WO2019041493A1

WO2019041493A1 - White balance adjustment method and device

Info

Publication number: WO2019041493A1
Application number: PCT/CN2017/107333
Authority: WO
Inventors: 袁全
Original assignee: 广东欧珀移动通信有限公司
Priority date: 2017-08-31
Filing date: 2017-10-23
Publication date: 2019-03-07

Abstract

The present invention provides a white balance adjustment method and device. The method comprises: calculating a first white balance gain value of an image by means of a human face white balance algorithm; calculating a plurality of second white balance gain values separately corresponding to the image if the image is separately obtained by imaging under a plurality of light sources; according to the first white balance gain value, selecting, from the plurality of second white balance gain values, a target white balance gain value close to the first white balance gain value; and carrying out white balance adjustment on the image according to the target white balance gain value. The present invention resolves the technical problem in the prior art of sudden change in a white balance gain value when there is a human face or there is no human face in the image the same scene.

Description

White balance adjustment method and device

Cross-reference to related applications

This application claims the Chinese patent application number “201710776063.X” submitted by Guangdong Opal Mobile Communication Co., Ltd. on August 31, 2017, and the invention name is “White Balance Adjustment Method and Device” and on August 31, 2017. The Chinese patent application number "201710775995.2", which is submitted under the name "White Balance Adjustment Method and Apparatus", has priority.

Technical field

The present invention relates to the field of imaging technologies, and in particular, to a white balance adjustment method and apparatus.

Background technique

In the related art, when the photographing device using the terminal device performs photographing, the color value obtained by the actual color image collection and the real color of the object may be deviated, and the cause of the deviation is mainly two, one is the color temperature of the light source environment. Change, different color temperature conditions, the same object's reflection spectrum is not the same, resulting in different colors of objects under different color temperature illumination, such as white objects appear blue in high color temperature environment, and in low color temperature environment In red. The other is due to the deviation of the gain of the color channel inherent in the shooting device itself. For example, the manual gain value of the B channel of GC0307 is 0x98, and the manual gain value of the R, G channel is 0x80.

Therefore, in the related art, in order to compensate for such color deviation, the white balance gain value of the color gain channel of the photographing device is changed by the related white balance algorithm, the color deviation caused by the color temperature environment, and the color channel inherent to the photographing device itself. The deviation of the gain is uniformly compensated, so that the acquired image can correctly reflect the true color of the object.

Among them, there are various white balance algorithms, which can be used to calculate the white balance gain value. In the scene shooting, in order to achieve better processing effect, different white balance algorithms are used for whitening based on human face and no face. The balance processing results in a significant change in the white balance gain value when the face is photographed and there is no face in the same scene when taking a picture, resulting in a sudden change in image color.

Summary of the invention

The invention provides a white balance adjustment method and device, which solves the technical problem that the white balance gain value is abrupt when the human face and the no-face are in the image in the same scene in the same scene, thereby causing color mutation.

An embodiment of the present invention provides a white balance adjustment method, including the following steps: using a face white balance algorithm to calculate a first white balance gain value of an image; and calculating if the image is obtained by imaging under multiple light sources, respectively Image Corresponding plurality of second white balance gain values respectively; selecting, according to the first white balance gain value, a target white that is close to the first white balance gain value from the plurality of second white balance gain values Balancing the gain value; using the target white balance gain value to perform white balance adjustment on the image.

Another embodiment of the present invention provides a white balance adjustment apparatus, including: a first calculation module, configured to calculate a first white balance gain value of an image by using a human face white balance algorithm; and a second calculation module, configured to calculate respectively a plurality of second white balance gain values respectively corresponding to the image when the image is obtained by using a plurality of light sources; and a first selecting module, configured to: according to the first white balance gain value, from the plurality of Selecting a second white balance gain value to obtain a target white balance gain value that is close to the first white balance gain value; and a first adjustment module, configured to perform white balance adjustment on the image by using the target white balance gain value .

A further embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores computer readable instructions, and when the instructions are executed by the processor, the processor performs the above implementation of the present invention. The white balance adjustment method described in the example.

Still another embodiment of the present invention provides a non-transitory computer readable storage medium having stored thereon a computer program that, when executed by a processor, implements a white balance adjustment method as described in the above embodiments of the present invention.

An embodiment of the present invention provides a white balance adjustment method, including the following steps: using a human face white balance algorithm to calculate a third white balance gain value of an image; and calculating if the image is obtained by imaging under multiple light sources, respectively a plurality of fourth white balance gain values corresponding to the image; and selecting, from the plurality of fourth white balance gain values, a target close to the third white balance gain value according to the third white balance gain value a white balance gain value; a fifth gray balance gain value of the image is calculated using a simple gray world algorithm; and the target balance gain value target white balance gain value and the fifth white balance gain value are The image is white balance adjusted.

Another embodiment of the present invention provides a white balance adjustment apparatus, including: a third calculation module, configured to calculate a third white balance gain value of an image by using a human face white balance algorithm; and a fourth calculation module, configured to calculate And a plurality of fourth white balance gain values corresponding to the image when the image is obtained by using a plurality of light sources, and a second selecting module, configured to: according to the third white balance gain value, from the plurality of Selecting a fourth white balance gain value to obtain a target white balance gain value close to the third white balance gain value; and a fifth calculation module, configured to calculate a fifth white balance of the image by using a simple gray world algorithm a gain value; a second adjustment module, configured to perform white balance adjustment on the image according to the target white balance gain value and the fifth white balance gain value.

The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

It suppresses the problem that the white balance gain value abruptly causes the screen to flicker when the face and the face are not in the same scene, and the damage to the human eye is avoided.

DRAWINGS

The above and/or additional aspects and advantages of the present invention will become apparent and readily understood from

1 is a flow chart of a white balance adjustment method according to an embodiment of the present invention;

2 is a flow chart of a white balance adjustment method according to another embodiment of the present invention;

3 is a schematic structural view of a white balance adjusting device according to an embodiment of the present invention;

4 is a schematic structural view of a white balance adjusting device according to another embodiment of the present invention;

FIG. 5 is a schematic structural diagram of a white balance adjusting apparatus according to still another embodiment of the present invention; FIG.

6 is a schematic structural diagram of an image processing circuit in a computer device according to an embodiment of the present invention;

7 is a flow chart of a white balance adjustment method according to an embodiment of the present invention;

FIG. 8 is a flowchart of a white balance adjustment method according to another embodiment of the present invention; FIG.

9 is a schematic structural view of a white balance adjusting device according to an embodiment of the present invention.

