TWI459220B - System and method of predicting color imagery - Google Patents

Description

Color image prediction system and method

The present invention relates to a color image prediction system and method, and more particularly to a color image prediction system and method capable of converting a color space system and a color psychology into each other.

One of the most famous color image spaces (Color Image Space) was proposed by the Nippon Color and Design Research Institute (NCD) in 1977 and has been widely accepted by the public for nearly two decades. And adopted by academic circles in the industry. With the establishment of the color image space, artists, designers and other related fields can understand the tendency of human emotions in the sense of color, and thus make it prudent to apply color. However, the color image scale value of this color image space is converted by the psychological experiment method of the five-level judgment. The scale values are mainly in the range of 0 to 3, so it is difficult for users to practice. The performance of psychological intensity is consistent, so that there is some degree of inadequacy and inconvenience in the accuracy of color image quantification and in achieving automatic prediction.

In summary, how to convert the color space system and the color psychological feeling into each other, which is consistent with the user's strong feeling of psychological feeling is the goal that is urgently needed.

The invention provides a color image prediction system and method, which converts a color space system and a color mental feeling into each other by a HRB color image space established by a relative semantic phrase, and can predict a color image of a user or provide a candidate color.

The color image prediction system according to an embodiment of the invention comprises an input module, a prediction module and an output module. The input module is used for a user to input a color information or a color image information. The prediction module is configured to map the color information to an HRB color image space to obtain a predicted image value, or to correspond the color image information to the HRB color image space to obtain a candidate color, wherein the HRB color image space includes an H axis and a R-axis and a B-axis, the H-axis represents hard-soft, deep-shallow, thick-thin, loose-tight, bright-dark, strong-weak, and the R-axis represents relaxed-tight, sturdy- Plain, natural - artificial, oppressive - free, gentle - ferocious, dirty - clean, B axis represents beauty - ugly, mournful - excited, luxurious - simple, happy - sad, sick - healthy, clear - fuzzy. The output module is electrically connected to the prediction module for outputting predicted image values or candidate colors.

A color image prediction method according to another embodiment of the present invention includes: an input step, a prediction step, and an output step. The input step is for inputting a color information or a color image information. The predicting step is configured to correspond the color information to an HRB color image space to obtain a predicted image value, or to correspond the color image information to the HRB color image space to obtain a candidate color, wherein the HRB color image space includes an H axis and an R The axis and a B axis, the H axis represents hard-soft, deep-shallow, thick-thin, loose-tight, bright-dark, strong-weak, and the R-axis represents relaxed-tight, awkward-- , nature - artificial, oppressive - free, gentle - ferocious, dirty - clean, B axis represents beauty - ugly, mournful - excitement, luxury - simplicity, happiness - sadness, sickness - health, clarity - blur, and H axis The R-axis and B-axis have a coordinate value of -100 to 100. The output module is electrically connected to the prediction module for outputting predicted image values or candidate colors.

A color image prediction method according to another embodiment of the present invention includes: an input step, a prediction step, and an output step. The input step is for inputting a color information or a color image information. The predicting step is configured to correspond the color information to an HRB color image space to obtain a predicted image value, or to correspond the color image information to the HRB color image space to obtain a candidate color, wherein the HRB color image space includes an H axis and an R The axis and a B axis, the H axis represents hard-soft, deep-shallow, thick-thin, loose-tight, bright-dark, strong-weak, and the R-axis represents relaxed-tight, awkward-- , nature - artificial, oppressive - free, gentle - ferocious, dirty - clean, B axis represents beauty - ugly, mournful - excitement, luxury - simplicity, happiness - sadness, sickness - health, clarity - blur, and H axis The R-axis and B-axis have a coordinate value of -100 to 100. The output step is for outputting a predicted image value or a candidate color.

The purpose, technical contents, features, and effects achieved by the present invention will become more apparent from the detailed description of the appended claims.

