JP3539883B2 - Image processing method and apparatus, recording medium, imaging apparatus, and image reproducing apparatus - Google Patents

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an image processing method and apparatus for converting a color of an image when a color image is reproduced on a color print or a color display, and a computer-readable recording medium storing a program for causing a computer to execute the image processing method. , An imaging device, and an image reproducing device.
[0002]
[Prior art]
2. Description of the Related Art In a digital electronic still camera (hereinafter referred to as a digital camera), an image obtained by shooting is stored as digital image data in a recording medium such as an internal memory or an IC card provided inside the digital camera, and the stored digital image data is stored. Based on the image, an image obtained by shooting as a hard copy such as a print or as a soft copy on a display can be reproduced. As described above, when reproducing an image obtained by a digital camera, it is expected that the image has the same high quality as a photograph printed from a negative film.
[0003]
On the other hand, when photographing with a digital camera, the photographing is performed under various photographing light sources such as tungsten light, fluorescent light, or outdoor daylight, so that the photographed image corresponds to the photographing light source. A digital camera having a function of automatically setting a white balance has been proposed. Further, since a main subject is imaged at the center position of the image, a method has been proposed in which the image is divided and the white balance is obtained by increasing the weight in the divided region at the center position (screen division is performed. Automatic white balance adjustment system used, Kikuchi et al., Report of the Institute of Television Engineers of Japan, 1990.6.26, PP19-24).
[0004]
When the automatic correction is performed in the digital camera as described above, for example, for an image acquired under tungsten light, adjustment is performed so that the RGB tristimulus values for tungsten light have the same balance. On the other hand, in human vision, chromatic adaptation is a mechanism that basically recognizes white paper as white even under daylight or under fluorescent lamps having a different spectral energy distribution from daylight. However, when white paper is viewed under tungsten light, the white paper is often perceived as slightly yellowish. This is a phenomenon that occurs because the adaptation is not completely performed, and is known as incomplete adaptation. Incomplete adaptation is likely to occur when the color of the light source is strong or when the luminance is low. Therefore, it is desired to convert the color of an image so that a reproduced image has the same appearance as that at the time of shooting, in consideration of incomplete adaptation and chromatic adaptation of human vision.
[0005]
For this reason, in Japanese Patent Application No. 9-207883, when photographing conditions such as a light source and brightness at the time of photographing with a digital camera are recorded on a recording medium together with image data, the photographing conditions are taken together with the image data and used as photographing conditions. An image processing method has been proposed in which an image faithful to the original scene is reproduced as a hard copy or a soft copy by performing appropriate image processing.
[0006]
[Problems to be solved by the invention]
However, the method proposed in Japanese Patent Application No. 9-207883 described above is based on the premise that the digital camera has a function of recording shooting conditions on a recording medium, and is effective in the case of a general digital camera. is not.
[0007]
On the other hand, in the apparatus for automatically setting the white balance as described above, it is necessary to provide a means for automatically setting the white balance in the camera, which increases the cost. Is often not provided. Furthermore, in order to obtain a high-quality playback image in a digital camera that does not have a function of automatically setting a white balance, an image is displayed on a display when reproducing image data, and while observing the displayed image, It is necessary to adjust the color balance and white balance. However, since such adjustment requires skill, it is difficult for a general user to adjust white balance and the like to obtain a satisfactory reproduced image.
[0008]
The present invention has been made in view of the above circumstances, and can be simplified in operation with a simple configuration, and the reproduced image has the same appearance as the original scene at the time of shooting. To provide an image processing method and apparatus capable of converting an image into an image, a computer-readable recording medium storing a program for causing a computer to execute the image processing method, an imaging apparatus, and an image reproducing apparatus It is.
[0009]
[Means for Solving the Problems]
A first image processing method according to the present invention calculates a physiological primary color of each pixel in an image represented by image data obtained by imaging,
The image data is subjected to image processing based on the physiological primary colors to obtain processed image data.
[0010]
Here, "obtain by imaging" means that a subject is photographed by a digital camera to obtain image data representing the subject image, and that a subject such as a negative film or a reversal film on which the subject image is recorded is read by a scanner or the like. This means that an image is read by an apparatus to obtain image data representing a subject image.
[0011]
Further, the second image processing method according to the present invention converts the RGB color of each pixel in the image represented by the image data acquired by the imaging into a chromaticity value,
Calculating a chromaticity average value that is an average value of the chromaticity values of the respective pixels;
The chromaticity average value is used as the physiological primary color.CorrespondingConvert to tristimulus values,
To the physiological primary colorsCorrespondingConsidering the tristimulus value as white of the photographing light source at the time of acquiring the image data, converting the image data so that the tristimulus value matches the chromaticity value of the white point of the reference photographing light source, and processing the processed image data. Is obtained.
[0012]
Here, “physiological primary colors” refer to primary colors perceived by visual cells of an eye when a human looks at an object. The phenomenon that white paper looks white under various light sources is considered to be achieved by adjusting the response amounts (tristimulus values) of red, green, and blue represented by physiological primary colors under the light source. ing.Specifically, the tristimulus values corresponding to the physiological primary colors can be calculated by converting the pixel values (for example, RGB) of each pixel in the image represented by the image data based on a predetermined conversion formula. .
[0013]
Further, the “reference photographing light source” refers to a photographing light source to be referred to when acquiring image data, unlike the photographing light source when acquiring image data. Specifically, in a general camera, daylight of D50 to D60 is used as a reference imaging light source.
[0014]
In the second image processing method according to the present invention, the physiological primary colorsCorrespondingIt is preferable that the degree of the coincidence is reduced as the tristimulus value differs from the chromaticity value of the white point of the reference imaging light source.
[0015]
Further, a third image processing method according to the present invention converts RGB colors of each pixel in an image represented by image data obtained by imaging into chromaticity values,
Calculating a chromaticity average value that is an average value of the chromaticity values of the respective pixels;
The chromaticity average value is converted to RGB tristimulus values,
The RGB tristimulus values are regarded as the white color of the imaging light source when the image data is obtained, and the image data is converted and processed so that the tristimulus values match the chromaticity values of the white point of the reference imaging light source. It is characterized by obtaining completed image data.
