CN106791758B - The judgment method and device of natural light mixing colour temperature in a kind of image - Google Patents

Abstract

The invention discloses a method and device for judging the mixed color temperature of natural light in an image. The method includes: dividing the image into a plurality of rectangular areas, and determining the value of a pair of natural light mixed color temperature evaluation parameters for each rectangular area; according to the parameters Determining each first coordinate point corresponding to each rectangular area; dividing the first coordinate point into a first cluster and a second cluster, and determining a first unit direction vector according to the center point of each cluster; when When the included angle between the first unit direction vector and the pre-stored second unit direction vector is smaller than the first set threshold, it is determined that the image to be corrected has a mixed color temperature of natural light. In the embodiment of the present invention, the angle between the first unit direction vector and the second unit direction vector reflects the possibility of the image having a mixed color temperature of natural light, so it is more accurate to judge whether the image to be corrected has a mixed color temperature of natural light.

Description

A method and device for judging the mixed color temperature of natural light in an image

technical field

The present invention relates to the technical field of image processing, in particular to a method and a device for judging the mixed color temperature of natural light in an image.

Background technique

With the continuous development of image processing technology, the white balance correction method is also becoming more and more perfect. In the existing technology, when performing white balance correction on the image to be corrected, the automatic white balance correction method is often used. Figure 1 shows the white balance correction method in the prior art Automatic white balance correction method, the specific process includes:

S101: Collect a reference white point of the image to be corrected, and determine the reference white point to draw a reference white area.

S102: Calculate the number of reference white points that currently fall into the reference white area of the image to be corrected. If the number of reference white points is less than or equal to a first preset threshold, go to S105; otherwise, go to S103.

S103: Determine whether the image to be corrected has a mixed color temperature of natural light, if so, go to S105, otherwise, go to S104.

S104: Calculate a white balance gain according to the number of white points falling into the reference white area, and use the white balance gain to perform color correction on the image to be corrected.

S105: Calculate a white balance gain using a preset algorithm strategy, and use the white balance gain to perform color correction on the image to be corrected.

This method dynamically clusters the white points of the image to be corrected that fall into the reference white area, and divides the white points into two clusters. When the number of white points in the two clusters satisfies the preset condition, then judge The image to be corrected has a mixed color temperature of natural light, but the law of mixed color temperature of natural light is not considered when making judgments. For example, when there are white roads and green plants in the image, the white and green are divided into two clusters through dynamic clustering. After judging that the number of white points in the white and green clusters meets the preset conditions, the method It is considered that the image has a mixed color temperature of natural light, and the white balance gain is calculated separately for white and green, but in reality, the image does not have a mixed color temperature of natural light. This makes it wrong to judge whether there is natural light mixed color temperature in the image, thereby reducing the accuracy of white balance correction.

Contents of the invention

Embodiments of the present invention provide a method and device for judging the mixed color temperature of natural light in an image to solve the problem of inaccurate judgment of the mixed color temperature of natural light in an image in the prior art.

The method of the present invention includes a method for judging the mixed color temperature of natural light in an image, the method comprising:

For the image to be corrected, divide the image into multiple rectangular areas according to the set rules, and determine a pair of natural light mixed color temperature evaluations for each rectangular area according to the mean values of red R, green G, and blue B in each rectangular area the value of the parameter;

Each first coordinate point corresponding to each rectangular area is determined in the constructed coordinate system according to the value of a pair of natural light mixed color temperature evaluation parameters of each rectangular area, wherein the coordinate system uses the pair of natural light mixed color temperature evaluation parameters as the abscissa axis and vertical axis;

Divide the first coordinate point in the coordinate system into a first cluster and a second cluster, and determine a first unit direction vector according to the center point of each cluster, wherein the direction of the first unit direction vector is From low color temperature to high color temperature;

judging whether the included angle between the first unit direction vector and the pre-saved second unit direction vector is smaller than the set first threshold, wherein the second unit direction vector is predetermined and saved according to the target image with mixed color temperature of natural light ;

If yes, it is determined that the image to be corrected has a mixed color temperature of natural light; otherwise, it is determined that the image to be corrected does not have a mixed color temperature of natural light.

Further, the process of determining the second unit direction vector includes:

For the pre-collected target image of mixed color temperature of natural light, the target image is divided into multiple rectangular areas according to the set rules, and a pair of natural light in each rectangular area is determined according to the R, G, and B mean values of each rectangular area. The value of the mixed color temperature evaluation parameter;

According to the value of a pair of natural light mixed color temperature evaluation parameters of each rectangular area, each second coordinate point corresponding to each rectangular area is determined in the constructed coordinate system, wherein the coordinate system uses the pair of natural light mixed color temperature evaluation parameters as horizontal and vertical axes;

According to the coordinates of each second coordinate point in the coordinate system, a second unit direction vector is determined, wherein the direction of the second unit direction vector is a direction from low color temperature to high color temperature.

Further, the determining the second unit direction vector according to the coordinates of each second coordinate point in the coordinate system includes:

Acquire coordinate information of two coordinate points input by the user, determine direction vectors of the two coordinate points, and use the unit vector corresponding to the direction vector as the second unit direction vector; or

According to the method of least squares, determine the equation of a line corresponding to the second coordinate point in the coordinate system, and determine the second unit direction vector corresponding to the equation of the line according to the equation of the line; or

According to the principal component analysis method, the covariance matrix of the point set composed of the second coordinate points is determined, and the second unit direction vector is determined according to the eigenvectors of the covariance matrix.

Further, after it is determined that the image to be corrected has a mixed color temperature of natural light, the method further includes:

The white balance gain is determined according to the natural light mixed color temperature white balance algorithm and the image is corrected.

Further, the determining the white balance gain according to the natural light mixed color temperature white balance algorithm includes:

According to the angle between the first unit direction vector and the second unit direction vector and the setting parameters, using a preset algorithm to determine the natural light mixing color temperature parameter value corresponding to the angle and the setting parameters;

Determine the first weight and the second weight corresponding to the single color temperature white balance algorithm and the natural light mixed color temperature white balance algorithm according to the natural light mixed color temperature parameter value, according to the single color temperature white balance algorithm and the natural light mixed color temperature white balance algorithm and the corresponding The first weight and the second weight determine the white balance gain.

Further, according to the natural light mixed color temperature parameter value, determining the first weight and the second weight corresponding to the single color temperature white balance algorithm and the natural light mixed color temperature white balance algorithm includes:

The value of the natural light mixed color temperature parameter is used as a second weight, and the difference between 1 and the second weight is used as a first weight.

Further, according to the angle between the first unit direction vector and the pre-saved second unit direction vector and the setting parameters, a preset algorithm is used to determine the natural light corresponding to the angle and the setting parameters. Mixed color temperature parameter values include:

According to the included angle between the first unit direction vector and the pre-stored second unit direction vector, calculate the product of the included angle and a preset parameter; determine the cosine value of the product as the natural light mixing color temperature parameter value.