Referring to FIG. 1 , a color image prediction system according to an embodiment of the present invention includes an input module 11 , a prediction module 12 , and an output module 13 . The input module 11 is configured to allow a user to input a color information C or a color image information I. If the user inputs the color information C, the prediction module 12 associates the color information C with an HRB color image space (as shown in FIG. 2) to obtain a predicted image value PI. Conversely, if the user inputs the color image information I, the prediction module 12 corresponds the color image information I to the HRB color image space to obtain a candidate color. Color CC. The output module 13 is electrically connected to the prediction module 12 for outputting the predicted image value PI or the candidate color CC generated by the prediction module 12.

The HRB color image space is a 207 textile color sample and 169 sets of relative semantic phrases, which are derived from the color image judgment experiment to obtain a new color image space. 207 textile color samples meet the following color sample selection conditions: 1) the color sample is evenly distributed in the CIE L*a*b* color space recommended by the International Commission on Illumination (CIE); and 2) the color sample should be included by the large Partially saturated, light and deep finishable colors are covered in this broad color gamut. The relative semantics of the 169 groups in the opposite sense are based on the relative adjectives in the latest dictionary published by Northwest Press, 1967, and refer to the relevant research results of imagery and semantics at home and abroad to select adjectives to ensure the formation of adjectives. The scale is suitable for the objectivity of color. The experimental method is to use the Psychological Semantic Differential Method and the Modified Magnitude Estimation Method of the Color Psychophysical Estimation Method. Finally, Factor Analysis is used to extract the common factors from many semantic scales to establish the HRB color image space.

The structure of the HRB color image space consists of three color image space axes, which are referred to herein as: Colour Weight, Color Impact, and Colour Potential. . The Colour Weight contains the following phrases: Hard-Soft, Deep-Shallow, Thick-Thin, Loose-Close (Loose) -Tight), Bright-Dark, Strong-Weak. Here, the "hard-soft" relative semantic phrase is used as the representative of this color image axis, and is called the H axis.

Color effects include phrases: Relaxed-Tense, Animalistic-Vegetal, Unmannered-Affected, Restrained-Free, Mild-Perpetual (Mild-Wild), Grimm-Cleanly. Here, the "easy-tight" relative semantic phrase is used as the representative of this color image axis, and is called the R axis.

Color energy contains the following phrases: Beautiful-Ugly, mourning-excited (Dejected-Encouraged), Gorgeous-Plain, Joyous-Grievous, Ill-Healthy, Clear-Vague. Here, the "beautiful-ugly" relative semantic phrase is used as the representative of this color image axis, and is called the B axis.

The maximum value of the H-axis, R-axis and B-axis of the HRB color image space is 100, that is, one end is positive 100 and the other end is negative 100. The use of the value of 100 is based on the common experience and habits of the public life, such as the perfect number of things, the full score in the early childhood education, the relatively 100% limit of things. In addition, the median value of a set of semantic scales is 0, which is the arithmetic mean value of the experimental values representing the meaning of the neutral image of the measured object. The advantages of the above-mentioned scale design not only match the user's psychological performance with the strong feeling of feelings in the practical application, but also the user is more likely to use the meaning of the actual size and size, which is beneficial to the use and The communication of image size.

Referring to FIG. 3 to FIG. 5, the three color images of the HRB color image space, the H, R, B and CIEL*a*b* color spaces, represent the color values of the three attributes L*, h*, and C*, respectively. Have a good relationship. At the same time, it also means that the HRB color image space and the CIEL*a*b* color space can be converted to each other. The conversion method can be calculated from formula (1) to formula (6): H=a ₁ L*+b ₁ ( 1)

R=a ₂ h* ³ +a ₃ h* ² +a ₄ h*+b ₂ (2)

B=a ₅ C*+b ₃ (3)

L*=a ₆ H+b ₄ (4)

h*=a ₇ R ³ +a ₈ R ² +a ₉ R+b ₅ (5)

C*=a ₁₀ B+b ₆ (6)

Among them, H, R, B are the coordinate values of the HRB color image space respectively, L*, h* and C* are the coordinate values of the CIEL*a*b* color space respectively, and the values of the coefficients a ₁ to a ₁₀ are - From 3000 to 3000, the values of the coefficients b ₁ to b ₆ range from -10000 to 10,000.

