior Art] Conventionally, a table such as a ROM is often used for color correction processing such as masking. Taking masking processing as an example, the ROM input of the processing is generally C _y.
The density values of (cyan), M (magenta), and Y _e (yellow) are used. At this time, since three inputs of C _y , M, and Y _e are input to the masking ROM, if each bit of C _y , M, and Y _e is expressed by 8 bits, there are 2 ²⁴ ways per 1 bit of output of the ROM. It will be a huge capacity. Therefore, C _y , M, and Y _e are quantized and the number of bits is reduced to be the input address of the masking ROM.

[Problems to be Solved by the Invention] However, conventionally, as a result of masking which is a color correction based on the output characteristic of the output means, for example, a color such as a skin color may be converted into a “skin color” that is slightly different. is there. That is, there is a problem in that the masking corrects a color that is often noticed by humans such as a skin color to a slightly different color, and deteriorates the quality of a reproduced image.

The present invention has been proposed in order to solve the above-mentioned problems of the prior art, and proposes an image processing method in which color correction processing including hue conversion and color correction produces a high-quality result. To aim.

[Means for Solving the Problems] An image for converting the color image data in order to visually output the color image data to a predetermined output means using a plurality of recording agents of the present invention for realizing the above-mentioned problems. The processing method is such that a digital color image composed of color image data each having an independent luminance component, saturation component and hue component is input, and a color image having a predetermined hue with respect to the input digital color image is input. A conversion process for converting the hue component of the data into a desired hue component is performed, and the digital color image data subjected to the conversion process is a color correction based on the characteristics of the recording material, and the conversion process is performed. Color correction is performed based on the color correction characteristics that do not depend on the hue of the generated digital color image data, and is composed of color components corresponding to the plurality of recording materials. It is characterized in that it is converted into image data.

Embodiments Embodiments according to the present invention will be described in detail below with reference to the accompanying drawings.

FIG. 1 is a block diagram of a color video printer as an embodiment to which the present invention is applied.
A masking circuit 14 for inputting the color difference signals RY and BY.
It has HC conversion ROM 13 in the previous stage. The outline of the whole operation is
The Y, RY, and BY input signals are A / D converted, and the image memory (1
Stored at 0,11,12). Then, the CPU 15 takes out Y, RY, and BY of the pixels to be printed from the image memory (10, 11, 12). The extracted RY and BY are converted into hue H and saturation C by the HC conversion ROM 13. The HC conversion ROM 13 stores data for converting the color difference signals RY and BY into the hue H and the saturation C by using the color difference signals as its address inputs.
Y, H, C pass through the masking circuit 14, and C _y ′ (cyan), M ′
(Magenta), Y _e ′ (yellow), D / A converted, and printed.

The HC conversion ROM 13 converts the signals R-Y 'and BY' into luminance Y and saturation C.
Is to be converted to. When converting to H (hue) and C (saturation) in this case, quantization as shown in FIG. 2 is performed. That is, the quantization is one in which the plane of RY and BY is divided into areas surrounded by circles and straight lines, and the values in the areas are represented by ".multidot." Points. At this time, However, if RY = 0, then H = 0.

The deviation from the reference axis is hue H, and the radius of the circle is saturation C
Becomes In FIG. 2, the output H is 4 bits,
A case where C is 3 bits is shown, and a shaded portion 100 shows an area in which a color close to skin color is distributed.

In the figure, h ₀ is the hue of the flesh color in the memory color, and h ₁ and h ₂ are the range of hues that should be felt as the flesh color in the quantization of FIG. Second
Quantization as shown in the figure is performed for H at equal angles and C
Is only taken as a representative value with equal intervals, so human vision is not taken into consideration in such quantization.

Therefore, the masking circuit 14 having the structure shown in FIG. 4 is used to achieve the reproduction of the skin color most noticeable to humans. The masking circuit 14 is constructed so that the HC decode 20, Y, H ', C'to Y', RY for converting the H, C signals into corrected H ', C'as shown in FIG. Conversion circuit 21 for reverse conversion into ‘, BY’, conversion circuit 22 for conversion into RGB, and C
A conversion circuit 23 for converting into _y , M, Y _e , a masking matrix circuit 24 for further performing a normal masking process, and the like.

First, the HC decode 20 will be explained. The HC conversion ROM 1
When 3 outputs H with h bits and C with k bits, generally, the hue and saturation for the area other than the skin color area 100 are 360 degrees × (h / 2 ⁿ ), C _max × C / (2 ^k −1). Here, C _max is the maximum saturation. In such a region 100, the HC decode 20
Does nothing.