Further, after it is determined that the image to be corrected does not have a mixed color temperature of natural light, the method further includes:

The white balance gain is determined according to the monochromatic temperature white balance algorithm and the image is corrected.

Further, before the judging whether the included angle between the first unit direction vector and the pre-saved second unit direction vector is smaller than the set first threshold, the method further includes:

Determine the ambient illuminance of the image according to the stored camera shutter time, gain parameters, and aperture, and when the ambient illuminance is greater than the second set threshold, continue to the subsequent steps; or

When it is judged that the distance between the center points of the first cluster and the second cluster is greater than the third set threshold, continue to the subsequent steps; or

Judging that the abscissas of the center points of the first cluster and the second cluster are both within the range of abscissas corresponding to the color temperature of natural light, and the ordinates of the center points of the first cluster and the second cluster are both within the range corresponding to the color temperature of natural light If it is within the range of the ordinate, continue to the next step.

Further, the values of a pair of natural light mixed color temperature evaluation parameters for each rectangular area include:

G/R value and G/B value for each rectangular area; or

the logarithmic values of G/R and the logarithmic values of G/B for each rectangular area; or

The logarithmic value of R/G and the logarithmic value of B/G of each rectangular area.

On the other hand, an embodiment of the present invention provides a device for judging the mixed color temperature of natural light in an image, the device comprising:

The first determination module is used to divide the image to be corrected into a plurality of rectangular areas according to the set rules, and determine each rectangular area according to the mean value of red R, green G, and blue B of each rectangular area. The value of a pair of natural light mixed color temperature evaluation parameters;

The second determination module is used to determine each first coordinate point corresponding to each rectangular area in the constructed coordinate system according to the value of a pair of natural light mixed color temperature evaluation parameters of each rectangular area, wherein the coordinate system is based on the pair The natural light mixed color temperature evaluation parameters are used as the horizontal axis and the vertical axis;

The third determination module is used to divide the first coordinate points in the coordinate system into the first cluster and the second cluster, and determine the first unit direction vector according to the center point of each cluster, wherein the first unit direction vector The direction of a unit direction vector is the direction from low color temperature to high color temperature;

The first judging module is used to judge whether the included angle between the first unit direction vector and the pre-saved second unit direction vector is smaller than the set first threshold, wherein the second unit direction vector is based on the mixed color temperature of natural light The target image is predetermined and saved;

The fourth determination module is configured to determine that the image to be corrected has a natural light mixed color temperature if the judgment result of the first judgment module is Yes, and determine that the image to be corrected does not have a natural light mixed color temperature if the judgment result of the first judgment module is No.

Further, the first determination module is also used for dividing the target image into a plurality of rectangular areas according to the set rules for the pre-collected target image of mixed color temperature of natural light, and according to the R, G , B mean value, determine the value of a pair of natural light mixed color temperature evaluation parameters for each rectangular area;

The second determination module is further configured to determine each second coordinate point corresponding to each rectangular area in the constructed coordinate system according to the value of a pair of natural light mixed color temperature evaluation parameters of each rectangular area, wherein the coordinate system is represented by The pair of natural light mixed color temperature evaluation parameters are used as the horizontal axis and the vertical axis;

The third determining module is further configured to determine a second unit direction vector according to the coordinates of each second coordinate point in the coordinate system, wherein the direction of the second unit direction vector is the direction from low color temperature to high color temperature .

Further, the third determining module is specifically configured to obtain coordinate information of two coordinate points input by the user, determine direction vectors of the two coordinate points, and use the unit vector corresponding to the direction vector as the first Two unit direction vectors; or, according to the least square method, determine the straight line equation corresponding to the second coordinate point in the coordinate system, and determine the second unit direction vector corresponding to the straight line equation according to the straight line equation; or, according to the main The component analysis method is to determine the covariance matrix of the point set composed of the second coordinate points, and determine the second unit direction vector according to the eigenvectors of the covariance matrix.

Further, the device also includes:

The fifth determination module is configured to determine the white balance gain according to the natural light mixed color temperature white balance algorithm and correct the image after it is determined that the image to be corrected has the mixed color temperature of natural light.

Further, the fifth determining module is specifically configured to determine the included angle and the set parameter by using a preset algorithm according to the included angle between the first unit direction vector and the second unit direction vector and setting parameters. The natural light mixed color temperature parameter value corresponding to the given parameter; the first weight and the second weight corresponding to the monochromatic temperature white balance algorithm and the natural light mixed color temperature white balance algorithm are determined according to the natural light mixed color temperature parameter value, according to the monochromatic temperature white balance Algorithm and natural light mixed color temperature white balance algorithm and corresponding first weight and second weight to determine white balance gain.

Further, the fifth determining module is specifically configured to use the natural light mixing color temperature parameter value as the second weight, and use the difference between 1 and the second weight as the first weight.

Further, the fifth determination module is specifically configured to calculate the product of the included angle and a preset parameter according to the included angle between the first unit direction vector and the pre-saved second unit direction vector; The cosine value of is determined as the natural light mixing color temperature parameter value.

Further, the device also includes:

The fifth determination module is configured to determine the white balance gain according to the single color temperature white balance algorithm and correct the image after determining that there is no natural light mixed color temperature in the image to be corrected.

Further, the device also includes:

The second judging module is used to determine the ambient illuminance of the image according to the saved camera shutter time, gain parameters, and aperture, and when the ambient illuminance is greater than the second set threshold, continue to the subsequent steps; or judge the first cluster and the second When the distance between the center points of the two clusters is greater than the third set threshold, continue to the subsequent steps; or judge that the abscissas of the center points of the first cluster and the second cluster are both within the range of abscissas corresponding to the color temperature of natural light , and when the ordinates of the center points of the first cluster and the second cluster are both within the range of ordinates corresponding to the color temperature of natural light, the first judging module is triggered.

Further, the first determination module is specifically used to determine the G/R value and G/B value of each rectangular area in each rectangular area; or, determine the logarithmic value of G/R and the G/B value of each rectangular area / logarithmic value of B; or, determine the logarithmic value of R/G and the logarithmic value of B/G of each rectangular area of each rectangular area.

Embodiments of the present invention provide a method and device for judging the mixed color temperature of natural light in an image. The method includes: for the image to be corrected, divide the image into a plurality of rectangular areas according to set rules, and divide the image into a plurality of rectangular areas according to each rectangular area. Determine the value of a pair of natural light mixed color temperature evaluation parameters for each rectangular area; according to the value of a pair of natural light mixed color temperature evaluation parameters for each rectangular area, determine each in the constructed coordinate system. Each first coordinate point corresponding to a rectangular area, wherein the coordinate system uses the pair of natural light mixed color temperature evaluation parameters as the horizontal axis and the vertical axis; the first coordinate point in the coordinate system is divided into the first cluster and For the second clustering, determine the first unit direction vector according to the center point of each cluster, wherein the direction of the first unit direction vector is the direction from low color temperature to high color temperature; judge the first unit direction vector and the pre-saved first unit direction vector Whether the included angle between the two unit direction vectors is smaller than the set first threshold, wherein the second unit direction vector is predetermined and saved according to the target image with mixed color temperature of natural light; if yes, it is determined that the image to be corrected has natural light Mixed color temperature, otherwise, make sure there is no natural light mixed color temperature in the image to be corrected. Since the second unit direction vector is determined for the target image with mixed color temperature of natural light, when the angle between the first unit direction vector and the second unit direction vector is smaller than the set first threshold, it indicates that the image to be corrected Similar to the target image with the mixed color temperature of natural light, the mixed color temperature of natural light also exists, so that the accuracy of judging the color temperature of the mixed natural light of the image can be ensured, and the accuracy of subsequent white balance correction can be further ensured.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings that need to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For Those of ordinary skill in the art can also obtain other drawings based on these drawings without making creative efforts.