In one embodiment, the color image prediction system of the present invention further defines a color image intensity value CI that can be used to represent the intensity of a person's perception of color imagery. The color image intensity value CI can be calculated by the formula (7): CI = (R ² + B ² ) ^1/2 (7)

Please refer to FIG. 6 to FIG. 13 , according to various color systems: red, red, yellow, yellow, yellow green, green, green blue, blue, blue red, color image intensity value CI and color represented by CIEL*a*b* color space. The chroma C* characteristic value of one of the attributes has a good relationship, and the relationship can be expressed by the formula (7) to the formula (23). The above various correspondences have a good correlation, and the correlation coefficient can reach 0.61 on average, wherein the minimum value is 0.40; and the maximum value is 0.83.

Cr*=a ₁₁ CIr+b ₇ (8)

Cry*=a ₁₂ CIry ⁵ +a ₁₃ CIry ⁴ +a ₁₄ CIry ³ +a ₁₅ CIry ² +a ₁₆ CIry+b ₈ (9)

Cy*=a ₁₇ CIy ⁴ +a ₁₈ CIy ³ +a ₁₉ CIy ² +a ₂₀ CIy+b ₉ (10)

Cyg*=a ₅₁ CIyg ³ +a ₂₂ CIyg ² +a ₂₃ CIyg+b ₁₀ (11)

Cg*=a ₂₄ CIg ³ +a ₂₅ CIg ² +a ₂₆ CIg+b ₁₁ (12)

Cgb*=a ₂₇ CIgb ⁴ +a ₂₈ CIgb ³ +a ₂₉ CIgb ² +a ₃₀ CIgb+b ₁₂ (13)

Cb*=a ₃₁ CIb ³ +a ₃₂ CIb ² +a ₃₃ CIb+b ₁₃ (14)

Cp*=a ₃₄ CIp ⁴ +a ₃₅ CIp ³ +a ₃₆ CIp ² +a ₃₇ CIp+b ₁₄ (15)

CIr=a ₃₈ Cr* ⁵ +a ₃₉ Cr* ⁴ +a ₄₀ Cr* ³ +a ₄₁ Cr* ² +a ₄₂ Cr*+b ₁₅ (16)

CIry=a ₄₃ Cry* ⁶ +a ₄₄ Cry* ⁵ +a ₄₅ Cry* ⁴ +a ₄₆ Cry* ³ +a ₄₇ Cry* ² +a ₄₈ Cry*+b ₁₆ (17)

CI y=a ₄₉ Cy* ⁶ +a ₅₀ Cy* ⁵ +a ₅₁ Cy* ⁴ +a ₅₂ Cy* ³ +a ₅₃ Cy* ² +a ₅₄ Cy*+b ₁₇ (18)

CI yg=a ₅₅ Cyg* ⁴ +a ₅₆ Cyg* ³ +a ₅₇ Cyg* ² +a ₅₈ Cyg*+b ₁₈ (19)

CI g=a ₅₉ Cg* ⁴ +a ₆₀ Cg* ³ +a ₆₁ Cg* ² +a ₆₂ Cg*+b ₁₉ (20)

CI gb=a ₆₃ Cgb* ⁵ +a ₆₄ Cgb* ⁴ +a ₆₅ Cgb* ³ +a ₆₆ Cgb* ² +a ₆₇ Cgb*+b ₂₀ (21)

CI b=a ₆₈ Cb* ⁶ +a ₆₉ Cb* ⁵ +a ₇₀ Cb* ⁴ +a ₇₁ Cb* ³ +a ₇₂ Cb* ² +a ₇₃ Cb*+b ₂₁ (22)