[0016]
In the third image processing method according to the present invention, it is preferable that the degree of the coincidence is reduced as the RGB tristimulus values are different from the chromaticity value of the white point of the reference photographing light source.
[0017]
In the second and third image processing methods according to the present invention, the chromaticity average value is calculated as a weighted average value according to a reciprocal of a distance of the chromaticity value from a CIE daylight locus or a black body locus. Is preferred.
[0018]
Here, “CIE daylight locus” is a trajectory indicating daylight on the CIE chromaticity diagram, and “blackbody locus” is a relation between the absolute temperature of the blackbody and the chromaticity at that time. It is a representation. These are both commonly used.
[0019]
Further, “weighting according to the reciprocal of the distance” means that the weighting becomes smaller as the chromaticity value of the pixel is farther from the daylight locus or the blackbody locus.
[0020]
Further, in the second and third image processing methods according to the present invention, it is preferable that the white point of the reference light source is set to have a value within a predetermined range based on the white point of the light source.
[0021]
Here, "setting to have a value in a predetermined range" means not setting the white point of the reference light source to only one point, but setting to have a value in a certain range.
[0022]
In the image processing method according to the present invention, it is preferable that the processed image data is further converted based on a reproduction target value of a reproduction device that reproduces the processed image data.
[0023]
A first image processing apparatus according to the present invention includes: a physiological primary color calculating unit configured to calculate a physiological primary color of each pixel in an image represented by image data acquired by imaging;
Image processing means for performing image processing on the image data based on the physiological primary colors to obtain processed image data.
[0024]
The second image processing apparatus according to the present invention further includes a chromaticity value conversion unit configured to convert the RGB colors of each pixel in the image represented by the image data acquired by the imaging into a chromaticity value;
Chromaticity average value calculating means for calculating a chromaticity average value which is an average value of the chromaticity values of the respective pixels,
The chromaticity average value is used as the physiological primary color.CorrespondingTristimulus value conversion means for converting to tristimulus values,
To the physiological primary colorsCorrespondingConsidering the tristimulus value as white of the photographing light source at the time of acquiring the image data, converting the image data so that the tristimulus value matches the chromaticity value of the white point of the reference photographing light source, and processing the processed image data. And a color adaptation conversion means for obtaining
[0025]
In the second image processing apparatus according to the present invention, the chromatic adaptation conversion unit converts the chromatic adaptationCorrespondingIt is preferable that the unit is a unit that reduces the degree of coincidence as the tristimulus value differs from the chromaticity value of the white point of the reference imaging light source.
[0026]
Further, a third image processing device according to the present invention includes a chromaticity value conversion unit that converts RGB colors of each pixel in an image represented by image data acquired by imaging into chromaticity values,
Chromaticity average value calculating means for calculating a chromaticity average value which is an average value of the chromaticity values of the respective pixels,
Tristimulus value conversion means for converting the chromaticity average value into RGB tristimulus values;
The RGB tristimulus values are regarded as the white color of the imaging light source when the image data is obtained, and the image data is converted and processed so that the tristimulus values match the chromaticity values of the white point of the reference imaging light source. Chromatic adaptation conversion means for obtaining finished image data.
[0027]
In the third image processing apparatus according to the present invention, the chromatic adaptation conversion unit reduces the degree of the coincidence as the RGB tristimulus values are different from the chromaticity value of the white point of the reference imaging light source. It is preferably a means.
[0028]
In the second and third image processing apparatuses according to the present invention, the chromaticity average value calculating means may calculate the chromaticity average value as a distance of the chromaticity value from a CIE daylight locus or a blackbody locus. It is preferable that the means is a means for calculating a weighted average value according to the reciprocal.
[0029]
Further, the second and third image processing devices according to the present invention further comprise a unit for setting a white point of the reference photographing light source to have a value within a predetermined range based on the white point of the photographing light source. Is preferred.
[0030]
The image processing apparatus according to the present invention preferably further includes an output conversion unit that converts the processed image data based on a reproduction target value of a reproduction apparatus that reproduces the processed image data.
[0031]
Note that the program for causing a computer to execute the first to third image processing methods according to the present invention may be provided by being recorded on a computer-readable recording medium.
[0032]
An imaging device according to the present invention is an imaging unit that captures a subject image and obtains image data representing the subject image,
An image processing apparatus according to the present invention for performing image processing on the image data,
Recording means for recording, on a recording medium, processed image data obtained by performing processing by the image processing apparatus.
[0033]
An image reproducing apparatus according to the present invention is an image reproducing apparatus that reproduces image data as a visible image,
Input means for inputting the image data,
An image processing apparatus according to the present invention for performing image processing on the image data input from the input unit;
And a reproducing means for reproducing processed image data obtained by performing image processing by the image processing apparatus.
[0034]
【The invention's effect】
According to the first image processing method and apparatus of the present invention, image processing is performed on image data based on physiological primary colors in pixels represented by image data, so that when a person sees an object, the image is recognized. The image data is subjected to image processing based on the color to be processed. For this reason, it is possible to perform image processing on the image data so that an image faithful to the appearance impression is reproduced, and to reproduce the processed image data on which the image processing has been performed, thereby giving the appearance impression. A faithful image can be reproduced.
[0035]
According to the second image processing method and apparatus of the present invention, the RGB colors of each pixel are converted into chromaticity values, and the chromaticity average value, which is the average of the chromaticity values, is calculated. Degree average value becomes physiological primary colorCorrespondingConverted to tristimulus values. In the present invention, the physiological primary colors determined from the image dataCorrespondingThe tristimulus value is regarded as white perceived when a human sees the object on the spot, that is, the white color of the imaging light source when imaging the image data, and the tristimulus value is set to match the white point of the reference imaging light source. First, the image data is converted. Here, the appearance of the image differs depending on the light source at the time of capturing the image data. For example, when the image is captured on a cloudy day or in a shade, the white color of the image is entirely blue. Looks like it. However, humans perceive as white by chromatic adaptation. Therefore, in the physiological primary colorsCorrespondingBy converting the image data so that the tristimulus values match the white point of the reference imaging light source, the white color represented by the image data will be converted to the same white color as obtained in the reference imaging light source. As a result, by reproducing the processed image data, it is possible to reproduce an image faithful to the impression of how the image looks when the image data is acquired.
