JPH0549049A - Color chart recognition device, gwa discrimination device and light source estimate device - Google Patents

Abstract

PURPOSE:To provide the color chart recognition device, the GWA discrimination device and the light source estimate device using the recognition device while solving a problem that the estimate of the light source does not follow a color charm often used for evaluation of a commodity or the like when the estimate of a light source is made accurate. CONSTITUTION:A picture signal sampled by a sampling means 1 in a color chart recognition device is converted into a chromaticity signal by a chromaticity conversion means 2, a projection means 3 projects the chromaticity signal in a direction perpendicular to a straight line defined on a chromaticity plane and converts the signal into a displacement of the component in the direction. Then a variance calculation means 4 calculates the variance of the displacement, a 3-dimension moment calculation means 5 calculates a 3-dimension moment of the displacement. Finally, threshold level processing means 61, 62 and an AND circuit 7 are used to implement prescribed processing to realize the color chart recognition device able to recognize a color chart. The excellent GWA discrimination device and the light source estimate device with high accuracy are realized by using the color chart recognition device.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color chart recognizing device for controlling a white balance adjusting device for reproducing a correct color tone in a video camera and the like, and a G chart recognizing device.
The present invention relates to a WA determination device and a light source estimation device.

[0002]

2. Description of the Related Art In recent years, with the spread of video cameras, means for following photographing conditions have become more and more important in order to obtain beautiful colors. Among them, it is very important for color reproduction to accurately estimate the light source at the time of shooting.

Conventionally, as means for estimating the photographing light source, there are known means for estimating using a plurality of photosensitive elements having different spectral characteristics and means for estimating through image processing. In general, the light source estimation device has been shifting from a means using a photosensitive element, which is restricted in terms of design, cost, etc., to a means for processing and estimating an image signal (hereinafter, referred to as AWB).

The basis of the method of estimating the light source based on image processing is that the color of the subject in the screen is achromatic when mixed and averaged (Gray World A).
(Sumulation: GWA). As a method for performing color adjustment based on this assumption, Japanese Patent Laid-Open No. 56-36291 is available.
The one described in Japanese Patent Publication is known.

The present inventor carried out the processing under the assumption that the simple arithmetic mean of the colors of the entire screen is always achromatic color, which the method based on the conventional image processing described above has. Therefore, when the above assumption is not satisfied, such as when the area occupied by a specific color on the screen is large, a method has been proposed that solves the drawback that the light source is often mistaken and the color balance is lost. That is, by an image determination device that determines a GWA, which includes a variance calculation unit that calculates the variance of the screen color distribution and a threshold processing unit that performs threshold processing depending on whether the output from the variance calculation unit is greater than a set value, It was to control the adoption / non-adoption of the light source estimation result.

[0006]

However, in the above configuration, the average of the color charts used for product evaluation is achromatic because the color chart has a very large dispersion in the color space. Also has a problem that it is determined that the GWA is not established and the light source estimation does not follow.

The present invention solves the above problems and provides a color chart recognition device for distinguishing an image of a daily life from a color chart, an image determination device using the same, and a light source estimation device.

[0008]

In order to achieve this object, a color chart recognition apparatus includes a sampling means for converting an image signal IS from a CCD or the like according to an appropriate sampling rule, and an image sampled from the sampling means. A chromaticity conversion means for converting a signal into a chromaticity signal, and a chromaticity signal perpendicular to a straight line approximating the trajectory drawn on the chromaticity plane when the color of the light source from the black body radiation is the temperature of the black body as a parameter Projection means for projecting in any direction (V1, V2) and converting to component magnitude in that direction, dispersion calculation means for calculating the image signal variance SGM of the displacement obtained from the sampling data, and displacement received from the projection means AND: third-order moment calculating means for obtaining a third-order moment, threshold processing means for receiving a signal, comparing it with a threshold stored internally, and outputting a determination signal, and It has a structure consisting of road. G
The WA determination device has a configuration in which a threshold value processing unit having a threshold value set so as to be suitable for GWA determination and an OR circuit are added to the above configuration. The light source estimation device further receives a light source estimation control signal from the GWA determination unit that estimates the light source from the image data sampled from the sampling unit, a light source estimation control signal from the GWA determination unit, and a light source estimation result from the light source estimation unit. And a light source estimation result holding means for receiving the comprehensive determination result from the comprehensive determination unit and outputting the internally held light source estimation result when the determination is impossible. have.