CIp=a ₇₄ Cp* ⁶ +a ₇₅ Cp* ⁵ +a ₇₆ Cp* ⁴ +a ₇₇ Cp* ³ +a ₇₈ Cp* ² +a ₇₉ Cp*+b ₂₂ (23)

Among them, Cr*, Cry*, Cy*, Cyg*, Cg*, Cgb*, Cb*, and Cp* are respectively the red (r), red-yellow (ry), and yellow (y) of the CIEL*a*b* color space. ), yellow-green (yg), green (g), green-blue (gb), blue (b), purple (p) color chroma values, CIr, CIry, CIy, CIyg, CIg, CIgb, CIb, CIp For the color image intensity values of the corresponding color systems, the values of the coefficients a ₁₁ to a ₇₉ range from -3000 to 3000, and the values of the coefficients b ₇ to b ₂₂ range from -10000 to 10000.

It should be noted that in the above embodiment, the CIEL*a*b* color space system is taken as an example, but is not limited thereto. In one embodiment, the color image prediction system of the present invention further includes a color conversion module that can be used to convert the color information C or the candidate color CC into another color space system. Those having ordinary skill in the art to which the present invention pertains can understand how other color space systems are converted to the CIEL*a*b* color space system, and will not be described herein. For example, other color space systems may be color space representations such as RGB, XYZ, and the like.

In one embodiment, the input module 11 can be a graphical user interface (GUI), and the user can directly input the value of the color information C or the color image information I in the graphical user interface. Alternatively, the user can also input a color image through the graphical user interface, and then perform analysis processing to obtain the desired color information C. In an example, the input module 11 can include a connection interface and an analysis unit. The connection interface is electrically connected to an external electronic device such that the color image prediction system of the present invention can receive the color image output by the external electronic device. The analysis unit is electrically connected to the connection interface for analyzing the color image to obtain the color information C. In an embodiment, the analysis unit may utilize statistical analysis methods (coefficient of variation, correlation coefficient, arithmetic mean, geometric mean, etc.), numerical methods (eg, interpolation, optimization, etc.), color space representation ( Such as RGB, XYZ, CIEL*a*b*, etc.) or A combination of the above methods analyzes the color image to obtain color information C. For example, by using a color measuring instrument, a digital camera, a camera, etc. to obtain a selected color or image data, and then using a specific color message conversion or calculation principle for further analysis to obtain a correct color or color image. Color separation planning data, and then use this color data as a benchmark to predict color imagery.

Referring to FIG. 14, a color image prediction method according to an embodiment of the present invention includes an input step S21, a prediction step S22, and an output step S23. The input step is for inputting a color information or a color image information. The predicting step is for mapping the color information to an HRB color image space to obtain a predicted image value, or to assign the color image information to the HRB color image space to obtain a candidate color. The detailed description of the HRB color image space has been described above and will not be described here. The output step is for outputting a predicted image value or a candidate color. In one embodiment, the color image prediction method of the present invention further includes a color conversion step for converting color information or candidate colors into another color space system.

In summary, the color image prediction system and method of the present invention establishes a HRB color image space by using relative semantic phrases, so that the color space system and the color mental feeling can be mutually converted, thereby predicting the user's color image or providing candidate colors. . Therefore, the field of application of the present invention can cover those related to the evaluation and prediction of the qualitative expression of the color psychological feeling tendency and the quantitative evaluation, such as: monochrome, multi-color matching clothing, daily necessities, overall appearance of the building, landscape environment , color TV, smart phones, network-related color image prediction and application, photography and painted images and other color image representation and prediction, as well as music and color co-feeling image representation and prediction, color and human sexual orientation, fortune and exploration Interesting life applications related to the island.