[0036]
Also, as the tristimulus value deviates from the chromaticity value of the white point of the reference imaging light source, as in the case of looking at a white paper under a tungsten light source, the tristimulus value and the chromaticity value of the white point of the reference imaging light source By reducing the degree of coincidence, the image data can be converted so as to leave the atmosphere of the shooting light source at the time of shooting, whereby an image faithful to the appearance of the original scene can be reproduced.
[0037]
Further, according to the third image processing method and apparatus of the present invention, the RGB colors of each pixel are converted into chromaticity values, and the chromaticity average value, which is the average of the chromaticity values, is calculated. The degree average value is converted to RGB tristimulus values. Then, similarly to the second image processing method and apparatus, the RGB tristimulus values are regarded as the white color of the imaging light source when the image data is captured, and the tristimulus values match the white point of the reference imaging light source. Thus, the image data is converted. Therefore, the white color represented by the image data is converted to the same white color as that obtained by the reference photographing light source, and as a result, by reproducing the processed image data, the image obtained when the image data is obtained It is possible to reproduce an image that is faithful to the impression of how the image looks. Further, in the second image processing method and apparatus, the physiological primary colorsCorrespondingSince the color adaptation conversion is performed based on the tristimulus values, it is necessary to convert the image data into physiological primary colors at the time of the color adaptation conversion. However, in the third image processing method and apparatus, the RGB image is converted to the RGB tristimulus values. Since the chromatic adaptation conversion is performed based on the image data, it is not necessary to perform the chromatic adaptation conversion by converting the image data into physiological primary colors, thereby reducing the calculation time and simplifying the configuration of the apparatus. Can be.
[0038]
Also, in the third image processing method and apparatus, as in the case of viewing white paper under a tungsten light source, the degree of coincidence increases as the RGB tristimulus values depart from the chromaticity value of the white point of the reference imaging light source. Is reduced, the image data can be converted so as to leave the atmosphere of the shooting light source at the time of shooting, whereby an image faithful to the appearance of the original scene can be reproduced.
[0039]
Further, in the CIE daylight locus or blackbody locus on the chromaticity diagram, a chromaticity value located farther away from these can be regarded as an object color such as red clothes and a blue sky rather than a light source color. Therefore, by calculating the average chromaticity value as a weighted average value corresponding to the reciprocal of the distance from the daylight locus or the blackbody locus, the physiological primary colorCorrespondingThe tristimulus values or the RGB tristimulus values more accurately represent the white color of the light source during imaging. Therefore, by obtaining tristimulus values based on the chromaticity average values thus obtained, more accurate image processing can be performed.
[0040]
Further, by setting the white point of the reference photographing light source to have a value within a predetermined range based on the white point of the photographing light source, it is possible to perform a process reflecting the color of the photographing light source at the time of acquiring the image data. Can be done. In addition, since the white point of the reference imaging light source has a value within a predetermined range, an appropriate white point and a physiological primary color within that range are obtained.CorrespondingBy matching the tristimulus values or the RGB tristimulus values, it is possible to prevent excessive chromatic adaptation conversion as compared with the case where the tristimulus values are matched to a certain white point. It is possible to obtain processed image data capable of reproducing a natural image.
[0041]
Further, since the processed image data is corrected so that its white point becomes a white point obtained under the reference imaging light source, the processed image data is further converted based on the reproduction target value of the reproduction apparatus. In addition, the processed image data can be converted to have an appropriate reproduction target value. Therefore, a higher quality reproduced image can be obtained by reproducing the converted processed image data.
[0042]
Furthermore, since the imaging device according to the present invention includes the image processing device according to the present invention, the image data acquired by imaging is recorded on a recording medium as processed image data capable of reproducing an impression of the appearance at the time of imaging. Will be done. Therefore, it is possible to obtain image data capable of reproducing an image faithful to the impression of the appearance at the time of imaging simply by imaging the subject.
[0043]
Still further, since the image reproducing apparatus according to the present invention includes the image processing apparatus according to the present invention, the input image data is subjected to image processing so as to reproduce an impression of appearance at the time of photographing and reproduced. Will be offered. Therefore, it is possible to reproduce an image faithful to the appearance at the time of shooting simply by inputting image data.
[0044]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a schematic diagram showing a configuration of an image processing system to which an image processing method and apparatus according to a first embodiment of the present invention are applied. As shown in FIG. 1, an image processing system to which the image processing method and apparatus according to the first embodiment is applied uses a reading device such as a scanner to read an image obtained by a digital camera or recorded on a negative film or a reversal film. A reading unit 2 for reading the image data S from a recording medium 1 such as an optical disk in which the image data S obtained by reading is stored; And a reproducing means 4 for reproducing the image data S2 subjected to the image processing in the image processing means 3 as a hard copy such as a print or a soft copy on a CRT monitor or the like.
[0045]
The image data S stored in the memory of the digital camera is copied and stored via a card reader or a cable in the recording medium 1. When the image data S is obtained in the reading device, the image data S is stored in the recording medium 1 in a data storage unit provided in the reading device.
[0046]
The image processing means 3 converts a tristimulus value R, G, B of each pixel when the image data S is reproduced as an image into a tristimulus value X, Y, Z and a chromaticity value u, v. An average value calculator 6 for calculating the average values u0 and v0 of the chromaticity values u and v obtained by the color converter 5, and an average value u0 and v0 calculated by the average value calculator 6 as the physiological primary colors L0. , M0, S0, and the physiological primaries L, M, S, which convert the tristimulus values X, Y, Z determined by the color converter 5 into physiological primaries L, M, S. Adaptation conversion based on the physiological primary colors L0, M0, S0 obtained by the physiological primary color conversion unit 7, and the physiological primary colors L ', M', S 'obtained by the color adaptation conversion unit 8. To an RGB conversion unit 9 for converting R ′, G ′, and B ′ into tristimulus values. Tristimulus values R obtained Te ', G', and an output unit 10 for performing image processing such as gradation conversion on B '.