FIG. 1 is a schematic diagram of an automatic white balance correction process provided by the prior art;

FIG. 2 is a schematic diagram of a judging process of the mixed color temperature of natural light in an image provided by Embodiment 1 of the present invention;

FIG. 3 is a schematic diagram of an image divided into rectangular regions provided by Embodiment 1 of the present invention;

FIG. 4 is a schematic diagram of a coordinate system corresponding to a target image provided in Embodiment 2 of the present invention;

Fig. 5 is a schematic diagram of a judging process of the mixed color temperature of natural light in an image provided by Embodiment 4 of the present invention;

FIG. 6 is a schematic structural diagram of an apparatus for judging the mixed color temperature of natural light in an image provided by an embodiment of the present invention.

Detailed ways

In order to make the judgment of the mixed color temperature of natural light in an image more accurate, and further improve the accuracy of subsequent white balance corrections, embodiments of the present invention provide a method and device for judging the mixed color temperature of natural light in an image.

The present invention will be further described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

Example 1:

Fig. 2 is a schematic diagram of a process of judging the mixed color temperature of natural light in an image provided by an embodiment of the present invention, and the process includes the following steps:

S201: For the image to be corrected, divide the image into multiple rectangular areas according to the set rules, and determine a pair of natural light mixtures for each rectangular area according to the mean values of red R, green G, and blue B in each rectangular area The value of the color temperature evaluation parameter.

A method for judging the mixed color temperature of natural light in an image provided by an embodiment of the present invention is applied to an electronic device. The electronic devices include PCs, smart TVs, cameras, and mobile phones.

The electronic device divides the image to be corrected into multiple rectangular areas according to the set rules, and when dividing the image to be corrected, the length of the image can be divided into n equal parts, and the width of the image can be divided into m equal parts, wherein The m and n are the same or different, and multiple rectangular areas are obtained after division. As shown in Figure 3, the length and width of the image are divided into 6 equal parts, that is, the image is divided into 36 rectangular areas. According to the red R, green G, and blue B values of each pixel in each rectangular area, determine the R, G, and B mean values of each rectangular area, and then determine the pair of natural light mixed color temperature evaluation parameters for each rectangular area value, the value of the pair of natural light mixed color temperature evaluation parameters can be the G/R and G/B values of each rectangular area, or the logarithmic value of G/R and the logarithmic value of G/B of each rectangular area , can also be the logarithmic value of R/G and the logarithmic value of B/G of each rectangular area.

S202: Determine each first coordinate point corresponding to each rectangular area in the constructed coordinate system according to the value of a pair of natural light mixed color temperature evaluation parameters of each rectangular area, wherein the coordinate system is based on the pair of natural light mixed color temperature evaluation parameters as horizontal and vertical axes.

The value of the pair of natural light mixed color temperature evaluation parameters can be the G/R and G/B values of each rectangular area, or the logarithmic value of G/R and the logarithmic value of G/B of each rectangular area, or It can be the logarithmic value of R/G and the logarithmic value of B/G of each rectangular area, and use the pair of natural light mixed color temperature evaluation parameters as the horizontal axis and the vertical axis to construct a coordinate system and determine each rectangular area corresponding to each first coordinate point.

For example, the pair of natural light mixed color temperature evaluation parameters are the G/R and G/B values of each rectangular area, and the coordinate system is constructed with G/R as the horizontal axis of the coordinate system and G/B as the vertical axis of the coordinate system axis. Therefore, for each rectangular area, the G/R and G/B values of the rectangular area can be considered as the abscissa and ordinate of the rectangular area, so the rectangular area is marked in the coordinate system, which can be considered as the abscissa is Its G/R value, the ordinate is the coordinate point of its G/B value.

For example, the value of the pair of natural light mixed color temperature evaluation parameters can also be the logarithmic value of G/R and the logarithmic value of G/B in each rectangular area, and the constructed coordinate system takes the logarithmic value of G/R as the horizontal axis of the coordinate system , taking the logarithmic value of G/B as the vertical axis of the coordinate system. The value of the pair of natural light mixed color temperature evaluation parameters can also be the logarithmic value of R/G and the logarithmic value of B/G of each rectangular area, and the constructed coordinate system takes the logarithmic value of R/G as the horizontal axis of the coordinate system, Take the logarithmic value of B/G as the vertical axis of the coordinate system.

S203: Divide the first coordinate point in the coordinate system into a first cluster and a second cluster, and determine a first unit direction vector according to the center point of each cluster, wherein the first unit direction vector The direction is from low color temperature to high color temperature.

According to the value of a pair of natural light mixed color temperature evaluation parameters in each rectangular area, after determining the first coordinate point corresponding to each rectangular area in the coordinate system, a clustering algorithm can be used to cluster the first coordinate points in the coordinate system Classes, specifically during clustering, the first coordinate point in the coordinate system is divided into a first cluster and a second cluster. Wherein, when clustering the first coordinate points in the coordinate system, general clustering methods, such as kmeans algorithm, K-means algorithm, etc., can be used.

In the embodiment of the present invention, the first coordinate points in the coordinate system may be divided into a first cluster and a second cluster by using a K-means algorithm. After being divided into two clusters, the center point of the first cluster and the second cluster can be respectively determined by using the mean shift method.

The higher the color temperature, the bluer the tone, the greater the B value; the lower the color temperature, the redder the tone, the greater the R value. The specific process of determining the direction from low color temperature to high color temperature belongs to the prior art, and will not be described in detail in the embodiment of the present invention.

Specifically, the clustering algorithm is used to divide the first coordinate points in the coordinate system into the first cluster and the second cluster, and the process of using the mean shift method to determine the center points of the first cluster and the second cluster belongs to the existing technology, and this process will not be described in detail in this embodiment of the present invention.

S204: Determine whether the included angle between the first unit direction vector and the pre-saved second unit direction vector is smaller than the set first threshold, wherein the second unit direction vector is determined in advance according to the target image with mixed color temperature of natural light and saved, if yes, go to S205, otherwise, go to S206.

In the embodiment of the present invention, the second unit direction vector is generated according to the target image with mixed color temperature of natural light. Because for the target image with natural light mixed color temperature, when it is divided into multiple rectangular areas according to the set rules, the value of a pair of natural light mixed color temperature evaluation parameters in each rectangular area is distributed approximately on a straight line in the coordinate system, Therefore, according to this feature, the unit vector corresponding to the straight line in the direction from low color temperature to high color temperature can be determined, and this unit vector can be used as the second unit direction vector.