[0009]

With this configuration, the color chart can be recognized by determining the color chart when the variance is large and the third moment is small. And
By recognizing the color chart together with the GWA determination, a device in which the light source estimation follows not only the natural features but also the color chart is realized.

[0010]

(Embodiment 1) An embodiment of the present invention will be described below with reference to the drawings.

First, the principle of the present invention will be described. AW
The essence of what enables B is the phenomenon that color mixture reduces the saturation of colors. Then, when thinking about this phenomenon, if a sufficient number of colors are mixed, the color becomes an achromatic color, that is, gray (white).

Based on this, the AWB performs processing under the following expectations. If the subject entering the screen contains enough colors, then taking the average of all the subjects' colors should be achromatic.
If the color of the subject is achromatic, the color of the screen reflects the light source color as it is. That is, the light source can be estimated by looking at the average color of the entire screen. Therefore, if the color reproduction system is adjusted so that the average color of the entire screen is achromatic, the color rendering of the light source can be corrected and the white balance can be adjusted.

Here, the point to be noted is that
Usually, it's not really emphasized, but as implied by "contains enough colors ...",
AWB is a statistical method (for one screen). The essence of AWB being a statistical method is that
Concerning the limits of WB, it is summarized in the well-discussed question, "Can you distinguish a white wall illuminated by a red light source from a red wall illuminated by a white light source?" As is well known, in order for statistical methods to be applicable, the subject must be sufficient to allow statistical discussion. Since the above settings deal with the case where the statistical method cannot be applied to the object, the light source cannot be properly estimated by the AWB. To be clear, the above question is a discussion that goes beyond the limits of AWB, and is meaningless.

When considering AWB, rather than discussing such statistically meaningless conditions, A
It is important to understand the application limits of WB. And
In implementing the AWB, it is most important to keep the light source estimation within the applicable limit in order to secure the estimation accuracy.

The present inventor considered the above and made the following consideration based on the data analysis.

Since chromaticity is two-dimensional, there are two dispersion axes that describe the distribution of data. Mathematically speaking, the axis doesn't matter. However, in the problem of white balance, in the chromaticity plane, the locus (light source curve) drawn by the color of the light emitted by the blackbody radiation according to the temperature of the blackbody, as shown in FIG. Directional equivalence does not hold because it has meaning. Now, a straight line that approximates the light source curve is named a light source line.

In fact, when looking at the chromaticity data of various images, there is a remarkable difference between the dispersion in the light source line direction (horizontal dispersion) and the dispersion in the direction perpendicular to the light source line (vertical dispersion). The dispersion parallel to the light source line is generally larger than the vertical dispersion, and the size of the dispersion changes greatly depending on whether or not the sky enters the screen in outdoor shooting. On the other hand, such a phenomenon is not observed in vertical dispersion. Therefore, for various subjects,
Vertical dispersion is stable and meaningful. Hereinafter, when the horizontal dispersion and the vertical dispersion are not distinguished, they are simply referred to as dispersion, and when specified, they are referred to as horizontal dispersion or vertical dispersion.

In the AWB, the color distribution of the subject entering the screen is a statistically natural distribution (for example, a normal distribution centered on achromatic color). At least, no matter what subject, the color distribution does not change. , Implicitly in the expression "many colors". This assumption is established with considerable accuracy when shooting a distant view outdoors. However, it is clear that the assumption that the color distribution is invariable is not always satisfied, considering a bust shot of a person wearing a highly saturated clothes against a distant view outdoors.

For a subject whose AWB assumption is broken, such as a bust shot of a person wearing such highly saturated clothes, the average color of the entire screen obtained by simple addition is strongly drawn to the color of the clothes. Will end up. This is a cause of incorrect estimation of the light source.

On the other hand, in an image such as a bust shot, the saturation of some of the sampling data is relatively high, and the sampling data in the hue plane is widely distributed. It appears as an increase in the variance of the sampling data. This phenomenon is conspicuous in the block average data obtained by dividing the screen into several blocks and performing the averaging operation in each of them.

When the variance is large as described above, the GWA is often not established, and the variance is effective as an index regarding an error. However, if you try to eliminate the influence of the sky,
It is better to use vertical dispersion for GWA determination. Considering this point, the explanation will focus on vertical dispersion, but this does not mean that horizontal dispersion is invalid.