Detailed ways

The embodiments of the present invention are described in detail below, and the examples of the embodiments are illustrated in the drawings, wherein the same or similar reference numerals are used to refer to the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the drawings are intended to be illustrative of the invention and are not to be construed as limiting.

It can be understood that in many application scenarios in a practical application, a user performs a photograph using an application in a terminal device such as a smart phone, wherein a photograph is taken in a portrait mode such as a front photographing mode or a portrait photographing mode of the terminal device, and When the photograph is taken in a non-portrait photographing mode such as a rear photographing mode, the white balance algorithm used is different because the color composition of the image is different in the face photographing mode and the non-face mode. Specifically, the Face Automatic White Balance (FaceAWB) algorithm is used in the portrait photographing mode. When there is a person in the image, the skin color change of one type of race is small, within an estimable range. Therefore, the corresponding correction operator can be determined according to the characteristics of the skin color of the face, thereby obtaining a more accurate white balance calculation result. Especially in large-area solid background and/or mixed light conditions, the white balance effect of the image can be effectively improved.

In the non-portrait camera mode, the Simple Gray World algorithm is used. The Simple Gray World algorithm is based on the gray world assumption. The assumption is that for an image with a large number of color changes, red The average values of the saturations of the three components (Red, R), green (Green), and blue (Blue, B) tend to be the same. A gray value. That is, the gray world algorithm assumes that the mean value of the average reflection of the natural scene for the light is generally constant, and the saturation of the three components R, G, and B tends to be uniform. When there are rich colors in the image, the image is processed by the gray world algorithm, and the influence of the ambient light can be better eliminated.

However, the white balance gain values obtained by using different white balance algorithms have a large difference. In the same application scenario, when the terminal device moves from a scene with a human face to a scene without a human face, the obtained white balance gain is obtained. The value difference is large, which leads to a sudden change in color, which is harmful to the human eye and has a poor visual effect.

In order to solve the above technical problem, the present invention proposes a white balance adjustment method and apparatus which can suppress the problem of sudden change of the white balance gain value.

A white balance adjustment method and apparatus according to an embodiment of the present invention will be described below with reference to the drawings.

1 is a flow chart of a white balance adjustment method according to an embodiment of the present invention. As shown in FIG. 1, the method includes the following steps:

In step 101, a first white balance gain value of the image is calculated by using a face white balance algorithm.

Specifically, in order to select a suitable gain value for the face image to perform white balance processing on the image, the color and the skin color of the face are consistent in the image processing result, and the image may be first calculated according to the face white balance algorithm. The gain value is used as a backup.

Step 102: Calculate a plurality of second white balance gain values corresponding to the images respectively when the images are imaged under a plurality of light sources.

The light source comprises: one or more combinations of a sunlight source, a fluorescent light source, a tungsten light source and a FAH light source, wherein the FAH light source is a light source between the A light and the H light, and the A light color temperature is 2850K, and the H light color temperature is It is 2350K.

Specifically, in order to select a suitable gain value for the non-face image to perform white balance processing on the image, the color of the non-face region is consistent with the natural color in the image processing result, and the image is obtained by imaging under various light sources respectively. When the images respectively correspond to a plurality of second white balance gain values for standby, the second white balance gain values are closer to the algorithm results of the grayscale world.

Step 103: Select, according to the first white balance gain value, a target white balance gain value that is close to the first white balance gain value from the plurality of second white balance gain values.

In step 104, white balance adjustment is performed on the image by using the target white balance gain value.

Specifically, in order to avoid a large difference in white balance gain values in the face photographing mode and the non-face photographing mode, the first white balance gain value is selected from the plurality of second white balance gain values and the first white Balancing the target white balance gain value close to the gain value, thereby adjusting the white balance of the image according to the target white balance gain value, on the one hand, considering the skin color of the face, and on the other hand, considering the rich color of the natural world (the grayscale world), Not only can the visual effect of image processing be improved, but also the target white balance gain value is closer to the white balance gain value of the grayscale world, avoiding the problem that the white balance gain value is abrupt and causing screen flicker.

It should be noted that, according to different application scenarios, different implementation manners may be adopted, and according to the first white balance gain value, a target white balance close to the first white balance gain value is selected from the plurality of second white balance gain values. The gain value is illustrated as follows:

In a first example, determining a first gain value of each color component in the first white balance gain value, and determining, for each second white balance gain value, a second gain value of each color component, for each second white balance Calculating the difference value between the gain value and the first white balance gain value, wherein the difference value is obtained by calculating the absolute difference between the first gain value and the second gain value in the same color component, and determining the absolute difference of each color component And obtaining, from among the plurality of second white balance gain values, selecting a target white balance gain value that is the smallest difference value from the first gain value.

In a second example, a first vector is generated according to a first gain value of the first white balance gain value on each color component, and a second gain value is generated according to each second white balance gain value on each color component. a plurality of second vectors, calculating a vector distance between the first vector and each of the second vectors, the vector distance including a Euclidean distance, and selecting a vector from the plurality of second white balance gain values according to the vector distance The minimum target white balance gain value.

In summary, the white balance adjustment method of the embodiment of the present invention uses the face white balance algorithm to calculate the first white balance gain value of the image, and if the images are respectively imaged under a plurality of light sources, the images respectively correspond to the images. a plurality of second white balance gain values, according to the first white balance gain value, selecting a target white balance gain value close to the first white balance gain value from the plurality of second white balance gain values, using the target white balance gain Value, white balance adjustment of the image. Thereby, the problem that the white balance gain value is abruptly caused by the sudden change of the white balance gain value when the face and the face are not in the same scene is suppressed, and the damage to the human eye is avoided.

Based on the above embodiment, for further detailed description, how to obtain a target white balance gain value close to the first white balance gain value from the plurality of second white balance gain values according to the first white balance gain value, The second example shows a gain-based vector determining target white balance gain value as an example for explanation.

2 is a flow chart of a white balance adjustment method according to another embodiment of the present invention. As shown in FIG. 2, the method includes:

In step 201, the first white balance gain value of the image is calculated by using a face white balance algorithm.

Specifically, the face recognition technology may be used to perform face recognition on the image to determine that the image includes a face region. For example, the face recognition method may be used to identify the face in the image to obtain a face region. Coordinate interval, wherein the face recognition algorithm has many implementations in the prior art, for example, the Adaboost model algorithm is used for face recognition, and other algorithms capable of expressly identifying the face region can be used to perform the face region. Identification. The implementation manner of the face recognition is not limited in this embodiment.