S205: Determine that the image to be corrected has a mixed color temperature of natural light.

S206: Determine that the image to be corrected does not have a mixed color temperature of natural light.

After the first unit direction vector is determined, according to the angle between the first unit direction vector and the second unit direction vector, it can be judged whether the image to be corrected has a mixed color temperature of natural light. When the angle is smaller than the first set threshold , it shows that the value of a pair of natural light mixed color temperature evaluation parameters in the divided rectangular area of the image also conforms to the distribution characteristics of the natural light mixed color temperature image, so it is determined that the image to be corrected has natural light mixed color temperature, when the included angle is not less than the first When setting the threshold, it is determined that there is no natural light mixed color temperature in the image to be corrected. Wherein the first set threshold may be a smaller value such as 3 degrees, 5 degrees, or the like.

Since in the embodiment of the present invention, it is judged whether the image to be corrected has a mixed color temperature of natural light according to the angle between the first unit direction vector and the second unit direction vector, the second unit direction vector is for the target image with a mixed color temperature of natural light. Determined and saved, therefore, when the angle between the first unit direction vector and the second unit direction vector is smaller than the set first threshold, it means that the image to be corrected is similar to the target image with natural light mixed color temperature, and there is also natural light Mixing color temperature, so as to ensure the accuracy of judging the mixed color temperature of natural light in the image, and further ensure the accuracy of subsequent white balance correction.

Example 2:

On the basis of the above-mentioned embodiments, in order to accurately determine the second unit direction vector according to the pre-collected target image of the mixed color temperature of natural light and improve the accuracy of subsequent judgments, in the embodiment of the present invention, the second unit direction is determined The vector process includes:

For the pre-collected target image of mixed color temperature of natural light, the target image is divided into multiple rectangular areas according to the set rules, and a pair of natural light in each rectangular area is determined according to the R, G, and B mean values of each rectangular area. The value of the mixed color temperature evaluation parameter;

According to the value of a pair of natural light mixed color temperature evaluation parameters of each rectangular area, each second coordinate point corresponding to each rectangular area is determined in the constructed coordinate system, wherein the coordinate system uses the pair of natural light mixed color temperature evaluation parameters as horizontal and vertical axes;

According to the coordinates of each second coordinate point in the coordinate system, a second unit direction vector is determined, wherein the direction of the second unit direction vector is a direction from low color temperature to high color temperature.

The electronic device pre-collects a target image with mixed color temperature of natural light. The target image contains information such as blue sky, sunlight, shadows, etc., and divides the target image into multiple rectangular areas according to the set rules. When dividing the target image into a plurality of rectangular regions, various methods can be adopted, as long as the method of dividing the rectangular region of the image in the above-mentioned embodiment is the same. For example, when dividing the target image, the length of the image can be divided into n equal parts, and the width of the image can be divided into m equal parts, wherein the m and n are the same or different, and multiple rectangular areas can be obtained after division. The specific division method Refer to Figure 3. According to the red R, green G, and blue B values of each pixel in each rectangular area, determine the R, G, and B mean values of each rectangular area, and then determine the value of a pair of natural light mixed color temperature evaluation parameters for each rectangular area .

The value of the pair of natural light mixed color temperature evaluation parameters can be the G/R and G/B values of each rectangular area, or the logarithmic value of G/R and the logarithmic value of G/B of each rectangular area, or It can be the logarithmic value of R/G and the logarithmic value of B/G of each rectangular area.

Taking the pair of natural light mixed color temperature evaluation parameters as the G/R and G/B values of each rectangular area as an example, after determining the G/R and G/B values of each rectangular area, construct the coordinate system Each second coordinate point corresponding to the G/R and G/B values of each rectangular area is determined in , for each second coordinate point determined in the coordinate system for a target image with mixed color temperature of natural light, refer to FIG. 4 .

According to Figure 4, there is a target image with mixed color temperature of natural light. After dividing it into multiple rectangular areas, each rectangular area is determined in the constructed coordinate system according to the G/R and G/B values of each rectangular area. After the corresponding second coordinate point, the distribution of the second coordinate point in the coordinate system is approximately a straight line, so the second unit direction vector can be determined according to the straight line, wherein the direction of the second unit direction vector is from low color temperature to High color temperature direction.

The specific method of determining a pair of mixed color temperature evaluation parameters for each rectangular area and constructing a coordinate system is the same as the method in Embodiment 1 of the present invention, and will not be repeated in this embodiment of the present invention.

In the embodiment of the present invention, the second unit direction vector can be determined by different methods, and according to the coordinates of each second coordinate point in the coordinate system, determining the second unit direction vector includes:

Acquire coordinate information of two coordinate points input by the user, determine direction vectors of the two coordinate points, and use the unit vector corresponding to the direction vector as the second unit direction vector; or

According to the method of least squares, determine the equation of a line corresponding to the second coordinate point in the coordinate system, and determine the second unit direction vector corresponding to the equation of the line according to the equation of the line; or

According to the principal component analysis method, the covariance matrix of the point set composed of the second coordinate points is determined, and the second unit direction vector is determined according to the eigenvectors of the covariance matrix.

The second coordinate point in the coordinate system is similar to a straight line, the user can select two coordinate points on the straight line, or the user inputs the coordinates of the two coordinate points, after the electronic device determines the coordinates of the two coordinate points, and then A direction vector from a low color temperature to a high color temperature direction of the two coordinate points may be determined, and a unit vector corresponding to the direction vector is used as the second unit direction vector.

For example, if the two coordinate points determined by the electronic device are A(gr1, gb1) and B(gr2, gb2), then it can be determined that the direction vectors corresponding to the two coordinate points A and B are C=B-A, where the direction is from low color temperature to High color temperature direction, then find the second unit direction vector e=C/|C|

It may also be that the electronic device determines the linear equation gb=k*gr+b corresponding to the second coordinate point in the coordinate system through the least square method according to the coordinates of each second coordinate point in the coordinate system, wherein, The process of fitting the linear equation by the least square method belongs to the prior art, and the implementation process will not be described in detail in the embodiment of the present invention. After the straight line equation is determined, determine any two coordinate points that satisfy the straight line equation, and then determine the direction vector from the low color temperature to the high color temperature direction of the two coordinate points, and use the unit vector corresponding to the direction vector as the second unit direction vector.

It is also possible to determine the covariance matrix of the point set composed of the second coordinate points according to each second coordinate point in the coordinate system through the method of principal component analysis, and obtain the eigenvector of the covariance matrix, According to the feature vector, a second unit direction vector is further determined, wherein the direction of the second unit direction vector is from a low color temperature to a high color temperature. The process of determining the covariance matrix through the method of principal component analysis belongs to the prior art, and this process will not be described in detail in the embodiment of the present invention.