The embodiments described above are only intended to illustrate the technical idea and the features of the present invention, and the purpose of the present invention is to enable those skilled in the art to understand the contents of the present invention and to implement the present invention. That is, the equivalent variations or modifications made by the spirit of the present invention should still be included in the scope of the present invention.

11. . . Input module

12. . . Prediction module

13. . . Output module

C. . . Color information

CC. . . Candidate color

H, R, B. . . Color image axis

I. . . Color image information

PI. . . Predicted image value

S21~S23. . . Steps for color image prediction

1 is a block diagram showing a color image prediction system in accordance with an embodiment of the present invention.

2 is a schematic diagram showing an HRB color image space of one of the color image prediction systems of an embodiment of the present invention.

3 is a diagram showing the relationship between the H value of the HRB color image space and the L* value of the CIEL*a*b* color space in an embodiment of the present invention.

4 is a diagram showing the relationship between the R value of the HRB color image space and the h* value of the CIEL*a*b* color space in an embodiment of the present invention.

Figure 5 is a diagram showing the relationship between the B value of the HRB color image space and the C* value of the CIEL*a*b* color space in accordance with an embodiment of the present invention.

6 to FIG. 13 are diagrams showing a color image intensity value CI and a CIEL*a*b* color space of a color image prediction system according to an embodiment of the present invention, red, yellow, yellow, yellow, green, green, and green. The relationship between the C* values of blue, blue, and purple.

Figure 14 is a flow chart showing a color image prediction method according to an embodiment of the present invention.

H, R, B. . . Color image axis

Claims (18)