[0047]
Next, the operation of the first embodiment will be described.
[0048]
First, the image data S is read from the recording medium 1 by the reading unit 2 and input to the image processing unit 3. In the image processing means 3, first, in the color conversion unit 5, the tristimulus values R, G, and B of each pixel when the image data S is reproduced as an image are expressed by the following formulas (1) and (2) according to the CIE1931XYZ tristimulus. Values and CIE1976uv chromaticity values. Here, the chromaticity value refers to the values of u and v in the figure representing the hue and saturation as shown in FIG. 2 among the three attributes of color, hue, saturation, and lightness, ie, the lightness direction is projected. It is the value that was calculated. Note that, in the CIE1976 chromaticity diagram, the notations u ′ and v ′ are used accurately, but in the present application, they are referred to as u and v.
[0049]
XR
Y = | A | · G (1)
Z B
u = 4X / (X + 15Y + 3Z)
v = 9Y / (X + 15Y + 3Z) (2)
Here, the matrix | A | is a matrix for converting the tristimulus values R, G, and B into tristimulus values X, Y, and Z. For example, the following values can be used.
[0050]
.4124 0.3576 0.1805
| A | = 0.2126 0.7152 0.0722 (3)
.0193 0.1192 0.9505
Note that, instead of the matrix | A |, the tristimulus values X, Y, and Z may be obtained from a look-up table.
[0051]
The signals r, g, b obtained from the digital camera are based on ITU-R BT. 709 (REC. 709), for example, expressed by encoding to the power of 0.45 as shown in the following equation (4).
[0052]
r = 255 × 4.500R (0 ≦ R <0.018)
r = 255 × (1.099R^0.45-0.099) (0.018 ≦ R ≦ 1) (4)
Here, R is a tristimulus value captured by the camera, and is a normalized signal (0 ≦ R ≦ 1). G and B are similarly encoded into g and b. In such a case, it is preferable that the above equation (4) is solved for R, converted into normalized tristimulus values R, G, B, and then subjected to color conversion.
[0053]
The average value calculation unit 6 calculates the average values u0 and v0 of the chromaticity values u and v obtained by the above equation (2) by the following equation (5).
[0054]
u0 = Σu / n
v0 = Σv / n (5)
Where n is the number of pixels
The physiological primary color conversion unit 7 converts the average values u0, v0 calculated by the average value calculation unit 6 into physiological primary colors L0, M0, S0 according to the following equations (6) to (8).
[0055]
sx = 9.0 × u0 (6.0 × u0-16.0 × v0 + 12.0)
sy = 4.0 × v0 (6.0 × u0-16.0 × v0 + 12.0) (6)
X0 = sx / sy
Y0 = 100.0 (7)
Z0 = (1.0−sx−sy) / sy
L0 X0
M0 = | B | · Y0 (8)
S0 Z0
Here, the matrix | B | is a matrix for converting the tristimulus values X0, Y0, Z0 into physiological primary colors L0, M0, S0, and for example, the following values can be used.
[0056]
.3897 0.6890 -0.0787
| B | = -0.2298 1.1834 0.0464 (9)
 0 1.0
In addition, the physiological primary color conversion unit 7 converts the tristimulus values X, Y, and Z of each pixel obtained by Expression (1) into physiological primary colors L, M, and S by Expression (8).
[0057]
The chromatic adaptation conversion unit 8 performs chromatic adaptation conversion on the physiological primary colors L, M, and S according to the following equations (10) and (11) to obtain converted physiological primary colors L ′, M ′, and S ′.
[0058]
L 'L
M ′ = | C | · M (10)
S 'S
However,
(a · Ln + (1-a) · L0) / L0
| C | = (a · Mn + (1-a) · M0) / M0 (11)
(aSn + (1-a) S0) / S0
Here, L0, M0, and S0 are physiological primary colors of the average value of the chromaticity values obtained by Expression (8), and can be regarded as the white color of the imaging light source when the image data S is obtained. If the image data S is obtained by reading an image recorded on a film or the like, the white color is obtained by combining both the light source of the photographing light source and the light source of the scanner. Ln, Mn, and Sn are the white chromaticities of the imaging light source to be referred to when acquiring the image data S. When this value is unknown, the color of the white point corresponding to the daylight of D50 to D60 is used. Degrees (un, vn) can be used. a is a value indicating the adaptation state, and usually 1 is used.
[0059]
For example, when viewing an original scene under a tungsten bulb or in the case of an evening scene, it may be preferable not to completely correct the color of the light source but to reproduce while retaining the color of the light source. . In this case, the value of a is preferably adjusted between 0 and 1. For example, in consideration of the distance between (u0, v0) and (un, vn) on the chromaticity diagram as shown in FIG. 2, instead of completely setting the value of a to 1, the following equation (12) is used. , (13), it is preferable to change the value of a so as to decrease as the distance between (u0, v0) and (un, vn) increases, so as to leave the color of the light source at the time of shooting.
= √ ((u0-un)²+ (V0-vn)²) (12)
a = 1.0−k × dist (13)
Where k is a constant for setting the value of a to be 0 or more and 1 or less.
Further, the value of a can be changed depending on the value of (u0, v0). For example, when it is desired to perform color correction completely, such as when using a tungsten light source or in a scene other than a sunset scene, when the value of u0 is equal to or more than a predetermined threshold value (for example, 0.2153), the equations (12) and (13) are used. Using the obtained value of a, it is also possible to set a = 1 when the value is less than a predetermined threshold. Thereby, a more preferable reproduced image can be obtained.
[0061]
In the RGB converter 9, the converted physiological primary colors L ', M', S 'are converted into processed tristimulus values R', G ', B' by the following equation (14).