After getting the face area, the skin color of a group of people changes little. For example, according to statistics, after the skin color RGB color space is converted to the YCbCr space, the Cb and Cr ranges of the face are [133, 173], [77, 127], respectively. That is, as long as the human skin color range can be determined, the image can be corrected according to the skin color range. Therefore, you can compare the faces in the image The color of the area and the preset skin color range are used to calculate the first gain value of the image.

Of course, the purpose of the foregoing embodiment for performing face recognition to determine the first gain value is to obtain a first gain value when performing white balance processing based on skin color in the face photographing mode, in fact, in the front camera photographing mode or in In the camera mode of the rear camera, the white balance processing is performed based on the face white balance algorithm. Therefore, it is also possible to determine the first white balance gain of the image by using the face white balance algorithm when the image is imaged by the front camera. Value, or, when determining that the image is imaged by the portrait mode of the rear camera, the first white balance gain value of the image is calculated by using the face white balance algorithm.

Step 202: Calculate a plurality of second white balance gain values corresponding to the images respectively when the images are imaged under a plurality of light sources.

Specifically, the white balance processing is performed by using a gray world algorithm under various light sources respectively, wherein the gray world algorithm is based on the assumption that for an image with a large number of color changes, three components of R, G, and B. The average of the saturations tends to the same gray value G. In practical applications, there are usually two ways to determine the gray value G.

As a possible implementation, a fixed value can be taken. For example, one half of the brightest gray value may be taken, that is, when the brightest gray value is 255, the gray value G may be 128. As another possible implementation manner, the average value of the three average values of R, G, and B in the image may be calculated as the gray value G. After determining the gradation value G, the second gain value of the image under the corresponding light source can be calculated by comparing the gradation value G with the average values of the three colors of R, G, and B, respectively.

Step 203: Generate a first vector according to the first gain value of the first white balance gain value on each color component.

Step 204: Generate a corresponding plurality of second vectors according to the second gain value of each second white balance gain value on each color component.

In practical applications, the first gain value and the second gain value can be accurately characterized using vectors in the color space. The color space can be varied, for example: RGB (red, green, blue) color space, that is, based on the color space of the device's three primary colors. In addition, it may also be an HSI color space, which is derived from a human visual system, and describes colors by Hue, Saturation or Chroma, and Intensity or Brightness. The HSI color space can be described by a conical space model. Certainly, other color spaces may be used for description, which are not described in this embodiment. As a possible implementation, the first gain value and the second gain value may be characterized by an RGB model in the color space.

Specifically, in the RGB model, each color appears in three color components of R, G, and B. This model is based on a Cartesian coordinate system, and the color space considered is a cube. One vertex of the cube can be used as the origin, black is at the origin, and white is at the vertex that is farthest from the origin in the cube. In this model, different colors are on the cube or inside the cube and can be characterized by a vector distributed from the origin.

As a possible implementation, assuming that all colors are normalized, the cube is a unit cube The body, that is, all R, G, and B values are in the range of [0, 1]. Therefore, the values of the first gain value and the second gain value on each of the R, G, and B color components may also take values in the range of [0, 1]. By combining the values of the first gain value on each color component, a first vector can be generated, and the values of the second gain value on each color component can be combined to generate a second vector. For example, if the first gain value has a value of 0.1 on the R component, a value of 0.2 on the G component, and a value of 0.3 on the B component, it can be on each color component according to the first gain value. The value of the first vector [0.1, 0.2, 0.3] is generated. If the second gain value has a value of 0.2 on the R component, a value of 0.2 on the G component, and a value of 0.2 on the B component, the second gain value can be taken on each color component. The value produces the first vector [0.2, 0.2, 0.2].

Step 205: Calculate a vector distance between the first vector and each of the second vectors, wherein the vector distance includes a Euclidean distance.

Specifically, after generating the first vector and the second vector, a quantized representation of the first gain value and the second gain value is achieved. When calculating the vector distance between the first vector and the second vector, the Euclidean distance can be used to describe the vector distance between the two vectors, or the cosine distance and the Pearson correlation coefficient can be used to describe the two vectors. Vector distance between vectors. Taking the vector distance between the first vector and the second vector using the Euclidean distance as an example, the following Euclidean distance formula can be used:

A vector distance between the first vector and the second vector is calculated. Where d(x, y) is the vector distance between the first vector and the second vector, and x _R , x _G , and x _B are the values on each color component in the first vector, respectively, y _R , y _G And y _B are the values on each color component in the second vector, respectively.

Furthermore, after calculating the vector distance between the first vector and the second vector, the vector distance between the first vector and the second vector can be determined, and the vector distance between the first vector and the second vector is more Large, it can be determined that the first gain value and the second gain value are less similar. If the vector distance between the first vector and the second vector is closer, it may be determined that the first gain value and the second gain value are similar.

Step 206: Select, according to the vector distance, a target white balance gain value that is the smallest vector distance from the plurality of second white balance gain values.

In step 207, the white balance adjustment is performed on the image by using the target white balance gain value.

Specifically, according to the vector distance, the target white balance gain value with the smallest vector distance is selected from the plurality of second white balance gain values, and the target white balance gain value calculated by the gray world algorithm is used to accurately white balance the image. Processing, on the one hand, considers the face color, on the other hand, considers the rich colors of nature, not only can improve the visual effect of image processing, but also based on the unified target white balance gain value for image processing, avoiding the white balance gain value mutation The problem that caused the screen to flicker.

In summary, the white balance adjustment method according to the embodiment of the present invention is based on adjusting a face in an image to a skin color. a face white balance algorithm, calculating a first gain value for the image, and calculating a plurality of second white balance gain values respectively corresponding to the image if the images are imaged under a plurality of light sources, respectively, according to the first white balance gain Generating, by the first gain value on each color component, a first vector, generating a corresponding plurality of second vectors according to a second gain value of each second white balance gain value on each color component, and calculating the first a vector distance between the vector and each of the second vectors; the vector distance includes a Euclidean distance, and from the plurality of second white balance gain values, the target white balance gain value having the smallest vector distance is selected according to the vector distance, and further The target white balance gain value is used to adjust the white balance of the image, so that the slow white balance convergence speed can effectively improve the problem of flickering when the white balance is adjusted by the white balance algorithm.

In order to achieve the above embodiment, the present invention also provides a white balance adjusting device. FIG. 3 is a schematic structural view of a white balance adjusting device according to an embodiment of the present invention. As shown in FIG. 3, the white balance adjusting device includes a first calculation. The module 100, the second calculation module 200, the first selection module 300, and the first adjustment module 400.

The first calculation module 100 is configured to calculate a first white balance gain value of the image by using a human face white balance algorithm.

The second calculating module 200 is configured to calculate a plurality of second white balance gain values respectively corresponding to the images when the images are respectively imaged under a plurality of light sources.