A color image prediction system includes: an input module for inputting a color information or a color image information by a user; and a prediction module for matching the color information to an HRB color image space Obtaining a predicted image value, or corresponding the color image information to the HRB color image space to obtain a candidate color, wherein the HRB color image space comprises an H axis, an R axis, and a B axis, the H axis representing a hard - soft, deep - shallow, thick - thin, loose - tight, bright - bleak, strong - weak, the R axis represents ease - tension, sputum - prime, nature - manipulation, oppression - freedom Mild - ferocious, dirty - clean, the B axis represents beauty - ugly, stunned - excited, luxurious - simple, happy - sad, sick - healthy, clear - fuzzy, and the H axis, the R axis and the B axis The coordinate value is -100 to 100; and an output module electrically connected to the prediction module for outputting the predicted image value or the candidate color. The color image prediction system according to claim 1, wherein the prediction module calculates the predicted image value according to the following formula: H=a ₁ L*+b ₁ R=a ₂ h* ³ +a ₃ h* ² + a ₄ h*+b ₂ B=a ₅ C*+b ₃ where H, R, and B are the coordinate values of the HRB color image space, respectively, and L*, h*, and C* respectively correspond to the CIEL *a*b* Coordinate values of the color space, the values of the coefficients a ₁ to a ₅ range from -3000 to 3000, and the values of the coefficients b ₁ to b ₃ range from -10000 to 10000. The color image prediction system of claim 1, wherein the prediction module calculates the candidate color according to the following formula: L*=a ₆ H+b ₄ h*=a ₇ R ³ +a ₈ R ² +a ₉ R+b ₅ C*=a ₁₀ B+b ₆ where L*, h*, and C* are the coordinate values of the CIEL*a*b* color space, respectively, and H, R, and B respectively correspond to the color image information. The coordinate value of the HRB color image space, the values of the coefficients a ₆ to a ₁₀ range from -3000 to 3000, and the values of the coefficients b ₄ to b ₆ range from -10000 to 10000. The color image prediction system according to claim 1, wherein the prediction module calculates a color image intensity value according to the following formula: CI=(R ² +B ² ) ^1/2 where CI is a color image intensity value, R and B are the coordinate values of the HRB color image space, respectively. The color image prediction system according to claim 4, wherein the relationship between the color image intensity value and the C* coordinate value of the CIEL*a*b* color space conforms to the following formula: Cr*=a ₁₁ CIr+b ₇ Cry*= a ₁₂ CIry ⁵ +a ₁₃ CIry ⁴ +a ₁₄ CIry ³ +a ₁₅ CIry ² +a ₁₆ CIry+b ₈ Cy*=a ₁₇ CIy ⁴ +a ₁₈ CIy ³ +a ₁₉ CIy ² +a ₂₀ CIy+b ₉ Cyg*=a ₂₁ CIyg ³ +a ₂₂ CIyg ² +a ₂₃ CIyg+b ₁₀ Cg*=a ₂₄ CIg ³ +a ₂₅ CIg ² +a ₂₆ CIg+b ₁₁ Cgb*=a ₂₇ CIgb ⁴ +a ₂₈ CIgb ³ +a ₂₉ CIgb ² +a ₃₀ CIgb+b ₁₂ Cb*=a ₃₁ CIb ³ +a ₃₂ CIb ² +a ₃₃ CIb+b ₁₃ Cp*=a ₃₄ CIp ⁴ +a ₃₅ CIp ³ +a ₃₆ CIp ² +a ₃₇ CIp+b ₁₄ where Cr*, Cry*, Cy*, Cyg*, Cg*, Cgb*, Cb*, Cp* are respectively CIEL*a*b* color space red, red yellow, yellow, yellow green , green, green, blue, purple, chroma values, CIr, CIry, CIy, CIyg, CIg, CIgb, CIb, CIp are the color image intensity values of the corresponding color system, the numerical range of the coefficients a ₁₁ to a ₃₇ The value ranges from -3000 to 3,000, and the values of the coefficients b ₇ to b ₁₄ range from -10000 to 10,000. The color image prediction system of claim 4, wherein the relationship between the color image intensity value