As is clear from the above discussion, it is possible to reduce the error rate of light source estimation by performing the following processing using vertical dispersion. That is, when the vertical dispersion, which is one of the indexes indicating the application limit of AWB, becomes larger than a certain set value, the GW is processed in the image to be processed.
It is expected that A will not hold. In this case, by not performing light source estimation and holding the current light source estimation result,
Suppress the occurrence of errors.

It may be argued that even if the variance is large, the average color of the entire screen becomes achromatic and the light source can be correctly estimated in some cases.
However, in such a case, when colors that occupy a large part of the screen are in a complementary color relationship with each other, it happens by chance in a statistical sense, and white balance processing based on GWA is performed. This is something that was not originally supposed to be. Of course, many such subjects do not exist, but a color chart that is often used in evaluation tests and product tests at the time of shipment is a typical example corresponding to this. It is desirable that the color chart can be recognized separately from other images for the purpose of use.

A detailed analysis of the color chart shows that the vertical distribution is very large, but the data distribution is evenly distributed around the mean value. For normal subjects, the data distribution in the color space is distorted when the variance is extremely large, and the balanced chromaticity distribution of the color chart is extremely characteristic, reflecting the artificial image. Target. The third-order moment is known as a statistic for extracting the distorted distribution. The third moment M3 is defined by M3 = Σ (x− <x>) ³ / N (where x is the value of data, <x> is the average value of x, and N is the number of data). According to the analysis by the present inventor, the dispersion of the color chart is almost twice or more the value of the dispersion threshold value that is determined to be the GWA being broken, and the magnitude of the third-order moment absolute value is determined to be OK in the GWA. The size is considerably smaller than that of the image, and compared with the third moment of the image having large variance, the magnitude is almost 10 times or more different. Therefore, the color chart can be recognized with extremely high accuracy by comprehensively determining the variance value of the chromaticity data and the third moment value, especially the absolute value of the third moment.

Although the same processing can be performed without using the absolute value of the third moment, the processing is simpler when the absolute value is used. Below, in order to simplify the explanation,
The absolute value of the second moment will be simply referred to as the third moment for explanation.

The above is the principle of the present invention. Next, the configuration of the first embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing the structure of the color chart recognition section. In FIG. 1, 1 is a sampling means for converting an image signal IS from a CCD or the like according to an appropriate sampling rule, and 2 is a chromaticity conversion means for converting the image signal sampled by the sampling means 1 into a chromaticity signal. Yes, 3 is a projection means for projecting the chromaticity signal in the direction (V1, V2) perpendicular to the light source line described in the explanation of the principle, and converting the component magnitude in that direction, that is, the displacement from the light source line, Reference numeral 4 is a dispersion calculating means for calculating the image signal dispersion SGM of the displacement obtained from the sampling data, and 5 is a third moment calculating means for receiving the displacement Xi from the projecting means 3 to obtain a third moment M3. Reference numerals 61 and 62 are threshold value processing means for receiving a signal and comparing it with a threshold value stored inside to output a determination signal. Reference numeral 7 is an ordinary AND circuit. Reference numeral 0 is a color chart determination unit which is the core of the present invention. Regarding the operation of the color chart determination device configured as described above,
This will be described with reference to FIG.

The input image signal IS will be described as (Y, RY, BY), but this does not restrict the present invention. ..