[0062]
R 'L'
G '= | A |^-1｜ B ｜^-1・ M '(14)
B 'S'
The output unit 10 converts the processed tristimulus values R ', G', B 'into tristimulus values R ", G", B "depending on the reproducing means 4 using a matrix or a lookup table. At this time, gradation processing may be performed on the processed tristimulus values R ′, G ′, and B ′, and processing such as changing the color and density of skin color, green, and sky blue may be performed. .
[0063]
The tristimulus values R ", G", B "(image data S") thus obtained are inputted to the reproducing means 4, and the image represented by the image data S "is reproduced as a hard copy or a soft copy. You.
[0064]
As described above, in the first embodiment, the physiological primary colors L0, M0, and S0 obtained by the equation (8) are regarded as the white point of the imaging light source when the image data S is obtained, and the equations (10) and (10) are used. According to (11), the physiological primary colors L0, M0, and S0 are converted so that the physiological primary colors L, M, and S of the respective pixels match the white chromaticities Ln, Mn, and Sn of the reference imaging light source. The image obtained by reproducing the obtained image data S ″ can be made substantially similar to the impression of how colors look when photographed under the reference photographing light source. Here, the image data S The appearance of the image differs depending on the imaging light source when the image is captured.For example, when the image is captured on a cloudy day or in a shade, the white in the image looks bluish as a whole, Even with a blue tint, human color adaptation Therefore, the physiological primary colors L0, M0, and S0 are converted so that the physiological primary colors L, M, and S of each pixel match the chromaticity values Ln, Mn, and Sn of the white point of the reference imaging light source. By doing so, the white color represented by the image data S ″ is converted into the same white color as that obtained by the reference imaging light source. As a result, by reproducing the image data S ″, it is possible to reproduce the same image as the impression of how the image looks when the image data S is acquired.
[0065]
In addition, the value of a in Expression (11) is changed so as to become smaller as the distance between (u0, v0) and (un, vn) becomes larger by Expressions (12) and (13). The color of the light source at the time can be left, so that an image faithful to the appearance of the original scene can be reproduced.
[0066]
Next, a second embodiment of the present invention will be described. FIG. 3 is a schematic diagram illustrating a configuration of an image processing system to which an image processing method and apparatus according to a second embodiment of the present invention are applied. Note that, in the image processing system according to the second embodiment, the same components as those of the image processing system according to the first embodiment are denoted by the same reference numerals, and detailed description is omitted. In the second embodiment, the tristimulus values R, G, and B are replaced by the CIE1976uv chromaticity instead of the color conversion unit 5, the physiological primary color conversion unit 7, the chromatic adaptation conversion unit 8, and the RGB conversion unit 9 in the first embodiment. A color conversion unit 35 for converting values into values, and a tristimulus value conversion for converting the average values u0, v0 calculated by the average value calculation unit 6 into tristimulus values R0, G0, B0 instead of the physiological primary colors L0, M0, S0. And a chromatic adaptation conversion unit 38 that performs chromatic adaptation conversion of the tristimulus values R, G, and B based on the tristimulus values R0, G0, and B0 obtained by the tristimulus value conversion unit 37. .
[0067]
Next, the operation of the second embodiment will be described.
[0068]
First, the image data S is read from the recording medium 1 by the reading unit 2 and input to the image processing unit 3. In the image processing means 3, first, in the color conversion section 35, the tristimulus values R, G, and B of each pixel when the image data S is reproduced as an image are converted into CIE1976uv chromaticity values by the above equations (1) and (2). Convert.
[0069]
The average value calculation unit 6 calculates the average values u0 and v0 of the chromaticity values u and v obtained by the above equation (2) using the above equation (5).
[0070]
The tristimulus value converter 37 converts the average values u0, v0 calculated by the average value calculator 6 into tristimulus values R0, G0, B0 according to the above equations (6), (7) and the following equation (15). I do.
[0071]
R0 X0
G0 = | A |^-1・ Y0 (15)
B0 Z0
Where the matrix | A |^-1Is the inverse of matrix A shown in equation (3) above.
[0072]
The chromatic adaptation conversion unit 38 performs chromatic adaptation conversion of the tristimulus values R, G, and B input to the image processing means 3 using equations (16) and (17), and converts the converted tristimulus values R ′, G ′, and B. '.
[0073]
R 'R
G ′ = | D | · G (16)
B 'B
However,
(a · Rn + (1-a) · R0) / R0
| D | = (aGn + (1-a) G0) / G0 (17)
(a · Bn + (1-a) · B0) / B0
Here, R0, G0, and B0 are tristimulus values of the average value of the chromaticity values obtained by Expression (15), and can be regarded as the white color of the imaging light source when the image data S is obtained. Further, Rn, Gn, and Bn are the chromaticities of the white of the photographing light source to be referred to when acquiring the image data S. When this value is unknown, the color of the white point corresponding to the daylight of D50 to D60 is obtained. Degrees (un, vn) can be used. a is a value indicating the adaptation state, and usually 1 is used. As in the first embodiment, the value of a may be changed by the equations (12) and (13) in consideration of the distance between (u0, v0) and (un, vn). The value of a may be changed depending on the value of (u0, v0).
[0074]
The output unit 10 converts the converted tristimulus values R ', G', B 'into tristimulus values R ", G", B "depending on the reproducing means 4 using a matrix or a lookup table.
[0075]
The tristimulus values R ", G", B "(image data S") thus obtained are inputted to the reproducing means 4, and the image represented by the image data S "is reproduced as a hard copy or a soft copy. You.
[0076]
As described above, in the second embodiment, the tristimulus values R0, G0, and B0 obtained by Expression (15) are regarded as the white point of the imaging light source when the image data S is obtained, and Expressions (16) and Equation (17) converts the tristimulus values R, G, and B of each pixel so that the tristimulus values R0, G0, and B0 match the white chromaticities Rn, Gn, and Bn of the reference imaging light source. As in the first embodiment, an image obtained by reproducing the finally obtained image data S ″ is substantially similar to the impression of how colors look when photographed under a reference photographing light source. Further, as in the first embodiment, the tristimulus values R, G, B of each pixel are converted into physiological primary colors L, M, S, and the physiological primary colors L ', M', Since it is not necessary to convert S ′ into the tristimulus values R ′, G ′, B ′, the ratio is different from that of the first embodiment. To be shortened operation time, also can be simplified the structure of the device since it is not necessary to provide the RGB converter 9.