and the C* coordinate value of the CIEL*a*b* color space conforms to the following formula: CIr=a ₃₈ Cr* ⁵ +a ₃₉ Cr* ⁴ +a ₄₀ Cr* ³ +a ₄₁ Cr* ² +a ₄₂ Cr*+b ₁₅ CIry=a ₄₃ Cry* ⁶ +a ₄₄ Cry* ⁵ +a ₄₅ Cry* ⁴ +a ₄₆ Cry* ³ +a ₄₇ Cry * ² +a ₄₈ Cry*+b ₁₆ CI y=a ₄₉ Cy* ⁶ +a ₅₀ Cy* ⁵ +a ₅₁ Cy* ⁴ +a ₅₂ Cy* ³ +a ₅₃ Cy* ² +a ₅₄ Cy*+ b ₁₇ CI yg=a ₅₅ Cyg* ⁴ +a ₅₆ Cyg* ³ +a ₅₇ Cyg* ² +a ₅₈ Cyg*+b ₁₈ CI g=a ₅₉ Cg* ⁴ +a ₆₀ Cg* ³ +a ₆₁ Cg* ² +a ₆₂ Cg*+b ₁₉ CI gb=a ₆₃ Cgb* ⁵ +a ₆₄ Cgb* ⁴ +a ₆₅ Cgb* ³ +a ₆₆ Cgb* ² +a ₆₇ Cgb*+b ₂₀ CI b=a ₆₈ Cb* ⁶ +a ₆₉ Cb* ⁵ +a ₇₀ Cb* ⁴ +a ₇₁ Cb* ³ +a ₇₂ Cb* ² +a ₇₃ Cb*+b ₂₁ CIp=a ₇₄ Cp* ⁶ +a ₇₅ Cp* ⁵ +a ₇₆ Cp* ⁴ + a ₇₇ Cp* ³ +a ₇₈ Cp* ² +a ₇₉ Cp*+b ₂₂ where Cr*, Cry*, Cy*, Cyg*, Cg*, Cgb*, Cb*, Cp* are respectively CIEL*a* b* color space red, red, yellow, yellow, green, green, blue, blue, blue and red C* coordinates, CIr, CIry, CIy, CIyg, CIg, CIgb, CIb, CIp are the corresponding colors Color image intensity value, coefficient a ₃₈ to a The value range of ₇₉ is -3000 to 3000, and the values of the coefficients b ₁₅ to b ₂₂ range from -10000 to 10000. The color image prediction system of claim 1, further comprising: a color conversion module for converting the color information or the candidate color into another color space system. The color image prediction system of claim 1, wherein the input module comprises: a connection interface for electrically connecting to an external electronic device to receive a color image output by the external electronic device; and An analysis unit electrically connected to the connection interface for analyzing the color image to obtain the color information. The color image prediction system of claim 8, wherein the analysis unit analyzes the color image using a statistical analysis method, a numerical method, a color space representation, or a combination of the above methods to obtain the color information. A color image prediction method includes: an input step for inputting a color information or a color image information; and a predicting step for corresponding the color information to an HRB color image space to obtain a predicted image value, Or corresponding to the color image information to the HRB color image space Obtaining a candidate color, wherein the HRB color image space comprises an H-axis, an R-axis, and a B-axis, the H-axis representing hard-soft, deep-shallow, thick-thin, loose-tight, Bright - bleak, strong - weak, the R axis represents relaxed - nervous, awkward - prime, natural - artificial, oppressive - free, gentle - ferocious, dirty - clean, the B axis represents beauty - ugly, mourning - Excitement, luxury - simplicity, happiness - sadness, illness - health, clarity - blur, and the H-axis, the R-axis and the B-axis have a coordinate value of -100 to 100; and an output step for outputting the prediction Image value or the candidate color. The color image prediction method according to claim 10, wherein the predicted image value is calculated according to the following formula: H = a ₁ L * + b ₁ R = a ₂ h * ³ + a ₃ h * ² + a ₄ h *+b ₂ B=a ₅ C*+b ₃ where H, R, and B are the coordinate values of the HRB color image space, respectively, and L*, h*, and C* respectively correspond to the CIEL*a* The coordinate value of the b* color space, the values of the coefficients a ₁ to a ₅ range from -3000 to 3000, and the values of the coefficients b ₁ to b ₃ range from -10000 to 10000. The color image prediction method according to claim 10, wherein the candidate color is calculated according to the following formula: L*=a ₆ H+b ₄ h*=a ₇ R ³ +a ₈ R ² +a ₉ R+b ₅ C*=a ₁₀ B+b ₆ where L*, h*, and C* are the coordinate values of the CIEL*a*b* color space, respectively, and H, R, and B respectively indicate that the color image information corresponds to the HRB color. The coordinate value of the image space, the values of the coefficients a ₆ to a ₁₀ range from -3000 to 3000, and the values of the coefficients b ₄ to b ₆ range from -10000 to 10000. The color image prediction method according to claim 10, wherein the prediction step further calculates a color image intensity value according to the following formula: CI=(R ² +B ² ) ^1/2 wherein CI is a color image intensity value, R And B are the coordinate values of the HRB color image space respectively. The color image prediction method according to claim 13, wherein the relationship between the color image intensity value and the C* coordinate value of the CIEL*a*b* color space conforms to the following formula: Cr*=a ₁₁ CIr+b ₇ Cry*= a ₁₂ CIry ⁵ +a ₁₃ CIry ⁴ +a ₁₄ CIry ³ +a ₁₅ CIry ² +a ₁₆ CIry+b ₈ Cy*=a ₁₇ CIy ⁴ +a ₁₈ CIy ³ +a ₁₉ CIy ² +a ₂₀ CIy+b ₉ Cyg*=a ₂₁ CIyg ³ +a ₂₂ CIyg ² +a ₂₃ CIyg+b ₁₀ Cg*=a ₂₄ CIg ³ +a ₂₅ CIg ² +a ₂₆ CIg+b ₁₁ Cgb*=a ₂₇ CIgb ⁴ +a ₂₈ CIgb ³ +a ₂₉ CIgb ² +a ₃₀ CIgb+b ₁₂ Cb*=a ₃₁ CIb ³ +a ₃₂ CIb ² +a ₃₃ CIb+b ₁₃ Cp*=a ₃₄ CIp ⁴ +a ₃₅ CIp ³ +a ₃₆ CIp ² +a ₃₇ CIp+b ₁₄ where Cr*, Cry*, Cy*, Cyg*, Cg*, Cgb*, Cb*, Cp* are respectively CIEL*a*b* color space red, red yellow, yellow, yellow green , green, green, blue, purple, chroma values, CIr, CIry, CIy, CIyg, CIg, CIgb, CIb, CIp are the color image intensity values of the corresponding color system, the numerical range of the coefficients a ₁₁ to a ₃₇ The value ranges from -3000 to 3,000, and the values of the coefficients b ₇ to b ₁₄ range from -10000 to 10,000. The color image prediction method according to claim 13, wherein the relationship between the color image intensity value and the C* coordinate value of the CIEL*a*b* color space conforms to the following formula: CIr=a ₃₈ Cr* ⁵ +a ₃₉ Cr* ⁴ +a ₄₀ Cr* ³ +a ₄₁ Cr* ² +a ₄₂ Cr*+b ₁₅ CIry=a ₄₃ Cry* ⁶ +a ₄₄ Cry* ⁵ +a ₄₅ Cry* ⁴ +a ₄₆ Cry* ³ +a ₄₇ Cry * ² +a ₄₈ Cry*+b ₁₆ CI y=a ₄₉ Cy* ⁶ +a ₅₀ Cy* ⁵ +a ₅₁ Cy* ⁴ +a ₅₂ Cy* ³ +a ₅₃ Cy* ² +a ₅₄ Cy*+b ₁₇ CI yg=a ₅₅ Cyg* ⁴ +a ₅₆ Cyg* ³ +a ₅₇ Cyg* ² +a ₅₈ Cyg*+b ₁₈ CI g=a ₅₉ Cg* ⁴ +a ₆₀ Cg* ³ +a ₆₁ Cg* ² +a ₆₂ Cg*+b ₁₉ CI gb=a ₆₃ Cgb* ⁵ +a ₆₄ Cgb* ⁴ +a ₆₅ Cgb* ³ +a ₆₆ Cgb* ² +a ₆₇ Cgb*+b ₂₀ CI b=a ₆₈ Cb* ⁶ +a ₆₉ Cb* ⁵ +a ₇₀ Cb* ⁴ +a ₇₁ Cb* ³ +a ₇₂ Cb* ² +a ₇₃ Cb*+b ₂₁ CIp=a ₇₄ Cp* ⁶ +a ₇₅ Cp* ⁵ +a ₇₆ Cp* ⁴ + a ₇₇ Cp* ³ +a ₇₈ Cp* ² +a ₇₉ Cp*+b ₂₂ where Cr*, Cry*, Cy*, Cyg*, Cg*, Cgb*, Cb*, Cp* are respectively CIEL*a* b* color space red, red, yellow, yellow, green, green, blue, blue, blue and red C* coordinates, CIr, CIry, CIy, CIyg, CIg, CIgb, CIb, CIp are the corresponding colors Color image intensity value, coefficient a ₃₈ to a The value range of ₇₉ is -3000 to 3000, and the values of the coefficients b ₁₅ to b ₂₂ range from -10000 to 10000. The color image prediction method of claim 10, further comprising: a color conversion step for converting the color information or the candidate color into another color space system. The color image prediction method of claim 10, wherein the inputting step is to receive a color image output by the external electronic device, and analyze the color image to obtain the color information. The color image prediction method of claim 17, wherein the input step analyzes the color image using a statistical analysis method, a numerical method, a color space representation, or a combination of the above methods to obtain the color information.