The first selection module 300 is configured to select, from the plurality of second white balance gain values, a target white balance gain value that is close to the first white balance gain value according to the first white balance gain value.

The first adjustment module 400 is configured to perform white balance adjustment on the image by using the target white balance gain value.

4 is a schematic structural diagram of a white balance adjusting apparatus according to another embodiment of the present invention. As shown in FIG. 4, the first selecting module 300 includes a first calculation. Unit 310 and first selection unit 320.

The first calculating unit 310 is configured to determine a first gain value of each color component in the first white balance gain value, and determine a second gain value of each color component for each second white balance gain value; Calculating a difference value between a second white balance gain value and a first white balance gain value, wherein the difference value is an absolute difference between the first gain value and the second gain value in the same color component, and each color component is calculated The absolute difference is summed.

The first selecting unit 320 is configured to select, from among the plurality of second white balance gain values, a target white balance gain value that is the smallest difference value from the first gain value.

Based on the above embodiment, FIG. 5 is a schematic structural diagram of a white balance adjusting apparatus according to still another embodiment of the present invention. As shown in FIG. 5, the first selecting module 300 includes a second calculating, as shown in FIG. Unit 330 and second selection unit 340.

The second calculating unit 330 is configured to determine, according to the first white balance gain value, a first gain value on each color component. Generating a first vector; generating a corresponding plurality of second vectors according to a second gain value of each second white balance gain value on each color component; calculating a vector distance between the first vector and each of the second vectors.

The second selecting unit 340 is configured to select, according to the vector distance, a target white balance gain value that is the smallest vector distance from the plurality of second white balance gain values.

It should be noted that the foregoing description of the method embodiments is also applicable to the device in the embodiment of the present invention, and the implementation principle is similar, and details are not described herein again.

The division of each module in the white balance adjustment device is for illustrative purposes only. In other embodiments, the white balance adjustment device may be divided into different modules as needed to complete all or part of the functions of the white balance adjustment device.

In summary, the white balance adjusting device of the embodiment of the present invention uses the face white balance algorithm to calculate the first white balance gain value of the image, and if the images are respectively imaged under a plurality of light sources, the images respectively correspond to the images. a plurality of second white balance gain values, according to the first white balance gain value, selecting a target white balance gain value close to the first white balance gain value from the plurality of second white balance gain values, using the target white balance gain Value, white balance adjustment of the image. Thereby, the problem that the white balance gain value is abruptly caused by the sudden change of the white balance gain value when the face and the face are not in the same scene is suppressed, and the damage to the human eye is avoided.

To achieve the above objective, an embodiment of the present invention further provides a computer device. The above computer device includes an image processing circuit, and the image processing circuit may be implemented by hardware and/or software components, and may include various processing units defining an ISP (Image Signal Processing) pipeline. Figure 6 is a schematic illustration of an image processing circuit in one embodiment. As shown in FIG. 6, for convenience of explanation, only various aspects of the image processing technique related to the embodiment of the present invention are shown.

As shown in FIG. 6, the image processing circuit includes an ISP processor 1040 and a control logic 1050. The image data captured by imaging device 1010 is first processed by ISP processor 1040, which analyzes the image data to capture image statistical information that may be used to determine and/or control one or more control parameters of imaging device 1010. Imaging device 1010 can include a camera having one or more lenses 1012 and image sensors 1014. Image sensor 1014 may include a color filter array (such as a Bayer filter) that may acquire light intensity and wavelength information captured with each imaging pixel of image sensor 1014 and provide a set of primitives that may be processed by ISP processor 1040 Image data. Sensor 1020 can provide raw image data to ISP processor 1040 based on sensor 1020 interface type. The sensor 1020 interface may utilize a SMIA (Standard Mobile Imaging Architecture) interface, other serial or parallel camera interfaces, or a combination of the above.

The ISP processor 1040 processes the raw image data pixel by pixel in a variety of formats. For example, each image pixel can have a bit depth of 8, 10, 12, or 14 bits, and the ISP processor 1040 can perform one or more image processing operations on the raw image data, collecting statistical information about the image data. Among them, image processing operations can be performed with the same or different bit depth precision.

ISP processor 1040 can also receive pixel data from image memory 1030. For example, the interface from sensor 1020 will The raw pixel data is sent to image memory 1030, and the raw pixel data in image memory 1030 is then provided to ISP processor 1040 for processing. Image memory 1030 can be part of a memory device, a storage device, or a separate dedicated memory within an electronic device, and can include DMA (Direct Memory Access) features.

When receiving raw image data from sensor 1020 interface or from image memory 1030, ISP processor 1040 can perform one or more image processing operations, such as time domain filtering. The processed image data can be sent to image memory 1030 for additional processing prior to being displayed. The ISP processor 1040 receives the processed data from the image memory 1030 and performs image data processing in the original domain and in the RGB and YCbCr color spaces. The processed image data may be output to display 1070 for viewing by a user and/or further processed by a graphics engine or a GPU (Graphics Processing Unit). Additionally, the output of ISP processor 1040 can also be sent to image memory 1030, and display 1070 can read image data from image memory 1030. In one embodiment, image memory 1030 can be configured to implement one or more frame buffers. Additionally, the output of ISP processor 1040 can be sent to encoder/decoder 1060 to encode/decode image data. The encoded image data can be saved and decompressed before being displayed on the display 1070 device. Encoder/decoder 1060 can be implemented by a CPU or GPU or coprocessor.

The statistics determined by the ISP processor 1040 can be sent to the control logic 1050 unit. For example, the statistical data may include image sensor 1014 statistical information such as auto exposure, auto white balance, auto focus, flicker detection, black level compensation, lens 1012 shading correction, and the like. Control logic 1050 can include a processor and/or a microcontroller that executes one or more routines, such as firmware, and one or more routines can determine control parameters and control of imaging device 1010 based on received statistical data. parameter. For example, the control parameters may include sensor 1020 control parameters (eg, gain, integration time for exposure control), camera flash control parameters, lens 1012 control parameters (eg, focus or zoom focal length), or a combination of these parameters. The ISP control parameters may include gain levels and color correction matrices for automatic white balance and color adjustment (eg, during RGB processing), as well as lens 1012 shading correction parameters.

The following are the steps to implement the white balance adjustment method using the image processing technique in FIG. 6:

In step 101', the first white balance gain value of the image is calculated by using a face white balance algorithm.

Step 102' calculates a plurality of second white balance gain values corresponding to the images respectively when the images are imaged under a plurality of light sources.

Step 103', selecting, according to the first white balance gain value, a target white balance gain value that is close to the first white balance gain value from the plurality of second white balance gain values.