Example 3:

In order to realize the white balance correction of the image, on the basis of the above-mentioned embodiments, in the embodiment of the present invention, after determining that the image to be corrected has a mixed color temperature of natural light, the method further includes:

The white balance gain is determined according to the natural light mixed color temperature white balance algorithm and the image is corrected.

The natural light mixed color temperature white balance algorithm may be a grayscale world algorithm. Specifically, the process of determining the white balance gain by using the grayscale world algorithm belongs to the prior art, and the process of correcting the image according to the white balance gain is also the prior art. This process is not described in detail in this embodiment of the present invention.

In order to make the white balance correction more accurate, the determination of the white balance gain according to the natural light mixed color temperature white balance algorithm in the embodiment of the present invention includes:

According to the angle between the first unit direction vector and the second unit direction vector and the setting parameters, using a preset algorithm to determine the natural light mixing color temperature parameter value corresponding to the angle and the setting parameters;

Determine the first weight and the second weight corresponding to the single color temperature white balance algorithm and the natural light mixed color temperature white balance algorithm according to the natural light mixed color temperature parameter value, according to the single color temperature white balance algorithm and the natural light mixed color temperature white balance algorithm and the corresponding The first weight and the second weight determine the white balance gain.

The directions of the first unit direction vector and the second unit direction vector are from low color temperature to high color temperature. According to the angle between the first unit direction vector and the second unit direction vector and the parameters set by the electronic device, the preset The established algorithm determines the included angle and the natural light mixed color temperature parameter value corresponding to the set parameter. Wherein, the preset algorithm can be to calculate the product of the angle and the preset parameter according to the angle between the first unit direction vector and the second unit direction vector; determine the cosine value of the product as the natural light Mixed color temperature parameter value.

Specifically, the angle between the first unit direction vector and the second unit direction vector is α, and the parameter is set as k. According to the angle α between the first unit direction vector and the pre-saved second unit direction vector, and the preset Assume the product of the parameter k; determine the cosine value of the product as the natural light mixed color temperature parameter value, that is, the natural light mixed color temperature parameter value is cos(kα).

In this embodiment, the single color temperature white balance algorithm may be a dynamic threshold algorithm, and the natural light mixed color temperature white balance algorithm may be a grayscale world algorithm.

After the natural light mixed color temperature parameter value is determined, the first weight of the single color temperature white balance algorithm and the second weight of the natural light mixed color temperature white balance algorithm can be determined according to the natural light mixed color temperature parameter value, and the natural light mixed color temperature parameter value can be determined. As the second weight, the difference between 1 and the second weight is used as the first weight, that is, the first weight corresponding to the dynamic threshold algorithm is 1-cos(kα), and the second weight corresponding to the grayscale world algorithm is cos(kα ). Of course, when determining the first weight and the second weight corresponding to the single color temperature white balance algorithm and the natural light mixed color temperature white balance algorithm, other methods can also be used, as long as the second weight can be guaranteed to be equal to the first weight and the second weight The ratio of the sum to the sum may be equal to the natural light mixed color temperature parameter value.

After the first weight and the second weight are determined, the final white balance gain of the image to be corrected is calculated through the monochromatic temperature white balance algorithm and the natural light mixed color temperature white balance algorithm, and the corresponding first weight and second weight, Its algorithm is shown in the following formula:

RGain=(1-cos(kα))*RGain1+cos(kα)*RGain2;

BGain=(1-cos(kα))*BGain1+cos(kα)*BGain2;

Among them, RGain1 and BGain1 are the white balance gains of R and B calculated according to the single color temperature white balance algorithm, RGain2 and BGain2 are the white balance gains of R and B calculated according to the natural light mixed color temperature white balance algorithm, and RGain and BGain are Computed final white balance gain for the image to be corrected.

The image is corrected according to the calculated final white balance gain of the image to be corrected. Specifically, the process of correcting the image according to the white balance gain belongs to the prior art, and will not be described in detail in this embodiment of the present invention.

In the embodiment of the present invention, when performing white balance correction on the image to be corrected, only the R and B components in the image are corrected, and the G component is kept unchanged, so that the brightness of the image to be corrected can be adjusted by white balance constant.

Since the final white balance gain is calculated according to the single color temperature white balance algorithm and the mixed color temperature white balance algorithm and their corresponding first weights and second weights, when performing white balance correction on an image to be corrected with mixed color temperatures of natural light, The image to be corrected can be corrected according to the mixed color temperature degree of natural light of the image to be corrected, so that the accuracy of white balance correction can be ensured.

In order to realize the white balance correction of the image, on the basis of the above-mentioned embodiments, in the embodiment of the present invention, after determining that the image to be corrected does not have a mixed color temperature of natural light, the method further includes:

The white balance gain is determined according to the monochromatic temperature white balance algorithm and the image is corrected.

The monochromatic temperature white balance algorithm may be a dynamic threshold algorithm. Specifically, the process of using the dynamic threshold algorithm to determine the white balance gain belongs to the prior art, and the process of correcting the image according to the white balance gain is also the prior art. In this paper This process is not described in detail in the embodiment of the invention.

Example 4:

On the basis of Embodiment 1, in order to further improve the accuracy of judging whether there is natural light mixed color temperature in the image, it is judged whether the angle between the first unit direction vector and the pre-saved second unit direction vector is smaller than the set first threshold, Wherein, before the second unit direction vector is pre-determined and saved according to the target image with mixed color temperature of natural light, the method further includes:

Judging the ambient illuminance of the image according to the saved camera shooting shutter time, gain parameters, and aperture, when the ambient illuminance is greater than the second set threshold, continue to the subsequent steps; or

When it is judged that the distance between the center points of the first cluster and the second cluster is greater than the third set threshold, continue to the subsequent steps; or

Judging that the abscissas of the center points of the first cluster and the second cluster are both within the range of abscissas corresponding to the color temperature of natural light, and the ordinates of the center points of the first cluster and the second cluster are both within the range corresponding to the color temperature of natural light If it is within the range of the ordinate, continue to the next step.

In the embodiment of the present invention, since the illuminance of the natural light mixed color temperature scene is relatively large, the electronic device judges the ambient illuminance of the image according to the saved camera shutter time, gain parameter, and aperture. When the ambient illuminance is greater than the second set threshold, Then it is considered that the image to be corrected is likely to have a mixed color temperature of natural light. At this time, continue to the subsequent step, that is, determine whether the angle between the first unit direction vector and the pre-saved second unit direction vector is smaller than the set first threshold.

Due to the large difference in color temperature in the natural light mixed color temperature scene, the first coordinate point in the coordinate system is divided into the first cluster and the second cluster. According to the center point of each cluster, if the two center points are close , it is easy to cause a large change in the first unit direction vector due to a small movement of one of the central points, so it is also likely to cause a relatively large change in the angle α between the first unit direction vector and the pre-saved second unit direction vector. Large changes, poor stability. Therefore, before judging whether the included angle between the first unit direction vector and the pre-saved second unit direction vector is smaller than the first set threshold, it is judged whether the distance between the center points of the two clusters is greater than the third set threshold, if If it is greater than, it is judged whether the included angle between the first unit direction vector and the pre-stored second unit direction vector is smaller than the set first threshold.

