JP2003235059A - Color calibration system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a color calibration system capable of minimizing a visual difference among ordinary users with respect to colors of an image displayed among computers at a remote place or in a network. <P>SOLUTION: Luminance characteristics of an image provider and a display side are measured, gradation data are corrected on the basis of the measured value and the corrected data are transmitted to a display apparatus. In order to correct the data by taking into account the color constancy of the visual sense in people, an image of an object under a virtual lighting is produced on the basis of a spectral reflectance of the object and the spectral radiance of the virtual illumination light. Further, a color pattern including the object and a plurality of reflection objects whose spectral reflectance is clear is photographed under the same photographing illumination, the spectral radiance of the illumination is obtained by utilizing the principal component analysis to derive the spectral reflectance of the object. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for minimizing visual color difference between a provider and a user and between users in displaying images between computers.

[0002]

2. Description of the Related Art In commerce on the Internet, fidelity of color expression on a computer screen has become a problem. For example, in mail order using the WWW, there is a considerable difference in the appearance of colors between the product viewed on the client screen and the actual product.

In the current technology, there is a method such as spectrophotometry using a measuring instrument to perform accurate adjustment. This method allows absolute color adjustment.

[0004]

However, the adjustment of the color development of the display device requires an expensive device and
It requires technically advanced and complicated operations.

Further, there is a problem in that the relative color development between the image forming side and the image displaying side is not adjusted.

Further, since the visual characteristics of individual users are not taken into consideration, there is a problem that the appearance of colors differs among users even if the physical characteristics of colored light are the same.

Further, since the color constancy is not taken into consideration, there is a problem that the appearance of colors varies depending on the lighting environment of the user.

According to the first aspect of the present invention, the visual color difference of the user can be simply measured by a measuring device or an experienced operator without requiring an expensive device on the image forming side and the image displaying side. It is possible to minimize not only the physical color difference that can be corrected but also the color difference that depends on the difference in visual characteristics between individuals.

According to the second aspect, an image is generated in consideration of the illumination on the display side without requiring an expensive device on the display side.
By generating an image that considers the illumination on the display side, it becomes an image that considers the color constancy that is inherent in the human visual sense, and the user has a much higher perception level at that perception level than when the illumination condition is not considered. It is possible to accurately perceive colors.

The principle of "color constancy" is that when the target object and the observer are under the same illumination, even if the characteristics of the illumination light change (for example, from sunlight to an incandescent bulb), It is that the visual system adapts to the illuminating light and that the human recognizes the illuminating environment, and perceives the color of the changed object as the physical color light to be the same color.

According to the third aspect, it is possible to derive the spectral reflectance of the object from the image which is the photographed result in a situation where the spectral reflectance of the object cannot be measured at the site where the image is photographed.

[0012]

According to the color calibration system of claim 1, the signal value of the color is defined by the brightness value based on the brightness characteristic value of the basic primary color defined by the side providing the image. It is characterized in that it is converted into each color luminance data of the color space, and then the luminance data is converted into a signal value of the color based on the luminance characteristic value of the basic primary color defined on the image display side and displayed.

The derivation of the luminance characteristic value of the basic primary colors is performed by selecting two colors each so as to cover all the defined basic primary colors.
It is characterized in that the measurement results of the signal values are compared so that the colors are compared with each other so that the brightness matches with human eyes.

There are already some methods for measuring the luminance ratio of two colors, and the HFP method or motion cancellation method is known. When these methods are used, since the measurement method relies on the human eyes, the difference in visual characteristics between users is automatically included in the measurement value.

The "defined basic primary color" is a combination of three or more independent colors that can be used in the display system, and in principle, any three or more colors can be selected. When the image providing side and the display side have different primary primaries, it is necessary to convert each color luminance data of the color space defined by the luminance value into the luminance data of the display-side basic primary color. Since this conversion can be performed in any color space, the following description will be made assuming that the image providing side and the display side use the same basic primary color.

The basic primary colors are n colors, which are C ₁ , C ₂ ,.
If C _n , the color is represented by the combination of the luminance intensities of these basic primary colors. Assuming that the signal intensity of the basic primary color C _i is ki and the luminance corresponding to it is Li (k _i ), the color is (L ₁ (k ₁ ), L ₂ ,
(k ₂ ), ..., L _n (k _n )).

When the maximum luminance of the basic primary color C _i is L _{i, max} , it can be expressed as in Expression 1. Here, r _i (k _i ) is a ratio of each primary color to the maximum brightness, and takes a value between 0 and 1. Formula 1 L _i (k _i ) ＝ r _i (k _i ) × L _{i, max}

The ratio ri (k _i ) of each primary color to the maximum brightness
Is derived using the existing method of measuring the brightness for each gradation.

Then, using a lookup table or the like,
The signal values k _i can be derived from r _i (k _i ) or, conversely, r _i (k _i ) from k _i .

Both signals are adjusted so that the luminances of the two primary colors C _a and C _b are the same, and the signals of C _a and C _b are K _{a, b} and K _, respectively.
_It is assumed that the brightness matches at _b and a. R _{a, b} = r _a (K _{a, b} ) / r
_When defined as _b (K _{b, a} ), it becomes as shown in Expression 2. Formula 2 R _{a, b} ＝ r _a (K _{a, b} ) / r _b (K _{b, a} ) ＝ L _{b, max} / L _{a, max From the} formula 2, suppose that the luminances of the two primary colors C _a and C _b are Even if the measurements are not consistent, if the results of C _a and C _c and C _b and C _c are available, then R _{a, b} = R _{a, c} / R _{b, c} You can ask. For this reason, the measurement in which the luminances of the two primary colors are matched needs to cover all the primary colors.