Step 104', using the target white balance gain value, to perform white balance adjustment on the image.

It should be noted that the foregoing description of the method embodiment is also applicable to the terminal device in this embodiment, and the implementation principle is similar, and details are not described herein again.

In summary, the terminal device in the embodiment of the present invention uses the face white balance algorithm to calculate the first white of the image. Balancing the gain value, and calculating a plurality of second white balance gain values respectively corresponding to the image when the images are imaged under a plurality of light sources respectively, and selecting from the plurality of second white balance gain values according to the first white balance gain value A target white balance gain value close to the first white balance gain value is obtained, and the image is subjected to white balance adjustment using the target white balance gain value. Thereby, the problem that the white balance gain value is abruptly caused by the sudden change of the white balance gain value when the face and the face are not in the same scene is suppressed, and the damage to the human eye is avoided.

The embodiment of the present invention further provides a non-transitory computer readable storage medium having stored thereon a computer program capable of implementing the white balance adjustment method as described in the foregoing embodiments when the computer program is executed by the processor.

In order to solve the above technical problem, the present invention also proposes a white balance adjustment method and apparatus, which can suppress the problem of sudden change of the white balance gain value.

FIG. 7 is a flowchart of a white balance adjustment method according to an embodiment of the present invention. As shown in FIG. 7, the method includes the following steps:

In step 301, a third white balance gain value of the image is calculated by using a face white balance algorithm.

Specifically, in order to select a suitable gain value for the face image to perform white balance processing on the image, the color of the face and the skin color are consistent in the image processing result, and the image may be calculated according to the face white balance algorithm. The gain value is used as a backup.

Step 302: Calculate a plurality of fourth white balance gain values corresponding to the image when the images are imaged under a plurality of light sources, respectively.

Specifically, in order to select a suitable gain value for the non-face image to perform white balance processing on the image, the color of the non-face region is consistent with the natural color in the image processing result, and the image is obtained by imaging under various light sources respectively. When the images respectively correspond to a plurality of fourth white balance gain values for standby, the fourth white balance gain value is closer to the algorithm result of the grayscale world.

Step 303: Select, according to the third white balance gain value, a target white balance gain value that is close to the third white balance gain value from the plurality of fourth white balance gain values.

Specifically, in order to avoid a large difference in white balance gain values in the face photographing mode and the non-face photographing mode, the third white balance gain value is selected from the plurality of fourth white balance gain values and the third white Balance the gain value of the target white balance gain value. Therefore, according to the target white balance gain value, the face color is considered on the one hand, and the rich color (grayscale world) of the natural world is considered on the other hand, which can not only improve the visual processing of the image processing. Effect, and based on the target Bai Ping The balance gain value is closer to the white balance gain value of the grayscale world, avoiding the problem that the white balance gain value is abrupt and causes the screen to flicker.

It should be noted that, according to different application scenarios, different implementation manners may be adopted, and according to the third white balance gain value, the target white balance close to the third white balance gain value is selected from the plurality of fourth white balance gain values. The gain value is illustrated as follows:

In a first example, a third gain value of each color component in the third white balance gain value is determined, and for each fourth white balance gain value, a fourth gain value of each color component is determined, for each fourth white balance The difference value between the gain value and the third white balance gain value is calculated, and the difference value is obtained by calculating the absolute difference between the third gain value and the fourth gain value in the same color component, and the absolute difference of each color component is obtained. And obtaining, from among the plurality of fourth white balance gain values, selecting a target white balance gain value that is the smallest difference value from the third gain value.

In a second example, a third vector is generated according to a third gain value of the third white balance gain value on each color component, and a fourth gain value is generated according to each fourth white balance gain value on each color component. a plurality of fourth vectors, calculating a vector distance between the third vector and each of the fourth vectors, the vector distance including a Euclidean distance, and selecting a vector from the plurality of fourth white balance gain values according to the vector distance The minimum target white balance gain value.

In step 304, a fifth gray balance gain value of the image is calculated using a simple gray world algorithm.

It should be understood that the gray world algorithm is based on the assumption that for an image with a large number of color variations, the average of the saturations of the three components R, G, and B tends to the same gray value G. In practical applications, there are usually two ways to determine the gray value G.

As a possible implementation, a fixed value can be taken. For example, one half of the brightest gray value may be taken, that is, when the brightest gray value is 255, the gray value G may be 128.

As another possible implementation manner, the average value of the three average values of R, G, and B in the image may be calculated as the gray value G. After determining the gradation value G, the fourth gain value of the image can be calculated by comparing the gradation value G with the average values of the three colors of R, G, and B, respectively.

Step 305: Perform white balance adjustment on the image according to the target white balance gain value and the fifth white balance gain value.

Specifically, since the target white balance value is a gain value that is slightly different from the third gain value, the processing effect on the face may be considered to a large extent, and when switching from including a face to a scene containing no face, It may be considered that a color change occurs due to a large change in the composition of the color. Therefore, in the embodiment of the present invention, further, the target white balance gain value and the fifth white balance gain value based on the gray world calculation are comprehensively considered. The image is white balance adjusted, so that when the face scene is changed to include no face scene, since the white balance gain value is closer to the white balance gain value of the gray world, the white balance gain value is effectively avoided. The problem of sudden color changes.

It should be noted that, according to different application scenarios, different implementation manners may be adopted, and white balance adjustment is performed on the image according to the target white balance gain value and the fifth white balance gain value, as illustrated below:

The first example:

A weighted average of the target white balance gain value and the fifth white balance gain value is calculated, and the image is white balance adjusted according to the weighted average value.

The second example:

Determining the weight of the target white balance gain value and the weight of the fifth white balance gain value according to the area ratio of the face area in the image, wherein the target white balance gain value and the gray world are determined due to the larger the face area ratio The difference of the fifth white balance gain value is larger. Therefore, in order to ensure that the primary color of the photographed face area matches the skin color, the weight between the target white balance gain value and the area ratio is positive, according to the target white balance gain value. The weights and the weights of the fifth white balance gain values are calculated to obtain a weighted average.

In summary, the white balance adjustment method of the embodiment of the present invention uses a face white balance algorithm to calculate a third white balance gain value of an image, and calculates an image corresponding to an image obtained by imaging under multiple light sources. a plurality of fourth white balance gain values, and selecting a target white balance gain value close to the third white balance gain value from the plurality of fourth white balance gain values according to the third white balance gain value, using a simple gray world algorithm The fifth white balance gain value of the image is calculated, and the image is white balance adjusted according to the target white balance gain value and the fifth white balance gain value. Thereby, the problem that the white balance gain value is abruptly caused by the sudden change of the white balance gain value when the face and the face are not in the same scene is suppressed, and the damage to the human eye is avoided.