In order to avoid mistaking a certain color in the image as natural light with a color temperature, the color temperature range of natural light is saved in the electronic device, and the color temperature range of natural light mixed with a pair of determined rectangular areas and the method of constructing a coordinate system are determined. The abscissa range corresponding to the color temperature range of natural light, and the ordinate range, if it is judged that the abscissa of the center points of the first cluster and the second cluster are both within the abscissa range corresponding to the color temperature of natural light, the first cluster and the second cluster When the ordinates of the central points of the two clusters are all within the range of ordinates corresponding to the color temperature of natural light, it is considered that the image to be corrected is likely to have a mixed color temperature of natural light. A step of whether the saved included angle of the second unit direction vector is smaller than the set first threshold.

Before judging whether the angle between the first unit direction vector and the pre-saved second unit direction vector is smaller than the set first threshold, any of the above methods can be used to judge, or a combination of two or three of the above can be used to determine To make a judgment, the process of using the combination of the above three methods to make a judgment is shown in FIG. 5 .

Fig. 5 is a schematic diagram of a process of judging the mixed color temperature of natural light in an image provided by an embodiment of the present invention, and the process includes the following steps:

S501: For the image to be corrected, divide the image into multiple rectangular areas according to the set rules, and determine a pair of natural light mixtures for each rectangular area according to the mean values of red R, green G, and blue B in each rectangular area The value of the color temperature evaluation parameter.

S502: Determine each first coordinate point corresponding to each rectangular area in the constructed coordinate system according to the value of a pair of natural light mixed color temperature evaluation parameters of each rectangular area, wherein the coordinate system is based on the pair of natural light mixed color temperature evaluation parameters as horizontal and vertical axes.

S503: Divide the first coordinate point in the coordinate system into a first cluster and a second cluster, and determine a first unit direction vector according to the density center of each cluster, wherein the first unit direction vector The direction is from low color temperature to high color temperature.

S504: Determine the ambient illuminance of the image according to the saved camera shutter time, gain parameter, and aperture. When the ambient illuminance is greater than the second set threshold, go to step S505; otherwise, go to step S509.

S505: When it is judged that the distance between the center of the first cluster and the center of the second cluster is greater than the third set threshold, go to step S506; otherwise, go to step S509.

S506: Judging that the abscissas of the center points of the first cluster and the second cluster are within the range of abscissas corresponding to the color temperature of natural light, and judging that the ordinates of the center points of the first cluster and the second cluster are corresponding to the color temperature of natural light If it is within the range of the ordinate, go to step S507; otherwise, go to step S509.

S507: Determine whether the included angle between the first unit direction vector and the pre-saved second unit direction vector is smaller than the set first threshold, wherein the second unit direction vector is determined in advance according to the target image with mixed color temperature of natural light and saved, if yes, go to S508; otherwise, go to S509.

S508: Determine that the image to be corrected has a mixed color temperature of natural light.

S509: Determine that the image to be corrected does not have a mixed color temperature of natural light.

Fig. 6 is a schematic structural diagram of an apparatus for judging the mixed color temperature of natural light in an image provided by an embodiment of the present invention, which is applied to electronic equipment, and the apparatus includes:

The first determination module 61 is used for dividing the image into a plurality of rectangular areas according to the set rules for the image to be corrected, and determining each rectangular area according to the mean value of red R, green G, and blue B of each rectangular area. The value of a pair of natural light mixed color temperature evaluation parameters in the area;

The second determination module 62 is configured to determine each first coordinate point corresponding to each rectangular area in the constructed coordinate system according to the value of a pair of natural light mixed color temperature evaluation parameters of each rectangular area, wherein the coordinate system is based on the one The evaluation parameters of the mixed color temperature of natural light are used as the horizontal axis and the vertical axis;

The third determining module 63 is configured to divide the first coordinate points in the coordinate system into first clusters and second clusters, and determine a first unit direction vector according to the center point of each cluster, wherein the The direction of the first unit direction vector is the direction from low color temperature to high color temperature;

The first judging module 64 is used to judge whether the included angle between the first unit direction vector and the pre-saved second unit direction vector is smaller than the set first threshold, wherein the second unit direction vector is based on the mixed color temperature of natural light The target image is pre-determined and saved;

The fourth determination module 65 is used to determine that the image to be corrected has natural light mixed color temperature if the judgment result of the first judgment module is yes, and determine that the image to be corrected does not have natural light if the judgment result of the first judgment module is no Mix color temperature.

The first determination module 61 is also used to divide the target image into a plurality of rectangular areas according to the pre-collected target image of mixed color temperature of natural light according to the set rules, and according to the R, G, B of each rectangular area Mean value, to determine the value of a pair of natural light mixed color temperature evaluation parameters for each rectangular area;

The second determination module 62 is further configured to determine each second coordinate point corresponding to each rectangular area in the constructed coordinate system according to the value of a pair of natural light mixed color temperature evaluation parameters of each rectangular area, wherein the coordinate system Taking the pair of natural light mixed color temperature evaluation parameters as the horizontal axis and the vertical axis;

The third determination module 63 is further configured to determine a second unit direction vector according to the coordinates of each second coordinate point in the coordinate system, wherein the direction of the second unit direction vector is from low color temperature to high color temperature direction.

The third determining module 63 is specifically configured to obtain coordinate information of two coordinate points input by the user, determine direction vectors of the two coordinate points, and use the unit vector corresponding to the direction vector as the second unit direction vector; or according to the least squares method, determine the line equation corresponding to the second coordinate point in the coordinate system, and determine the second unit direction vector corresponding to the line equation according to the line equation; or according to the method of principal component analysis , determine the covariance matrix of the point set composed of the second coordinate points, and determine the second unit direction vector according to the eigenvectors of the covariance matrix.

The device also includes:

The fifth determination module 66 is configured to determine the white balance gain according to the natural light mixed color temperature white balance algorithm and correct the image after it is determined that the image to be corrected has the mixed color temperature of natural light.

The fifth determining module 66 is specifically used to determine the angle and the setting parameters according to the angle between the first unit direction vector and the second unit direction vector and the setting parameters, using a preset algorithm The corresponding natural light mixed color temperature parameter value; determine the first weight and the second weight corresponding to the monochromatic temperature white balance algorithm and the natural light mixed color temperature white balance algorithm according to the natural light mixed color temperature parameter value, according to the monochromatic temperature white balance algorithm and The natural light mixed color temperature white balance algorithm and the corresponding first weight and second weight determine the white balance gain.

The fifth determination module 66 is specifically configured to use the natural light mixing color temperature parameter value as the second weight, and use the difference between 1 and the second weight as the first weight.

The fifth determining module 66 is specifically configured to calculate the product of the included angle and a preset parameter according to the included angle between the first unit direction vector and the pre-saved second unit direction vector; The value is determined as the natural light mixing color temperature parameter value.