[0077]
Here, when the chromatic adaptation conversion is performed in the space of the physiological primary colors LMS as in the first embodiment, and when the chromatic adaptation conversion is performed in the space of the tristimulus values RGB as in the second embodiment, both of them are used. Differences appear in chromatic colors. That is, while the white point matches the case of the physiological primary color LMS and the case of the tristimulus value RGB, the chromatic color depends on the saturation of the target color and the distance between the white point of the imaging light source and the white point of the reference light source. Will make a difference. In this case, the chromatic adaptation conversion can be performed so that the physiological primary color space that takes into account the perception of the visual cells of the human eye is closer to the real impression, but especially in the case of natural images. In the case of an image in which there is not much high-saturation color, the difference between the two is small, so that there is not much problem visually.
[0078]
In the first and second embodiments, the average (u0, v0) of the chromaticity values is obtained by Expression (5). However, the present invention is not limited to this. Is imaged at the center of the image, the center of the image may be selected and the average value (u0, v0) may be obtained. v0) may be obtained. In this case, the weighting is changed according to the position of the image. However, the weighting may be performed as follows according to the color of each pixel.
[0079]
That is, in the white balance adjustment of the photographing light source, for example, when a green portion is included in the image, whether the green portion is caused by green of a tree (an actually existing object) or the photographing light source is a fluorescent lamp For this reason, it is difficult to determine whether the white portion is caused by the appearance of green. Also, the blue part is the object color of the blue sky or the result of shooting in the shade, or the red part is the object color of red clothes or the like, or it is obtained by shooting under tungsten light. It is also difficult to determine the same. Here, the chromaticity value of the photographing light source is generally located on the CIE daylight locus, plankian locus or blackbody locus as shown in FIG. 4, and the object color is the daylight locus (or plankian). Locus, blackbody locus). For this reason, for the chromaticity value of a certain pixel, the distance from the chromaticity value to the daylight locus (or Planckian locus, black body locus) is considered, and the reciprocal of this distance is used as a weight and the average value (u0, v0) Can be asked. By calculating the average value (u0, v0) in this way, it is possible to reduce the weighting of a point distant from the daylight locus, that is, a pixel having a chromaticity value that is considered to have a high possibility of the object color, and , Average value (u0, v0) can be obtained as a value that strongly reflects the tint of the imaging light source. Specifically, assuming that the daylight locus of CIE is represented by two straight lines, the distance d from a certain pixel to the daylight locus can be obtained by the following equation (18).
[0080]
When u ≧ 0.2153
d = abs (v−0.6895 × u−0.3478) /1.2147
When u <0.2153
d = abs (v−1.6733 × u−0.1372) /1.9442 (18)
The weight w for each pixel can be represented by the following equation (19).
[0081]
w = 1.0−k ′ × d (19)
From Expressions (18) and (19), the weighted average value (u0, v0) can be obtained by Expression (20) below.
[0082]
u0 = (Σu · w) / (Σw)
v0 = (Σv · w) / (Σw) (20)
In particular, when it is desired not to be affected by the blue sky, the weight at the time of u <0.193 is set to 0. In order to prevent the influence of the vivid red, the weight is set to u> 0.235. The average value (u0, v0) may be obtained by changing the weight according to the value of u, such as setting the weight at that time to 0.
[0083]
In the first and second embodiments, only one point value is used as the chromaticity of white of the imaging light source to be referred to when acquiring the image data S. It may be used as a value having a certain range based on points. This range can be set, for example, to a specific range on the daylight locus shown in FIG. That is, assuming that the daylight locus is represented by two straight lines, the perpendicular and the daylight locus when the perpendicular is drawn from the chromaticity average value (u0, v0) to the daylight locus as shown in FIG. Can be obtained by the following equation (21).
[0084]
u0 ≧ 0.2153 = (0.6895 × v0 + u0−0.6895 × 0.3478) / (0.6895 × 0.6895 + 1.0) = (0.6895 × 0.6895 × v0 + 0.6895 × u0 + 0.3478) / (0.6895 × 0.6895 + 1.0)
u0 <0.2153 = (1.6673 × v0 + u0−1.673 × 0.1372) / (1.6673 × 1.6673 + 1.0) = (1.6673 × 1.6673 × v0 + 1.6773 × u0 + 0.1372) / (1.6673 × 1.6673 + 1.0)
(21)
Then, the white color of the reference imaging light source can be determined depending on ug. That is, when ug> 0.2028, the chromaticity value (0.2028, 0.4781) at a color temperature of 5700 K on the daylight locus is set as the white point of the reference imaging light source, and the physiological primary colors Ln, Mn, Sn or the tristimulus values Rn, Gn, Bn Convert to
[0085]
When ug <0.2007, the chromaticity value (0.2007, 0.4743) at a color temperature of 6000 K on the daylight locus is set as the white point of the reference photographing light source and converted into the physiological primary colors Ln, Mn, Sn or the tristimulus values Rn, Gn, Bn. I do.
[0086]
.2007 ≦ ug ≦ 0.2028, G (ug, uv) is set as the white point of the reference imaging light source, and is converted into physiological primary colors Ln, Mn, Sn or tristimulus values Rn, Gn, Bn.
[0087]
Here, when the white point of the reference photographing light source is one point, the physiological primary colors L, M, S or the tristimulus values R, G, B are subjected to chromatic adaptation conversion as shown in FIG. When the white point of the reference light source is within a certain range, the physiological primary colors L, M, S or the tristimulus values R, G, B are subjected to chromatic adaptation conversion as shown in FIG. . Therefore, it is possible to prevent the physiological primary colors L, M, S or the tristimulus values R, G, B from being excessively corrected at the time of the chromatic adaptation conversion, and to reproduce an image having a natural appearance. Further, by obtaining the white point of the reference photographing light source based on the white point of the photographing light source, it is possible to perform chromatic adaptation conversion reflecting the color of the photographing light source.