Based on the above embodiment, for further detailed description, how to obtain a target white balance gain value close to the third white balance gain value from the plurality of fourth white balance gain values according to the third white balance gain value, which is combined with the above The second example shows a gain-based vector determining target white balance gain value as an example for explanation.

FIG. 8 is a flowchart of a white balance adjustment method according to another embodiment of the present invention. As shown in FIG. 8, the method includes:

In step 401, a third white balance gain value of the image is calculated by using a face white balance algorithm.

After getting the face area, the skin color of a group of people changes little. For example, according to statistics, after the skin color RGB color space is converted to the YCbCr space, the Cb and Cr ranges of the face are [133, 173], [77, 127], respectively. That is, as long as the human skin color range can be determined, the image can be corrected according to the skin color range. Therefore, the third gain value of the image can be calculated by comparing the color of the face region in the image with the preset skin color range.

Of course, the purpose of the foregoing embodiment for performing face recognition to determine the third gain value is to obtain a third gain value when the white balance processing is performed based on the skin color in the face photographing mode, in fact, in the front camera photographing mode or In the camera mode of the rear camera, the white balance processing is performed based on the white balance algorithm of the face. Therefore, it is also possible to determine the third white balance gain of the image by using the face white balance algorithm when the image is imaged by the front camera. Value, or, when determining that the image is imaged by the portrait mode of the rear camera, the face white balance algorithm is used to calculate the third white balance gain value of the image.

Step 402: Calculate a plurality of fourth white balance gain values corresponding to the image when the images are imaged under a plurality of light sources, respectively.

Step 403: Generate a third vector according to the third gain value of the third white balance gain value on each color component.

Step 404: Generate a corresponding plurality of fourth vectors according to the fourth gain value of each fourth white balance gain value on each color component.

In practical applications, the third gain value and the fourth gain value can be accurately characterized using vectors in the color space. The color space can be varied, for example: RGB (red, green, blue) color space, that is, based on the color space of the device's three primary colors. In addition, it may also be an HSI color space, which is derived from a human visual system, and describes colors by Hue, Saturation or Chroma, and Intensity or Brightness. The HSI color space can be described by a conical space model. Certainly, other color spaces may be used for description, which are not described in this embodiment. As a possible implementation, the third gain value and the fourth gain value may be characterized by an RGB model in the color space.

As a possible implementation, assuming that all colors are normalized, the cube is a unit cube, that is, all R, G, and B values are in the range of [0, 1]. Therefore, the values of the third gain value and the fourth gain value on each of the R, G, and B color components may also take values in the range of [0, 1]. By combining the values of the third gain value on each color component, a third vector can be generated, and the values of the fourth gain value on each color component are combined to generate a fourth vector. For example, if the third gain value has a value of 0.1 on the R component, a value of 0.2 on the G component, and a value of 0.3 on the B component, it can be on each color component according to the third gain value. The value of the third vector [0.1, 0.2, 0.3] is generated. If the fourth gain value has a value of 0.2 on the R component, a value of 0.2 on the G component, and a value of 0.2 on the B component, it can be taken on each color component according to the fourth gain value. The value produces a third vector [0.2, 0.2, 0.2].

Step 405, calculating a vector distance between the third vector and each of the fourth vectors; the vector distance includes Euclidean distance.

Specifically, after generating the third vector and the fourth vector, quantitative characterization of the third gain value and the fourth gain value is achieved. When calculating the vector distance between the third vector and the fourth vector, the Euclidean distance can be used to describe the vector distance between the two vectors, or the cosine distance and the Pearson correlation coefficient can be used to describe the two vectors. Vector distance between vectors. Taking the vector distance between the third vector and the fourth vector using the Euclidean distance as an example, the following Euclidean distance formula can be used:

A vector distance between the third vector and the fourth vector is calculated. Where d(x, y) is the vector distance between the third vector and the fourth vector, and x _R , x _G , and x _B are the values on each color component in the third vector, respectively, y _R , y _G And y _B are the values on each color component in the fourth vector, respectively.

Furthermore, after calculating the vector distance between the third vector and the fourth vector, the vector distance between the third vector and the fourth vector can be determined, and the vector distance between the third vector and the fourth vector is more Large, it can be determined that the third gain value and the fourth gain value are less similar. If the vector distance between the third vector and the fourth vector is closer, it may be determined that the third gain value and the fourth gain value are similar.

Step 406: Select, according to the vector distance, a target white balance gain value that is the smallest vector distance from the plurality of fourth white balance gain values.

Step 407, using a simple gray world algorithm, calculating a fifth white balance gain value of the image.

Step 408: Perform white balance adjustment on the image according to the target white balance gain value and the fifth white balance gain value.

Specifically, according to the vector distance, the target white balance gain value with the smallest vector distance is selected from the plurality of fourth white balance gain values, and the fifth white balance gain value and the target white balance gain value calculated by the gray world algorithm are combined. Accurate white balance processing on the image, on the one hand considers the face color, on the other hand considers the rich colors of nature, not only can improve the visual effect of image processing, avoiding the problem of white balance gain value abrupt and causing screen flicker.

In summary, the white balance adjustment method of the embodiment of the present invention calculates a third gain value for the image according to a face white balance algorithm for adjusting a face in an image to a skin color, and if the calculation is respectively When imaging the image under the light source, the image respectively corresponds to a plurality of fourth white balance gain values, and according to the third gain value of the third white balance gain value on each color component, a third vector is generated, according to each fourth a fourth gain value of the white balance gain value on each color component, generating a corresponding plurality of fourth vectors, and calculating a vector distance between the third vector and each of the fourth vectors; the vector distance includes a Euclidean distance, According to the vector distance, the target white balance gain value with the smallest vector distance is selected from the plurality of fourth white balance gain values, and then the fifth white balance gain value of the image is calculated by using the simple gray world algorithm, according to the target white The balance gain value and the fifth white balance gain value are used to adjust the white balance of the image. Therefore, with the slower white balance convergence speed, the problem of flickering when there is no face can be effectively improved when the white balance adjustment is performed by the face white balance algorithm.

In order to achieve the above embodiment, the present invention also provides a white balance adjusting device. FIG. 9 is a schematic structural view of a white balance adjusting device according to an embodiment of the present invention. As shown in FIG. 9, the white balance adjusting device includes a third calculating. The module 10, the fourth calculation module 20, the second selection module 30, the fifth calculation module 40, and the second adjustment module 50.