First, the image signal IS from the CCD or the like enters the sampling means 1 and is converted according to an appropriate sampling rule. As an example, as shown in FIG. 3, the screen is divided into m × n blocks, and a plurality of (k) data points are extracted from each block and averaged as shown in FIG. IS> i = Σ (Yj, R-Yj, B-Yj) / k (where i = 1, ..., N, N (= m × n) is the sampling number, and k is sampled in each block. It is conceivable to obtain the number of data). The chromaticity calculation means 2 is
Image signal sampled from the sampling means 1 <
IS> i = (YYi, BYi, RYi) is received, and C1i = BYi / YYi C2i = RYi / YYi is converted into a chromaticity signal CSi = (C1i, C2i).
The projection means 3 receives the chromaticity signal CSi and projects it in the direction (V1, V2) perpendicular to the light source line described in the description of the principle,
The component magnitude in that direction, that is, the displacement from the light source line, is converted into Xi, Xi = C1i · V1 + C2i · V2. The variance calculation means 4 calculates the displacement Xi obtained from the sampling data for one screen, i = 1, ..., N,
The image signal variance SGM of is calculated. The third moment calculating means 5 also receives the displacement Xi from the projecting means 3 and receives the third moment M3. Ask for. The threshold processing means 61 receives the image signal variance SGM from the variance calculating means 4 and compares it with the image variance threshold θTC stored inside. When SGM <θTC, TSGM = 0: non-color chart Outputs a color chart variance determination signal TSGM, with TSGM = 1: possible color chart. The threshold value processing means 62 receives the third moment M3 from the third moment calculating means 5 and stores the third moment M3 therein.
Compared with the next moment threshold value θM3, when M3 <θM3, TM3 = 1: There is a possibility of color chart. In other cases, TM3 = 0: Non-color chart, the third moment judgment signal TM3 of color chart is output. .. The AND circuit 7 receives the color chart variance determination signal TSGM from the threshold processing means 61 and the color chart third moment determination signal TM3 from the threshold processing means 62,
The logical product is obtained and the color chart determination signal T: T = 1: color chart T = 0: not color chart is output.

In this way, the color chart is determined from the image signal. The vector (V1, V2) that gives a direction perpendicular to the upper light source line may be approximately (V1, V2) to (1, 1), and need not be strictly determined. Depending on the parameters, the accuracy of the AWB hypothesis judgment will vary somewhat,
The impact is not significant as a result of data analysis.

The essence of the calculation of the vertical dispersion SGM is that the image signal is (RGB) or (Y, RY, BY).
It goes without saying that it does not depend on whether or not

Further, in the above calculation, the image signal is dropped into the two-dimensional chromaticity space and then the vertical dispersion is calculated. The image signal can also be used directly for calculation.

In addition, when a three-dimensional image signal is directly used, an algorithm can be designed so as to remove the effect of a specific color such as the sky, which has a great influence on the light source estimation described in the principle. In this case, the dispersion in the light source line direction can also have a sufficient meaning.
It is not necessary to specify the statistic to be determined as the vertical variance. The present invention embodies the processing based on the principle obtained from the analysis result for the image described above, and its configuration is not limited to the one described here and can take various configurations. In particular, there are many methods for setting the determination condition. As one of them, there is a method of setting CC = α · SGM−β · | M3 | 0 <α << β and setting CC as a color chart if CC is equal to or less than a certain value.

In addition, two projection means 3 may be provided to change the axis of the signal input to the variance calculation means 4 and the axis of the signal input to the third moment calculation means 5.

Since these matters are the same in the following embodiments, they will not be repeated. (Embodiment 2) Next, the configuration of the second embodiment of the present invention will be described.

In Example 1, in order to recognize the color chart, the third moment in the direction perpendicular to the light source line was calculated. However, there is a simpler method for recognizing the color chart by using the statistics in the axial direction. It is a method that positively utilizes the property that the entire average of the color chart screen is achromatic.

The effect on the color of the light source is mainly to move the color along the light source line, and there is almost no effect on the direction perpendicular thereto. Therefore, the average <X> of the color signals projected in the straight direction on the light source line is maintained at almost 0. As a result, the third moment M3 is Can be approximated by

In the second embodiment of the present invention, instead of the third moment, the vertical third-order sum M3 ' Is used to recognize a color chart. In the configuration, in the first embodiment, the direction parameter of the projection means may be set so as to be perpendicular to the light source line, and the third-order moment calculation means 5 may be changed to one for calculating the vertical cubic sum.

It is also possible to provide two projecting means 3 and change the axis of the signal input to the variance calculating means 4 and the axis of the signal input to the vertical cubic sum calculating means.

(Third Embodiment) A third embodiment of the present invention will be described below with reference to the drawings.

FIG. 5 is a block diagram showing the configuration of an image determination device incorporating a color chart determination device.