[0088]
In the above embodiment, the image processing method and apparatus according to the present invention have been described as hardware, but a program for executing the processing of the image processing method according to the present invention is recorded on a recording medium and provided. Is also good.
[0089]
Next, an embodiment of an image reproducing apparatus according to the present invention will be described.
[0090]
FIG. 7 is a diagram showing a configuration of an image reproducing device including the image processing device according to the present invention. As shown in FIG. 7, an image reproducing apparatus including an image processing apparatus according to the present invention includes a reading unit 11 for reading image data S from a recording medium on which image data S obtained by a digital camera or the like is recorded, A color correction unit (AWB) 12 including a processing device, a density correction unit 13 for correcting the density of the image data S subjected to color correction in the color correction unit 12, and a An image processing unit 15 that performs image processing such as gradation processing with reference to the LUT 14 and a reproduction unit 16 that reproduces image data S ′ obtained by performing the image processing. Note that the image reproducing device in the present embodiment may be a printer or a CRT monitor.
[0091]
Next, the operation of the present embodiment will be described. The image data S read by the reading unit 11 is subjected to image processing in the color correction unit 12 in the same manner as in the above-described embodiment, and further, the density is corrected by the density correction unit 13 and input to the image processing unit 15. . The image processing unit 15 performs image processing such as gradation conversion on the image data whose density has been corrected with reference to the LUT 14, and reproduces image data S ′ obtained by performing the image processing. The data is reproduced as a hard copy or a soft copy in the unit 16.
[0092]
As described above, since the image reproducing apparatus according to the present embodiment includes the above-described image processing apparatus according to the present invention, the input image data S is converted so as to reproduce the impression of the appearance at the time of shooting. Will be used for reproduction. Therefore, it is possible to reproduce an image faithful to the appearance at the time of shooting simply by inputting the image data S.
[0093]
Next, an embodiment of an imaging device according to the present invention will be described.
[0094]
FIG. 8 is a block diagram showing a configuration of a digital camera which is an embodiment of an imaging device including the image processing device of the present invention. As shown in FIG. 8, the digital camera according to the present embodiment includes an imaging unit 21 such as a lens and an aperture, a CCD 22 that photoelectrically converts a captured subject image to obtain image data S representing the subject image, and an image data S A gain adjustment circuit 23 for adjusting the gain of the image processing apparatus, a color correction unit (AWB) 24 including the image processing apparatus according to the present invention, and video processing for performing video processing on the image data S including processing by the color correction unit 24. Circuit 25, a gamma conversion unit 26 for performing gamma conversion on the image data processed by the video processing circuit 25, and image data S 'obtained by performing the gamma conversion on a recording medium or the like. And a signal recording circuit 27.
[0095]
Next, the operation of the present embodiment will be described. The subject image picked up by the image pickup unit 21 is photoelectrically converted into image data S by the CCD 22. The gain of the image data S is adjusted by the gain adjustment circuit 23, and the image processing is performed by the color correction unit 24 and the video processing circuit 25 as described above. The image data that has been subjected to the image processing is subjected to gamma conversion in a gamma conversion unit 26, and the image data S ′ obtained by performing the gamma conversion is recorded in a signal recording circuit 27 on a recording medium.
[0096]
As described above, since the imaging device according to the present embodiment includes the image processing device according to the present invention, the image data S acquired by imaging is processed image data S that can reproduce the impression of the appearance at the time of imaging. 'Is recorded on the recording medium. Therefore, it is possible to obtain image data capable of reproducing an image faithful to the impression of the appearance at the time of imaging simply by imaging the subject.
[Brief description of the drawings]
FIG. 1 is a schematic diagram showing a configuration of an image processing system to which an image processing method and apparatus according to a first embodiment of the present invention are applied;
FIG. 2 is a diagram showing a chromaticity diagram;
FIG. 3 is a diagram showing a daylight locus and a planchian locus.
FIG. 4 is a schematic diagram showing a configuration of an image processing system to which an image processing method and apparatus according to a second embodiment of the present invention are applied;
FIG. 5 is a diagram for explaining a state of finding an intersection when a perpendicular is drawn to the daylight locus from the chromaticity average value.
FIG. 6 is a diagram illustrating a state of color adaptation.
FIG. 7 is a block diagram showing a configuration of an image reproducing apparatus according to an embodiment of the present invention.
FIG. 8 is a block diagram illustrating a configuration of an imaging device according to an embodiment of the present invention.