The device also includes:

The fifth determining module 66 is configured to determine a white balance gain according to a single color temperature white balance algorithm and correct the image after determining that the image to be corrected does not have a mixed color temperature of natural light.

The device also includes:

The second judging module 67 is used to determine the ambient illuminance of the image according to the saved camera shutter time, gain parameters, and aperture, and when the ambient illuminance is greater than the second set threshold, continue the subsequent steps; or judge the first clustering and When the distance between the center points of the second cluster is greater than the third set threshold, continue to the subsequent steps; or determine that the abscissas of the center points of the first cluster and the second cluster are both within the abscissa range corresponding to the color temperature of natural light , and the ordinates of the central points of the first cluster and the second cluster are both within the range of ordinates corresponding to the color temperature of natural light, the first judging module 64 is triggered.

The first determining module 61 is specifically used to determine the G/R value and G/B value of each rectangular area in each rectangular area; or, determine the logarithmic value of G/R and G/B of each rectangular area or, determine the logarithm value of R/G and the logarithm value of B/G of each rectangular area of each rectangular area.

Embodiments of the present invention provide a method and device for judging the mixed color temperature of natural light in an image. The method includes: for the image to be corrected, divide the image into a plurality of rectangular areas according to set rules, and divide the image into a plurality of rectangular areas according to each rectangular area. Determine the value of a pair of natural light mixed color temperature evaluation parameters for each rectangular area; according to the value of a pair of natural light mixed color temperature evaluation parameters for each rectangular area, determine each in the constructed coordinate system. Each first coordinate point corresponding to a rectangular area, wherein the coordinate system uses the pair of natural light mixed color temperature evaluation parameters as the horizontal axis and the vertical axis; the first coordinate point in the coordinate system is divided into the first cluster and For the second clustering, determine the first unit direction vector according to the center point of each cluster, wherein the direction of the first unit direction vector is the direction from low color temperature to high color temperature; judge the first unit direction vector and the pre-saved first unit direction vector Whether the included angle between the two unit direction vectors is smaller than the set first threshold, wherein the second unit direction vector is predetermined and saved according to the target image with mixed color temperature of natural light; if yes, it is determined that the image to be corrected has natural light Mixed color temperature, otherwise, make sure there is no natural light mixed color temperature in the image to be corrected. Since the second unit direction vector is predetermined and saved for the target image with the mixed color temperature of natural light, when the angle between the first unit direction vector and the second unit direction vector is smaller than the set first threshold, it indicates that the pending The corrected image is similar to the target image that has the mixed color temperature of natural light, and also has the mixed color temperature of natural light, thereby ensuring the accuracy of judging the mixed color temperature of natural light in the image, and further ensuring the accuracy of subsequent white balance correction.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each procedure and/or block in the flowchart and/or block diagram, and a combination of procedures and/or blocks in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions may be provided to a general purpose computer, special purpose computer, embedded processor, or processor of other programmable data processing equipment to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing equipment produce a An apparatus for realizing the functions specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to operate in a specific manner, such that the instructions stored in the computer-readable memory produce an article of manufacture comprising instruction means, the instructions The device realizes the function specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions can also be loaded onto a computer or other programmable data processing device, causing a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process, thereby The instructions provide steps for implementing the functions specified in the flow chart or blocks of the flowchart and/or the block or blocks of the block diagrams.

While preferred embodiments of the invention have been described, additional changes and modifications to these embodiments can be made by those skilled in the art once the basic inventive concept is appreciated. Therefore, it is intended that the appended claims be construed to cover the preferred embodiment as well as all changes and modifications which fall within the scope of the invention.

Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these modifications and variations.

Claims (18)