In FIG. 5, reference numeral 1 is a sampling means for converting an image signal IS from a CCD or the like according to an appropriate sampling rule, and 2 is a chromaticity for converting the image signal sampled by the sampling means 1 into a chromaticity signal. Reference numeral 3 is a conversion means, and 3 is a projection means for projecting the chromaticity signal in the direction (V1, V2) perpendicular to the light source line described in the description of the principle and converting the component magnitude in that direction, that is, the displacement from the light source line. 4 is a dispersion calculation means for calculating the image signal dispersion SGM of the displacement obtained from the sampling data, and 5 is 3 when the displacement Xi is received from the projection means 3.
Third-order moment calculating means for obtaining the second-order moment M3, and 61 and 62 are threshold value processing means for receiving a signal and comparing it with a threshold value stored inside to output a determination signal.
Is a normal AND circuit, and the above is the same as the configuration of FIG. 1. The threshold processing means 63 added here is 61,
The same threshold processing means as 62, 8 is an ordinary OR circuit, and 200 is an image determination unit.

The operation of the GWA determining device configured as above will be described with reference to FIG.

First, the image signal IS from the CCD or the like enters the sampling means 1 and is converted according to an appropriate sampling rule. The chromaticity calculation means 2 receives the image signals (YYi, BY) sampled by the sampling means 1.
i, RYi), and C1i = BYi / YYi C2i = RYi / YYi is converted into a chromaticity signal CSi = (C1i, C2i).
The projection means 3 receives the chromaticity signal CSi and projects it in the direction (V1, V2) perpendicular to the light source line described in the description of the principle,
The component magnitude in that direction, that is, the displacement from the light source line, is converted into Xi, Xi = C1i · V1 + C2i · V2. The variance calculation means 4 calculates the displacement Xi obtained from the sampling data for one screen, i = 1, ..., N,
The image signal variance SGM of is calculated. The third moment calculating means 5 also receives the displacement Xi from the projecting means 3 to obtain the third moment M3. The threshold processing means 61 receives the image signal variance SGM from the variance calculating means 4 and compares it with the image variance threshold θTC stored inside. When SGM <θTC, TSGM = 0: non-color chart Outputs a color chart variance determination signal TSGM, with TSGM = 1: possible color chart. The threshold value processing means 62 receives the third moment M3 from the third moment calculating means 5 and stores the third moment M3 therein.
Compared with the next moment threshold value θM3, when M3 <θM3, TM3 = 1: There is a possibility of color chart. In other cases, TM3 = 0: Non-color chart, the third moment judgment signal TM3 of color chart is output. .. The AND circuit 7 receives the color chart variance determination signal TSGM from the threshold processing means 61 and the color chart third moment determination signal TM3 from the threshold processing means 62,
The logical product is obtained and the color chart determination signal T: T = 1: color chart T = 0: not color chart is output. The threshold processing means 63 receives the image signal variance SGM from the variance calculating means 4 and compares it with the image variance threshold θSGM stored inside. When SGM <θSGM, J = 1: AWB hypothesis holds (O
K) In other cases, J = 0: AWB disagreement holds (N
O), the GWA determination signal J is output. The OR circuit 8 is
The GWA determination signal J is received from the threshold value processing means 63, the color chart determination signal T is received from the AND circuit 7, and a logical sum is calculated to obtain a light source estimation control signal C: C = 1: light source estimation result OK C = 0: light source estimation result. Output NG.

In this way, the image judging device including the color chart judging device is realized. (Embodiment 4) Hereinafter, a fourth embodiment of the present invention will be described with reference to the drawings.

FIG. 7 is a block diagram showing the configuration of the light source estimation device. In FIG. 7, 200 is the GWA determination unit described in the second embodiment, 9 is a light source estimation unit that estimates the light source from the image data sampled by the sampling unit 1, and 10 is a light source estimation control from the GWA determination unit 200. Reference numeral 11 denotes a comprehensive determination unit that receives a light source estimation result from the light source estimation unit 9 to make a comprehensive determination. Reference numeral 11 receives the overall determination result from the comprehensive determination unit, and if the determination cannot be made, internally. The light source estimation result holding means outputs the held light source estimation result.

The operation of the light source estimation device configured as described above will be described with reference to FIG. Comprehensive judgment section 1
0 receives the light source estimation result S1 from the light source estimation unit 9 and the light source estimation control signal C from the GWA determination unit 200, and when C = 1: OK S2 = S1 C = 0: NG S2 = (determination impossible ), The secondary light source estimation result S2 is output. The light source estimation result holding unit 11 receives the secondary light source estimation result S2 from the comprehensive determination unit 10, and when S2 ≠ (decision impossible) Sf = S2 Sbuf = S2 S2 = (decision impossible) Sf = Sbuf Sbuf = As Sbuf, the secondary light source estimation result is held in Sbuf, and at the same time, the final light source estimation result Sf is output.