The third calculation module 10 is configured to calculate a third white balance gain value of the image by using a face white balance algorithm.

The fourth calculating module 20 is configured to calculate a plurality of fourth white balance gain values corresponding to the image when the images are respectively imaged under a plurality of light sources.

The second selecting module 30 is configured to select, according to the third white balance gain value, a target white balance gain value that is close to the third white balance gain value from the plurality of fourth white balance gain values;

The fifth calculating module 40 is configured to calculate a fifth white balance gain value of the image by using a simple gray world algorithm.

The second adjustment module 50 is configured to perform white balance adjustment on the image according to the target white balance gain value and the fifth white balance gain value.

In summary, the white balance adjusting device of the embodiment of the present invention uses a human face white balance algorithm to calculate a third white balance gain value of an image, and calculates an image corresponding to an image obtained by imaging under multiple light sources. a plurality of fourth white balance gain values, and selecting a target white balance gain value close to the third white balance gain value from the plurality of fourth white balance gain values according to the third white balance gain value, using a simple gray world algorithm The fifth white balance gain value of the image is calculated, and the image is white balance adjusted according to the target white balance gain value and the fifth white balance gain value. Thereby, the problem that the white balance gain value is abruptly caused by the sudden change of the white balance gain value when the face and the face are not in the same scene is suppressed, and the damage to the human eye is avoided.

To achieve the above objective, an embodiment of the present invention further provides a computer device. The above computer device includes an image processing circuit, and the image processing circuit may be implemented by hardware and/or software components, and may include various processing units defining an ISP (Image Signal Processing) pipeline. Continuing to refer to the schematic diagram of the image processing circuit of one embodiment shown in FIG. As shown in FIG. 6, for convenience of explanation, only various aspects of the image processing technique related to the embodiment of the present invention are shown.

As shown in FIG. 6, the image processing circuit includes an ISP processor 1040 and a control logic 1050. The image data captured by imaging device 1010 is first processed by ISP processor 1040, which analyzes the image data to capture Image statistics may be used to determine and/or image one or more control parameters of imaging device 1010. Imaging device 1010 can include a camera having one or more lenses 1012 and image sensors 1014. Image sensor 1014 may include a color filter array (such as a Bayer filter) that may acquire light intensity and wavelength information captured with each imaging pixel of image sensor 1014 and provide a set of primitives that may be processed by ISP processor 1040 Image data. Sensor 1020 can provide raw image data to ISP processor 1040 based on sensor 1020 interface type. The sensor 1020 interface may utilize a SMIA (Standard Mobile Imaging Architecture) interface, other serial or parallel camera interfaces, or a combination of the above.

ISP processor 1040 can also receive pixel data from image memory 1030. For example, raw pixel data is sent from the sensor 1020 interface to the image memory 1030, and the raw pixel data in the image memory 1030 is then provided to the ISP processor 1040 for processing. Image memory 1030 can be part of a memory device, a storage device, or a separate dedicated memory within an electronic device, and can include DMA (Direct Memory Access) features.

The statistics determined by the ISP processor 1040 can be sent to the control logic 1050 unit. For example, the statistical data may include image sensor 1014 statistical information such as auto exposure, auto white balance, auto focus, flicker detection, black level compensation, lens 1012 shading correction, and the like. Control logic 1050 can include a processor and/or a microcontroller that executes one or more routines, such as firmware, and one or more routines can determine control parameters and control of imaging device 1010 based on received statistical data. parameter. For example, the control parameters may include sensor 1020 control parameters (eg, gain, integration time for exposure control), camera flash control parameters, lens 1012 control parameters (eg, focus or zoom focal length), or groups of these parameters Hehe. The ISP control parameters may include gain levels and color correction matrices for automatic white balance and color adjustment (eg, during RGB processing), as well as lens 1012 shading correction parameters.

In step 201', a third white balance gain value of the image is calculated by using a face white balance algorithm.

Step 202' calculates a plurality of fourth white balance gain values corresponding to the image when the images are imaged under a plurality of light sources, respectively.

Step 203', selecting, according to the third white balance gain value, a target white balance gain value that is close to the third white balance gain value from the plurality of fourth white balance gain values.

Step 204', using a simple grayscale world algorithm, calculates a fifth white balance gain value for the image.

Step 205', performing white balance adjustment on the image according to the target white balance gain value and the fifth white balance gain value.

In summary, the terminal device in the embodiment of the present invention uses a face white balance algorithm to calculate a third white balance gain value of an image, and calculates multiple images corresponding to images obtained by imaging under multiple light sources. The fourth white balance gain value is selected from the plurality of fourth white balance gain values according to the third white balance gain value to obtain a target white balance gain value close to the third white balance gain value, and is calculated by using a simple gray world algorithm. The fifth white balance gain value of the image is obtained, and the image is white balance adjusted according to the target white balance gain value and the fifth white balance gain value. Thereby, the problem that the white balance gain value is abruptly caused by the sudden change of the white balance gain value when the face and the face are not in the same scene is suppressed, and the damage to the human eye is avoided.

In the description of the present specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" and the like means a specific feature described in connection with the embodiment or example. A structure, material or feature is included in at least one embodiment or example of the invention. In the present specification, the schematic representation of the above terms is not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. In addition, various embodiments or examples described in the specification and features of various embodiments or examples may be combined and combined without departing from the scope of the invention.

In addition, the terms "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or The implied indicates the number of technical features indicated. Thus, features defining "first" and "second" may include at least one of the features, either explicitly or implicitly. In the description of the present invention, the meaning of "a plurality" is at least two, such as two, three, etc., unless specifically defined otherwise.

Any process or method description in the flowcharts or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a custom logic function or process. And the scope of the preferred embodiments of the invention includes additional implementations, in which the functions may be performed in a substantially simultaneous manner or in an opposite order depending on the functions involved, in the order shown or discussed. It will be understood by those skilled in the art to which the embodiments of the present invention pertain.

The logic and/or steps represented in the flowchart or otherwise described herein, for example, may be considered as an ordered list of executable instructions for implementing logical functions, and may be embodied in any computer readable medium, Used in conjunction with, or in conjunction with, an instruction execution system, apparatus, or device (eg, a computer-based system, a system including a processor, or other system that can fetch instructions and execute instructions from an instruction execution system, apparatus, or device) Or use with equipment. For the purposes of this specification, a "computer-readable medium" can be any apparatus that can contain, store, communicate, propagate, or transport a program for use in an instruction execution system, apparatus, or device, or in conjunction with the instruction execution system, apparatus, or device. More specific examples (non-exhaustive list) of computer readable media include the following: electrical connections (electronic devices) having one or more wires, portable computer disk cartridges (magnetic devices), random access memory (RAM), Read only memory (ROM), erasable editable read only memory (EPROM or flash memory), fiber optic devices, and portable compact disk read only memory (CDROM). In addition, the computer readable medium may even be a paper or other suitable medium on which the program can be printed, as it may be optically scanned, for example by paper or other medium, followed by editing, interpretation or, if appropriate, other suitable The method is processed to obtain the program electronically and then stored in computer memory.