1. A method for judging the mixed color temperature of natural light in an image, characterized in that the method comprises: For the image to be corrected, divide the image into multiple rectangular areas according to the set rules, and determine a pair of natural light mixed color temperature evaluations for each rectangular area according to the mean values of red R, green G, and blue B in each rectangular area the value of the parameter; Each first coordinate point corresponding to each rectangular area is determined in the constructed coordinate system according to the value of a pair of natural light mixed color temperature evaluation parameters of each rectangular area, wherein the coordinate system uses the pair of natural light mixed color temperature evaluation parameters as the abscissa axis and vertical axis; Divide the first coordinate point in the coordinate system into a first cluster and a second cluster, and determine a first unit direction vector according to the center point of each cluster, wherein the direction of the first unit direction vector is From low color temperature to high color temperature; judging whether the included angle between the first unit direction vector and the pre-saved second unit direction vector is smaller than the set first threshold, wherein the second unit direction vector is predetermined and saved according to the target image with mixed color temperature of natural light ; If so, determine that the image to be corrected has a natural light mixed color temperature, otherwise, determine that the image to be corrected does not have a natural light mixed color temperature; The process of determining the second unit direction vector includes: For the pre-collected target image of mixed color temperature of natural light, the target image is divided into multiple rectangular areas according to the set rules, and a pair of natural light in each rectangular area is determined according to the R, G, and B mean values of each rectangular area. The value of the mixed color temperature evaluation parameter; According to the value of a pair of natural light mixed color temperature evaluation parameters of each rectangular area, each second coordinate point corresponding to each rectangular area is determined in the constructed coordinate system, wherein the coordinate system uses the pair of natural light mixed color temperature evaluation parameters as horizontal and vertical axes; According to the coordinates of each second coordinate point in the coordinate system, a second unit direction vector is determined, wherein the direction of the second unit direction vector is a direction from low color temperature to high color temperature. 2. The method according to claim 1, wherein said determining the second unit direction vector according to the coordinates of each second coordinate point in the coordinate system comprises: Acquire coordinate information of two coordinate points input by the user, determine direction vectors of the two coordinate points, and use the unit vector corresponding to the direction vector as the second unit direction vector; or According to the method of least squares, determine the equation of a line corresponding to the second coordinate point in the coordinate system, and determine the second unit direction vector corresponding to the equation of the line according to the equation of the line; or According to the principal component analysis method, the covariance matrix of the point set composed of the second coordinate points is determined, and the second unit direction vector is determined according to the eigenvectors of the covariance matrix. 3. The method according to claim 1, characterized in that, after determining that the image to be corrected has a mixed color temperature of natural light, the method further comprises: The white balance gain is determined according to the natural light mixed color temperature white balance algorithm and the image is corrected. 4. The method according to claim 3, wherein the determining the white balance gain according to the natural light mixed color temperature white balance algorithm comprises: According to the angle between the first unit direction vector and the second unit direction vector and the setting parameters, using a preset algorithm to determine the natural light mixing color temperature parameter value corresponding to the angle and the setting parameters; According to the natural light mixed color temperature parameter value, determine the first weight and the second weight corresponding to the single color temperature white balance algorithm and the natural light mixed color temperature white balance algorithm, according to the single color temperature white balance algorithm and the natural light mixed color temperature white balance algorithm and The white balance gain is determined corresponding to the first weight and the second weight. 5. The method according to claim 4, wherein, according to the natural light mixed color temperature parameter value, determining the first weight and the second weight corresponding to the single color temperature white balance algorithm and the natural light mixed color temperature white balance algorithm include : The value of the natural light mixed color temperature parameter is used as a second weight, and the difference between 1 and the second weight is used as a first weight. 6. The method according to claim 4, characterized in that, according to the included angle between the first unit direction vector and the pre-saved second unit direction vector and setting parameters, a preset algorithm is used to determine the The above-mentioned included angle and the corresponding natural light mixing color temperature parameter value of the above-mentioned setting parameters include: According to the included angle between the first unit direction vector and the pre-stored second unit direction vector, calculate the product of the included angle and a preset parameter; determine the cosine value of the product as the natural light mixing color temperature parameter value. 7. The method according to claim 1, wherein after determining that the image to be corrected does not have a mixed color temperature of natural light, the method further comprises: The white balance gain is determined according to the monochromatic temperature white balance algorithm and the image is corrected. 8. The method according to claim 1, wherein before the judgment of whether the included angle between the first unit direction vector and the pre-saved second unit direction vector is smaller than the set first threshold, the method further comprises : Determine the ambient illuminance of the image according to the stored camera shutter time, gain parameters, and aperture, and when the ambient illuminance is greater than the second set threshold, continue to the subsequent steps; or When it is judged that the distance between the center points of the first cluster and the second cluster is greater than the third set threshold, continue to the subsequent steps; or Judging that the abscissas of the center points of the first cluster and the second cluster are both within the range of abscissas corresponding to the color temperature of natural light, and the ordinates of the center points of the first cluster and the second cluster are both within the range corresponding to the color temperature of natural light If it is within the range of the ordinate, continue to the next step. 9. The method according to claim 1, wherein the values of a pair of natural light mixed color temperature evaluation parameters for each rectangular area include: G/R value and G/B value for each rectangular area; or the logarithmic values of G/R and the logarithmic values of G/B for each rectangular area; or The logarithmic value of R/G and the logarithmic value of B/G of each rectangular area. 10. A device for judging the mixed color temperature of natural light in an image, characterized in that the device comprises: The first determination module is used to divide the image to be corrected into a plurality of rectangular areas according to the set rules, and determine each rectangular area according to the mean value of red R, green G, and blue B of each rectangular area. The value of a pair of natural light mixed color temperature evaluation parameters; The second determination module is used to determine each first coordinate point corresponding to each rectangular area in the constructed coordinate system according to the value of a pair of natural light mixed color temperature evaluation parameters of each rectangular area, wherein the coordinate system is based on the pair The natural light mixed color temperature evaluation parameters are used as the horizontal axis and the vertical axis; The third determination module is used to divide the first coordinate points in the coordinate system into the first cluster and the second cluster, and determine the first unit direction vector according to the center point of each cluster, wherein the first unit direction vector The direction of a unit direction vector is the direction from low color temperature to high color temperature; The first judging module is used to judge whether the included angle between the first unit direction vector and the pre-saved second unit direction vector is smaller than the set first threshold, wherein the second unit direction vector is based on the mixed color temperature of natural light The target image is predetermined and saved; The fourth determination module is used to determine that the image to be corrected has a natural light mixed color temperature if the judgment result of the first judgment module is Yes, and if the judgment result of the first judgment module is No, determine that the image to be corrected does not have a natural light mixed color temperature; The first determining module is also used to divide the target image into a plurality of rectangular areas according to the pre-collected target image of mixed color temperature of natural light according to the set rules, and according to the R, G, and B mean values of each rectangular area , to determine the value of a pair of natural light mixed color temperature evaluation parameters for each rectangular area; The second determination module is further configured to determine each second coordinate point corresponding to each rectangular area in the constructed coordinate system according to the value of a pair of natural light mixed color temperature evaluation parameters of each rectangular area, wherein the coordinate system is represented by The pair of natural light mixed color temperature evaluation parameters are used as the horizontal axis and the vertical axis; The third determining module is further configured to determine a second unit direction vector according to the coordinates of each second coordinate point in the coordinate system, wherein the direction of the second unit direction vector is the direction from low color temperature to high color temperature . 11. The device according to claim 10, wherein the third determining module is specifically configured to obtain coordinate information of two coordinate points input by the user, determine the direction vectors of the two coordinate points, and set The unit vector corresponding to the direction vector is used as the second unit direction vector; or, according to the least squares method, determine the straight line equation corresponding to the second coordinate point in the coordinate system, and determine the straight line equation according to the straight line equation The corresponding second unit direction vector; or, according to the principal component analysis method, determine the covariance matrix of the point set composed of the second coordinate points, and determine the second unit direction vector according to the eigenvectors of the covariance matrix. 12. The device according to claim 10, further comprising: The fifth determination module is configured to determine the white balance gain according to the natural light mixed color temperature white balance algorithm and correct the image after it is determined that the image to be corrected has the mixed color temperature of natural light. 13. The device according to claim 12, wherein the fifth determining module is specifically configured to adopt a preset According to the algorithm, determine the angle and the natural light mixed color temperature parameter value corresponding to the set parameter; determine the first weight sum corresponding to the single color temperature white balance algorithm and the natural light mixed color temperature white balance algorithm according to the natural light mixed color temperature parameter value The second weight is to determine the white balance gain according to the single color temperature white balance algorithm and the natural light mixed color temperature white balance algorithm and the corresponding first weight and second weight. 14. The device according to claim 13, wherein the fifth determination module is specifically configured to use the natural light mixed color temperature parameter value as the second weight, and use the difference between 1 and the second weight as the second weight. a weight. 15. The device according to claim 13, wherein the fifth determination module is specifically configured to calculate the included angle according to the included angle between the first unit direction vector and the pre-saved second unit direction vector. The product of the angle and the preset parameter; the cosine value of the product is determined as the parameter value of the natural light mixing color temperature. 16. The device according to claim 10, further comprising: The fifth determination module is configured to determine the white balance gain according to the single color temperature white balance algorithm and correct the image after determining that there is no natural light mixed color temperature in the image to be corrected. 17. The device according to claim 10, further comprising: The second judging module is used to determine the ambient illuminance of the image according to the saved camera shutter time, gain parameters, and aperture, and when the ambient illuminance is greater than the second set threshold, continue to the subsequent steps; or judge the first cluster and the second When the distance between the center points of the two clusters is greater than the third set threshold, continue to the subsequent steps; or judge that the abscissas of the center points of the first cluster and the second cluster are both within the range of abscissas corresponding to the color temperature of natural light , and when the ordinates of the center points of the first cluster and the second cluster are both within the range of ordinates corresponding to the color temperature of natural light, the first judging module is triggered. 18. The device according to claim 10, wherein the first determination module is specifically configured to determine the G/R value and the G/B value of each rectangular area of each rectangular area; or, determine each The logarithmic value of G/R and the logarithmic value of G/B of each rectangular area; or, determine the logarithmic value of R/G and the logarithmic value of B/G of each rectangular area of each rectangular area.