Finally, the process at the time of rising will be described. At the time of rising, if the signal from the GWA determination unit is C = 0 (NG), separate processing is required. As the processing at the time of rising, first, there is a method in which J = OK is set at least once after the rising until SGM <θ. Secondly, there is a method in which Sbuf of the light source estimation result holding unit is provided with a standard setting at the time of rising. This standard setting may be configured to depend on the light source estimation result S1. Thirdly, there is a method of holding the previous final estimation result in Sbuf by using an SRAM or the like as the means for holding Sbuf in the estimation result holding unit.

As described above, according to this embodiment, by determining the establishment of GWA based on the variance of image signals,
Since the light source determination when the GWA is not established is abandoned because the light source estimation is likely to be erroneous, a light source estimating apparatus with less error as a whole is realized.

The vector (V1, V2) that gives a direction perpendicular to the above light source line may be approximately (V1, V2) to (1, 1), and need not be strictly determined. Depending on the parameters, the accuracy of the AWB hypothesis judgment will vary somewhat,
The impact is not significant as a result of data analysis.

The essence of the calculation of the vertical dispersion SGM is that the image signal is (RGB) or (Y, RY, BY).
It goes without saying that it does not depend on whether or not

Further, in the above calculation, the image signal is dropped into the two-dimensional chromaticity space and then the vertical dispersion is calculated. The image signal can also be used directly for calculation.

In addition, when a three-dimensional image signal is directly used, an algorithm can be constructed so as to remove the effect of a specific color such as the sky, which has a great influence on the light source estimation described in the principle. In this case, the dispersion in the light source line direction can also have a sufficient meaning.
It is not necessary to specify the statistic to be determined as the vertical variance. The present invention embodies the processing based on the principle obtained from the analysis result for the image described above, and its configuration is not limited to the one described here and can take various configurations. In particular, there are many methods for setting the determination condition.

[0054]

As described above, according to the present invention, the color chart recognition device has the sampling means for converting the image signal IS from the CCD or the like according to an appropriate sampling rule and the chromaticity of the image signal sampled by the sampling means. Chromaticity conversion means for converting into a signal, and a direction (V1, V2) perpendicular to the light source line described in the explanation of the principle of the chromaticity signal
Projection means for projecting to and converting to the component size in that direction,
Dispersion calculation means for calculating the image signal dispersion SGM of the displacement obtained from the sampling data, and displacement X from the projection means.
The third-order moment calculating means for receiving the third-order moment M3 in response to i, the threshold processing means for receiving the signal and comparing with the internally stored threshold value and outputting the determination signal, and the AND circuit, the chromaticity It is possible to realize an excellent color chart recognition device capable of recognizing a color chart by taking out the feature of the color chart in which the variance of 1 is large and the third moment is small. Further, by adding a threshold value processing unit having a threshold value set to be suitable for GWA determination and an OR circuit to the above configuration, an excellent GWA determination apparatus capable of color chart recognition can be realized. In addition, the GWA determination device receives a light source estimation control signal from the light source estimation unit that estimates the light source from the image data sampled from the sampling unit, the GWA determination unit, and the light source estimation result from the light source estimation unit to make a comprehensive determination. The GWA is provided by adding a comprehensive determination unit for performing and a light source estimation result holding unit that receives the comprehensive determination result from the comprehensive determination unit and outputs the internally held light source estimation result when the determination cannot be performed. It is possible to realize an excellent light source estimation device that performs light source estimation while performing determination and color chart recognition.

[Brief description of drawings]

FIG. 1 is a block connection diagram showing a configuration of a color chart recognition device according to a first embodiment of the present invention.

FIG. 2 is a flowchart showing the operation of the color chart recognition device in the first embodiment of the present invention.

FIG. 3 is a conceptual diagram showing an example of a sampling format of image data.

FIG. 4 is a conceptual diagram showing an example of a sampling format of image data.

FIG. 5 is a block connection diagram showing a configuration of a GWA determination device according to a second embodiment of the present invention.

FIG. 6 is a flowchart showing the operation of the GWA determining device in the second embodiment of the present invention.

FIG. 7 is a block connection diagram showing a configuration of a light source estimation device in a third embodiment of the present invention.

FIG. 8 is a flowchart showing the operation of the light source estimation device in the third embodiment of the present invention.