[Explanation of symbols]
1 Recording medium
2,11 readout unit
3 Image processing means
4 Reproduction means
5,35 color conversion unit
6 Average value calculation unit
7 Physiological primary color converter
8,38 color adaptation converter
9 RGB converter
10 Output section
12, 24 color correction unit
13 Density correction unit
14 LUT
15 Image processing unit
16 Reproduction unit
21 Imaging unit
22 CCD
23 Gain adjustment circuit
25 Video processing circuit
26 Gamma converter
27 Signal recording circuit
37 Tristimulus value converter

Claims (23)

Converting the RGB color of each pixel in the image represented by the image data obtained by the imaging into a chromaticity value,
Calculating a chromaticity average value that is an average value of the chromaticity values of the respective pixels;
Converting the chromaticity average value to tristimulus values corresponding to the physiological primary colors,
The tristimulus values corresponding to the physiological primary colors are regarded as the white color of the imaging light source when the image data is obtained, and the image data is converted so that the tristimulus values match the chromaticity values of the white point of the reference imaging light source. Image processing method for obtaining processed image data. 2. The image processing method according to claim 1, wherein the degree of the coincidence is reduced as the tristimulus value corresponding to the physiological primary color is different from the chromaticity value of the white point of the reference photographing light source . Converting the RGB color of each pixel in the image represented by the image data obtained by the imaging into a chromaticity value,
Calculating a chromaticity average value that is an average value of the chromaticity values of the respective pixels;
The chromaticity average value is converted to RGB tristimulus values,
The RGB tristimulus values are regarded as the white color of the imaging light source when the image data is acquired, and the image data is converted such that the RGB tristimulus values match the chromaticity value of the white point of the reference imaging light source. An image processing method characterized by obtaining processed image data by performing the following steps . 4. The image processing method according to claim 3, wherein the degree of the matching is reduced as the RGB tristimulus values are different from the chromaticity value of the white point of the reference photographing light source . The chromaticity average value is calculated as a weighted average value according to a reciprocal of a distance of the chromaticity value from a CIE daylight locus or a black body locus. image processing method. The white point of the reference imaging light source, and based on the white point of the imaging light source, an image processing method according to any one of claims 1 5, characterized in that the set to have a value in a predetermined range . The image processing method according to claim 1, further comprising converting the processed image data based on a reproduction target value of a reproduction device that reproduces the processed image data . Chromaticity value conversion means for converting the RGB color of each pixel in the image represented by the image data acquired by the imaging into a chromaticity value;
Chromaticity average value calculating means for calculating a chromaticity average value which is an average value of the chromaticity values of the respective pixels,
Tristimulus value conversion means for converting the chromaticity average value to a tristimulus value corresponding to the physiological primary color,
The tristimulus values corresponding to the physiological primary colors are regarded as the white color of the imaging light source when the image data is obtained, and the image data is converted so that the tristimulus values match the chromaticity values of the white point of the reference imaging light source. An image processing apparatus comprising: a chromatic adaptation conversion unit that obtains processed image data . 9. The apparatus according to claim 8, wherein the chromatic adaptation conversion unit reduces the degree of the coincidence as the tristimulus value corresponding to the physiological primary color is different from the chromaticity value of the white point of the reference imaging light source. The image processing apparatus according to claim 1. Chromaticity value conversion means for converting the RGB color of each pixel in the image represented by the image data acquired by the imaging into a chromaticity value;
Chromaticity average value calculating means for calculating a chromaticity average value which is an average value of the chromaticity values of the respective pixels,
Tristimulus value conversion means for converting the chromaticity average value into RGB tristimulus values;
The RGB tristimulus values are regarded as the white color of the imaging light source when the image data is acquired, and the image data is converted such that the RGB tristimulus values match the chromaticity value of the white point of the reference imaging light source. An image processing apparatus comprising: a color adaptation conversion unit that obtains processed image data by using a color adaptation conversion unit. The image according to claim 10, wherein the chromatic adaptation conversion unit is a unit that reduces the degree of coincidence as the RGB tristimulus values are different from the chromaticity value of the white point of the reference imaging light source. Processing equipment. The chromaticity average value calculating means is means for calculating the chromaticity average value as a weighted average value according to a reciprocal of a distance of the chromaticity value from a CIE daylight locus or a blackbody locus. The image processing apparatus according to any one of claims 8 to 11, wherein: 13. The apparatus according to claim 8, further comprising: a unit configured to set a white point of the reference photographing light source to have a value within a predetermined range based on the white point of the photographing light source. the image processing apparatus. 14. The apparatus according to claim 8, further comprising an output conversion unit configured to convert the processed image data based on a reproduction target value of a reproduction apparatus that reproduces the processed image data. Image processing device. Converting RGB color of each pixel into a chromaticity value in an image represented by image data obtained by imaging;
Calculating a chromaticity average value that is an average value of the chromaticity values of the respective pixels;
Converting the chromaticity average value to a tristimulus value corresponding to the physiological primary color,
The tristimulus values corresponding to the physiological primary colors are regarded as the white color of the imaging light source when the image data is obtained, and the image data is converted so that the tristimulus values match the chromaticity values of the white point of the reference imaging light source. And a program for causing a computer to execute a procedure for obtaining processed image data by performing a computer-readable recording medium . The step of obtaining the processed image data is a step of reducing the degree of the coincidence as the tristimulus value corresponding to the physiological primary color is different from the chromaticity value of the white point of the reference imaging light source. The computer-readable recording medium according to claim 15 . Converting RGB color of each pixel into a chromaticity value in an image represented by image data obtained by imaging ;
Calculating a chromaticity average value that is an average value of the chromaticity values of the respective pixels;
Converting the chromaticity average into RGB tristimulus values;
The RGB tristimulus values are regarded as the white color of the imaging light source when the image data is obtained , and the image data is converted and processed so that the tristimulus values match the chromaticity values of the white point of the reference imaging light source. A computer-readable recording medium on which a program for causing a computer to execute a procedure for obtaining used image data is recorded. 18. The method according to claim 17, wherein the step of obtaining the processed image data is a step of reducing the degree of coincidence as the RGB tristimulus values are different from the chromaticity value of the white point of the reference imaging light source. The recording medium according to the above. The step of calculating the chromaticity average value is a step of calculating the chromaticity average value as a weighted average value according to a reciprocal of a distance of the chromaticity value from a CIE daylight locus or a black body locus. A computer-readable recording medium according to any one of claims 15 to 18, characterized in that: The method according to any one of claims 15 to 19, further comprising a step of setting a white point of the reference photographing light source to have a value within a predetermined range based on the white point of the photographing light source. Computer readable recording medium. 21. The program according to claim 15, further comprising a step of converting the processed image data based on a reproduction target value of a reproduction device that reproduces the processed image data. Computer-readable recording medium. Imaging means for capturing a subject image to obtain image data representing the subject image;
The image processing apparatus according to claim 8, wherein image processing is performed on the image data.
Recording means for recording, on a recording medium, processed image data obtained by performing processing by the image processing apparatus . An image reproducing device that reproduces image data as a visible image,
Input means for inputting the image data,
The image processing apparatus according to claim 8, wherein image processing is performed on the image data input from the input unit.
An image reproducing apparatus comprising: reproducing means for reproducing processed image data obtained by performing image processing by the image processing apparatus .

Images