From Expression 2, an arbitrary primary color C on the image providing side_iMaximum of
Brightness Lv_{i, max}Is a certain primary color C _oThe highest brightness Lv_{o, max}Next using
It is expressed by the formula. Lv_{i, max}= 1 / Rv_{i, o}× Lv_{o, max} Substituting this equation into equation 1 yields equation 3. Formula 3 Lv_i(kv_i) ＝ rv_i(kv_i) / Rv_{i, o}× Lv_{o, max}

Similarly, on the image display side, equation 4 is established using the maximum luminance of a certain primary color Cp. Formula 4 Lu _i (ku _i ) ＝ ru _i (ku _i ) / Ru _{i, p} × Lu _{p, max}

When the luminances of the corresponding primary colors on the image providing side and the display side of the pixels forming the image are compared, if the ratio is constant regardless of the type of pixel or primary color, it is the same color. If this is expressed by a formula, formula 5 is obtained. Formula 5 Lu _i (ku _i ) = q × Lv _i (kv _i ) However, q is a value that does not depend on pixels or colors.

Substituting equations 3 and 4 into equation 5 and transforming it yields equation 6
become. Formula 6 ru _i (ku _i ) = rv _i (kv _i ) × Ru _{i, p} / Rv _{i, o} × Q, where Q = Lv _{o, max} / Lu _{p, max} × q.

The terms other than Q in equation 6 can be calculated before the image is displayed.
Because eventually any primary color C _iThe right-hand side of Equation 5 for is
The value Q is expressed as an unknown. Primary color C_iSignal maximum of
K_{i, max}Then, the maximum of the signal on the image providing side and the display side
Ratio to value kv_i/ Kv_{i, max}And ku_i/ Ku_{i, max}Becomes closer
To choose Q. ru_i(ku_i), Look up
Using a cable etc._iCan be asked. This signal k
u_iIf the image is sent to the device on the display side,
The same color can be displayed.

In the color correction system according to the second aspect, the spectral reflectance of the object is measured on the image providing side, and from the result and the spectral radiance of the light of the virtual illumination, under the virtual illumination. It is characterized by generating an image of the object.

First, a distribution map of the spectral reflectance of the object is created. Since the spectral reflectance differs depending on the material and color, an image of the object is taken and the image is decomposed for each material and color to create a spectral reflectance distribution map. Then, the spectral reflectance of each part is measured based on the distribution chart of the spectral reflectance of the object.

Next, the spectral radiance of the virtual illumination light is designated. Specifically, there are methods such as based on the model number of the luminaire, based on the spectral radiance table of the luminaire, or actually measuring the target illumination if it can be measured.

Chromaticity and luminance data of the object is generated using the spectral radiance of the designated illumination light and the measured spectral reflectance of the object. By calculating the gradation value based on these data, it is possible to generate an image of the object under virtual illumination.

In the spectral reflectance deriving system according to a third aspect of the present invention, a color pattern including an object and a plurality of reflective objects whose spectral reflectances are apparent is photographed by the same photographing illumination, and the principal component analysis is used to photograph the color pattern. It is characterized in that the spectral radiance of the illumination is obtained and then the spectral reflectance of the object is derived.

A color pattern including a plurality of reflecting objects having clear spectral reflectances is photographed in the same environment as the illumination environment of the object, and the image is used as environmental information.

The spectral radiance of the illumination light in the photographing environment is estimated from the color and luminance information of the color pattern image and the spectral reflectance of the reflecting object included in the color pattern. As an estimation method, a database of the spectral radiance of illumination is created in advance, and three or more main principal components are obtained in advance by principal component analysis, which is a statistical method, so that each error is minimized. It is determined by approximating the spectral radiance of the illumination light by adjusting the principal component weighting coefficient.

By measuring an image obtained by photographing a general object with a photometer, information about color and brightness of the image such as tristimulus values (X, Y, Z) is obtained. The spectral reflectance of each object is estimated from these values and the spectral radiance of the illumination light estimated above. As an estimation method, a database of spectral reflectances of the object and its analogues is created in advance, and three or more main principal components are obtained in advance by a principal component analysis, which is a statistical method, and the error is minimized. The spectral reflectance of the object is approximated by adjusting each principal component weighting coefficient so that

Examples of the invention 【The invention's effect】

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Toshio Nishimura             2-37 Chitosedai, Setagaya-ku, Tokyo F-term (reference) 2G020 AA08 DA05 DA12 DA34 DA65                 5B057 CA01 CA08 CA12 CA16 CB08                       CB12 CB16 CE11 CE16                 5C061 BB11 CC01                 5C066 AA01 AA03 CA17 EB01 FA02                       KA12 KD07 KE03 KE04 KE09

Claims (3)

[Claims] 1. A color calibration system for displaying an image by performing brightness correction based on the brightness characteristic value on the image creating side and the brightness characteristic value on the display side. 2. A color correction system for an image under virtual illumination based on a spectral reflectance of an object and a spectral radiance of virtual illumination. 3. A system for deriving a spectral reflectance of an object from an image.