FIG. 9 is a conceptual diagram showing a light source line in a color space for explaining the principle of the second embodiment of the present invention.

FIG. 10 is a color chart for explaining the principle of the present invention and a conceptual diagram of distribution of a normal image and distribution in the third moment plane.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Sampling means 2 Chromaticity conversion means 3 Projection means 4 Variance calculation means 5 Third moment calculation means 61 Threshold processing means 62 Threshold processing means 63 Threshold processing means 7 AND circuit 8 OR circuit 9 Light source estimation section 10 Overall determination section 11 Light source estimation Result holding unit 100 Color chart recognition unit 200 GWA determination unit

Claims (21)

[Claims] 1. A color chart recognition device for recognizing a color chart based on a statistic of an image signal. 2. The color chart recognition device according to claim 1, wherein the statistic is a variance of a color distribution and a third moment. 3. The statistic is perpendicular to a straight line that approximates the trajectory drawn on the chromaticity plane when the color of the light source from the black body radiation is used as a parameter for the dispersion of the color distribution and the chromaticity signal. The color chart recognition device according to claim 1, wherein the color chart recognition device projects in any direction and is a cubic sum of the magnitudes of components in that direction. 4. Sampling means for converting the image signal IS from the reading means according to an appropriate sampling rule, chromaticity conversion means for converting the image signal sampled by the sampling means into chromaticity signals, and chromaticity signals. When the color of the light source from the black body radiation is the parameter of the temperature of the black body, it is projected in the direction perpendicular to the straight line approximating the locus drawn on the chromaticity plane and converted into the displacement which is the magnitude of the component in that direction. Projection means, dispersion calculation means for calculating the image signal variance of the displacement obtained from the sampling data, third moment calculation means for obtaining the third moment by receiving the displacement from the projection means, and receiving the signal and storing it internally. A color chart recognition device including a threshold value processing means for outputting a determination signal in comparison with the threshold value and an AND circuit. 5. The projection means for generating a signal input to the variance calculation means and the projection means for generating a signal input to the third moment calculation means have independent parameters. Color chart recognition device. 6. A sampling means for converting the image signal IS from the reading means according to an appropriate sampling rule, a chromaticity conversion means for converting the image signal sampled by the sampling means into a chromaticity signal, and a chromaticity signal. Projection means for projecting in a direction perpendicular to a straight line approximating the locus drawn on the chromaticity plane when the color of the light source from the black body radiation is the parameter of the temperature of the black body, and converting to a component size in that direction,
Dispersion calculating means for calculating the image signal dispersion of the displacement obtained from the sampling data, vertical third-order sum calculating means for obtaining the vertical third-order sum by receiving the displacement from the projecting means, and a threshold value internally stored by receiving the signal. And a color chart recognizing device including a threshold value processing unit that outputs a determination signal in comparison with. 7. The projection means for generating a signal input to the variance calculation means and the projection means for generating a signal input to the vertical cubic sum calculation means have independent parameters. Color chart recognition device. 8. A GWA determination device that determines whether the average of all screens is gray and whether it is a color chart, based on the statistics of image signals. 9. The GWA determination device according to claim 8, wherein the statistics are variance and third moment. 10. The GWA determining device according to claim 8, wherein the statistic is a variance and a vertical cubic sum. 11. Sampling means for converting the image signal IS from the reading means according to an appropriate sampling rule, chromaticity conversion means for converting the image signal sampled by the sampling means into chromaticity signals, and chromaticity signals. Projection means for projecting in the direction perpendicular to the light source line described in the explanation of the principle and converting to the component size in that direction, dispersion calculation means for calculating the image signal dispersion of the displacement obtained from the sampling data, and projection means Third moment calculating means for receiving a third moment by receiving displacement, threshold processing means for receiving a signal and comparing with a threshold value stored inside and outputting a determination signal, and logic of signals from two threshold processing means A logical product circuit for taking a product, threshold processing means having a threshold value set so as to be suitable for the GWA determination of whether or not the overall average of the screen is gray, and a logic A GWA judging device comprising a logical sum circuit for taking the logical sum of the output from the product circuit and the output of the threshold value processing means for judging the GWA. 12. The projection means for generating a signal input to the variance calculation means and the projection means for generating a signal input to the third moment calculation means have independent parameters. GWA determination device. 