On the other hand, the image signal included in the skin color area 100 is converted into the HC conversion ROM 13
Output to the HC decode 20 as (h bits, k bits), the HC decode 20 is closer to h ₀ if h is between h ₁ and h ₂ , as shown in FIG. Output as h '. Furthermore, if consideration is also given to the saturation of the skin color region 100, h is between h ₁ and h ₂ , and 2 ^k
The correction may be applied only to the output of the HC conversion ROM 13 in the range of ≦ C ₀ . An HC decode 20 that converts the input shown in Fig. 2 into the output shown in Fig. 3
It is easy to configure as a ROM.

In this way, by setting the hue of the image signal that seems to be a skin color to a value closer to the hue of the “preferred skin color”, it becomes possible to collect the generally distributed skin color area into a more “preferred skin color” area. The reproduction of can be improved.

The Y, H, and C signals thus obtained are input to the conversion circuit 21 which receives the three signals and subjected to the following conversion.

(R−Y) ′ = C × CosH (B−Y) ′ = C × SinH Further, by the conversion circuit 22, R = (R−Y) ′ + Y ′ G = − {0.3 (R−Y) ′ + 0. 11 (B-Y) '} + Y' B = (B-Y) '+ Y' is outputted, by further conversion circuit 23, the conversion of _{C y = -logR 'M = -logG} ' Y e = -logB ' Is given. Then, in order to correct the ink asymmetric color component, by the masking matrix circuit 24, Perform Here, a ₁₁ ~a ₃₃ place are constants.

It should be noted that the HC conversion ROM 13 and the HC decode 20 can be configured by one ROM. Although the above embodiment describes the video printer, a more preferable skin color can be reproduced by displaying the output of the RGB conversion circuit 22 on the CRT when outputting to a CRT device or the like.

According to the above embodiment, the conversion process (HC decoding 20)
The color correction result including the color correction processing (masking matrix 24, etc.) can be made high quality.

In the conversion process, as shown in FIGS. 2 and 3, the saturation component is corrected and the hue component is converted to a desired hue. That is, in this embodiment, processing suitable for each component is performed for each component, and high-quality color correction can be performed.

Further, since the conversion process is performed only on the area (see FIG. 2) defined by the saturation and the hue indicating the skin color, high-quality color correction can be performed.

[Effects of the Invention] As described above, according to the image processing method of the present invention, the conversion processing of converting the hue component of the image data having a predetermined hue of the digital color image into the desired hue component is performed. Further, the hue-converted image data is color-corrected by the color-correction characteristic set based on the characteristic of the recording material used in the above-mentioned means.

Therefore, for example, when a certain color may be corrected to a different color by the color correction process, by performing the above conversion process in advance, by converting the certain color into another desired color, The above fear can be prevented.

Further, more preferable high quality color reproduction can be performed based on the characteristics of the image data of the input digital color image. In particular, since the conversion process is performed based on the hue component, the conversion process can be performed with high accuracy.

Furthermore, according to the present invention, since the color correction processing is performed after the conversion processing is performed, this processing can be performed efficiently.

[Brief description of drawings]

FIG. 1 is a block diagram of an image processing apparatus according to an embodiment of the present invention, FIG. 2 is a diagram for explaining quantization and skin color distribution by an HC conversion ROM, and FIG. 3 is a diagram showing the hue of an image signal in a skin color region. FIG. 4 is a diagram for explaining the correction for making the color closer to the memory color, and FIG. 4 is a detailed diagram of the masking circuit of the embodiment. In the figure, 10 to 12 ... Image memory, 13 ... HC conversion ROM, 14 ... Masking circuit, 16 ... Head driver.

Claims (4)

(57) [Claims] 1. An image processing method for converting color image data in order to visually output the color image data to a predetermined output means using a plurality of recording materials, the method comprising an independent luminance component, saturation component, and A conversion for inputting a digital color image composed of color image data having a hue component, and converting the input digital color image into the desired hue component of the hue component of the color image data having a predetermined hue. A color correction characteristic of the digital color image data which has been subjected to the conversion processing and which has been subjected to the conversion processing, which is color correction based on the characteristics of the recording material, and which does not depend on the hue of the converted digital color image data. An image processing method characterized by performing color correction based on the above, and converting into image data composed of color components corresponding to the plurality of recording materials. . 2. The image processing method according to claim 1, wherein the conversion process stores a saturation component of image data of the input digital color image. 3. The conversion processing is performed on the input color image,
The image processing method according to claim 1, wherein the image processing is performed on image data existing in a color region defined by a predetermined hue component and saturation component. 4. The image processing method according to claim 1, wherein the conversion process is a color masking process of the recording agent.