13. A sampling means for converting an image signal IS from a CCD or the like according to an appropriate sampling rule, a chromaticity conversion means for converting an image signal sampled by the sampling means into a chromaticity signal, and a chromaticity signal. Projection means for projecting in the direction perpendicular to the light source line described in the explanation of the principle and converting to the component size in that direction, dispersion calculation means for calculating the image signal dispersion of the displacement obtained from the sampling data, and projection means A vertical third-order sum calculation means for receiving a displacement to obtain a vertical third-order sum, a threshold processing means for receiving a signal and outputting a determination signal in comparison with a threshold value stored inside, and signals from two threshold processing means. Output of the AND circuit, the threshold value processing means having a threshold value set so as to be suitable for the GWA determination, and the output of the AND circuit and the threshold value processing means for performing the GWA determination. A GWA determination device configured with a logical sum circuit that takes a logical sum with a force. 14. The projection means for generating a signal input to the variance calculation means and the projection means for generating a signal input to the vertical cubic sum calculation means have independent parameters. GWA determination device. 15. A light source estimation device for estimating a light source by evaluating a light source estimation result by determining whether the total average of a screen is gray and whether it is a color chart based on a statistical amount of an image signal. 16. The light source estimation apparatus according to claim 15, wherein the statistics are variance and third moment. 17. The light source estimation apparatus according to claim 15, wherein the statistic is a variance and a vertical cubic sum. 18. A sampling means for converting the image signal IS from the reading means according to an appropriate sampling rule, a chromaticity conversion means for converting the image signal sampled by the sampling means into a chromaticity signal, and a chromaticity signal. Projection means for projecting in a direction perpendicular to a straight line approximating the locus drawn on the chromaticity plane when the color of the light source from the black body radiation is the parameter of the temperature of the black body, and converting to a component size in that direction, Dispersion calculating means for calculating the image signal dispersion of the displacement obtained from the sampling data, third moment calculating means for obtaining the third moment by receiving the displacement from the projecting means, and comparison with the threshold value stored internally upon receiving the signal A threshold value processing unit for outputting a determination signal, a logical product circuit for obtaining a logical product of the signals from the two threshold value processing units, and a threshold value suitable for GWA determination. A threshold value processing unit, a logical sum circuit that takes the logical sum of the output from the logical product circuit and the output of the threshold value processing unit that performs the GWA determination, and a light source estimation unit that estimates the light source from the image data sampled by the sampling unit. When the light source estimation control signal is received from the logical sum circuit and the light source estimation result is received from the light source estimation unit to make a comprehensive determination, and the comprehensive determination result from the comprehensive determination unit is received and it is impossible to make a determination A light source estimation result holding means and a light source estimation device for outputting the light source estimation result held inside. 19. The projection means for generating a signal input to the variance calculation means and the projection means for generating a signal input to the third moment calculation means have independent parameters. Light source estimation device. 20. Sampling means for converting the image signal IS from the reading means according to an appropriate sampling rule, chromaticity conversion means for converting the image signal sampled by the sampling means into chromaticity signals, and chromaticity signals Projection means for projecting in a direction perpendicular to a straight line approximating the locus drawn on the chromaticity plane when the color of the light source from the black body radiation is used as a parameter and converting it to the magnitude of the component in that direction. , A variance calculating means for calculating the image signal variance of the displacement obtained from the sampling data, a vertical third-order sum calculating means for obtaining the vertical third-order sum by receiving the displacement from the projecting means, and a signal stored internally. A threshold processing unit that outputs a determination signal by comparing with a threshold value, an AND circuit that obtains a logical product of signals from two threshold processing units, and a threshold value set with a threshold value suitable for GWA determination. A processing unit, a logical sum circuit that takes the logical sum of the output from the logical product circuit and the output of the threshold value processing unit that determines the GWA, a light source estimation unit that estimates the light source from the image data sampled by the sampling unit, and a logic An internal judgment unit that receives a light source estimation control signal from a summing circuit and receives a light source estimation result from a light source estimation unit to make a comprehensive judgment, and an internal judgment when the comprehensive judgment result from the comprehensive judgment unit cannot be judged A light source estimation device comprising a light source estimation result holding means for outputting the light source estimation result held in the. 21. The projection means for generating a signal input to the variance calculation means and the projection means for generating a signal input to the vertical cubic sum calculation means have independent parameters. The light source estimation device described.