It should be understood that portions of the invention may be implemented in hardware, software, firmware or a combination thereof. In the above-described embodiments, multiple steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware and in another embodiment, it can be implemented by any one or combination of the following techniques well known in the art: discrete with logic gates for implementing logic functions on data signals Logic circuits, application specific integrated circuits with suitable combinational logic gates, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), and the like.

One of ordinary skill in the art can understand that all or part of the steps carried by the method of implementing the above embodiments can be completed by a program to instruct related hardware, and the program can be stored in a computer readable storage medium. When executed, one or a combination of the steps of the method embodiments is included.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above integrated module is It can be implemented in the form of hardware or in the form of a software function module. The integrated modules, if implemented in the form of software functional modules and sold or used as stand-alone products, may also be stored in a computer readable storage medium.

The above mentioned storage medium may be a read only memory, a magnetic disk or an optical disk or the like. Although the embodiments of the present invention have been shown and described, it is understood that the above-described embodiments are illustrative and are not to be construed as limiting the scope of the invention. The embodiments are subject to variations, modifications, substitutions and variations.

Claims

A white balance adjustment method, comprising the steps of:

Using the face white balance algorithm, the first white balance gain value of the image is calculated;

Calculating a plurality of second white balance gain values corresponding to the images respectively when the images are imaged under a plurality of light sources;

Determining, according to the first white balance gain value, a target white balance gain value that is close to the first white balance gain value from the plurality of second white balance gain values;

The image is subjected to white balance adjustment using the target white balance gain value.
The white balance adjustment method according to claim 1, wherein the first white balance gain is selected from the plurality of second white balance gain values according to the first white balance gain value The target white balance gain value close to the value, including:

Determining, in the first white balance gain value, a first gain value of each color component;

Determining, for each of the second white balance gain values, a second gain value of each color component;

Calculating a difference value between each of the second white balance gain values and the first white balance gain value, the difference value being the first gain value and the second gain value in the same color component After the absolute difference is calculated, the absolute difference of each color component is summed;

From among the plurality of the second white balance gain values, a target white balance gain value having a smallest difference value from the first gain value is selected.
The white balance adjustment method according to claim 1 or 2, wherein the first white balance gain value is selected from the plurality of second white balance gain values and the first white A target white balance gain value that approximates the gain value, including:

Generating a first vector according to the first gain value of the first white balance gain value on each color component;

Generating a corresponding plurality of second vectors according to a second gain value of each second white balance gain value on each color component;

Calculating a vector distance between the first vector and each of the second vectors; the vector distance includes a Euclidean distance;

And selecting, according to the vector distance, a target white balance gain value that is the smallest vector distance from the plurality of second white balance gain values.
The white balance adjustment method according to any one of claims 1 to 3, wherein before the calculating the first white balance gain value of the image by using the face white balance algorithm, the method further comprises:

Performing face recognition on the image to determine that the image includes a face region;

Or determining that the image is obtained by using a front camera;

Alternatively, it is determined that the image is imaged using a portrait mode of a rear camera.
The white balance adjustment method according to any one of claims 1 to 4, wherein the light source comprises one or more combinations of a daylight source, a fluorescent source, a tungsten filament source, and an F-A-H source.
A white balance adjusting device, comprising:

a first calculating module, configured to calculate a first white balance gain value of the image by using a human face white balance algorithm;

a second calculating module, configured to calculate a plurality of second white balance gain values respectively corresponding to the images when the images are obtained under a plurality of light sources respectively;

a first selecting module, configured to select, according to the first white balance gain value, a target white balance gain value that is close to the first white balance gain value from the plurality of second white balance gain values;

The first adjustment module is configured to perform white balance adjustment on the image by using the target white balance gain value.
The white balance adjustment device according to claim 6, wherein the first selection module comprises:

a first calculating unit, configured to determine a first gain value of each color component in the first white balance gain value; and determine, for each second white balance gain value, a second gain value of each color component; Calculating a difference value between the second white balance gain value and the first white balance gain value, the difference value being an absolute difference between the first gain value and the second gain value in the same color component After calculation, the sum of the absolute differences of the color components is obtained;

And a first selecting unit, configured to select, from among the plurality of the second white balance gain values, a target white balance gain value that is the smallest difference value from the first gain value.
The white balance adjustment device according to claim 6 or 7, wherein the first selection module comprises:

a second calculating unit, configured to generate a first vector according to the first gain value of the first white balance gain value on each color component; and a second gain on each color component according to each second white balance gain value a value, generating a corresponding plurality of second vectors; calculating a vector distance between the first vector and each of the second vectors;

a second selecting unit, configured to select, according to the vector distance, a target white balance gain value that is the smallest vector distance from the plurality of second white balance gain values.
A white balance adjustment method, comprising the steps of:

Using the face white balance algorithm, the third white balance gain value of the image is calculated;

Calculating a plurality of fourth white balance gain values corresponding to the image if the image is obtained by imaging under a plurality of light sources, respectively;

Determining, according to the third white balance gain value, a target white balance gain value that is close to the third white balance gain value from the plurality of fourth white balance gain values;

Calculating a fifth white balance gain value of the image using a simple gray world algorithm;

And white balance adjustment is performed on the image according to the target white balance gain value and the fifth white balance gain value.
The white balance adjustment method according to claim 9, wherein the third white balance gain is selected from the plurality of fourth white balance gain values according to the third white balance gain value. The target white balance gain value close to the value, including:

Determining, in the third white balance gain value, a third gain value of each color component;

Determining a fourth gain value of each color component for each fourth white balance gain value;

Calculating a difference value between each of the fourth white balance gain value and the third white balance gain value, wherein the difference value is for the third gain value and the fourth gain value of the same color component After the absolute difference is calculated, the absolute difference of each color component is summed;

From among the plurality of the fourth white balance gain values, a target white balance gain value having a smallest difference value from the third gain value is selected.
The white balance adjustment method according to claim 9 or 10, wherein the third white balance gain value is selected from the plurality of fourth white balance gain values and the third white A target white balance gain value that approximates the gain value, including:

Generating a third vector according to the third gain value of the third white balance gain value on each color component;

Generating a corresponding plurality of fourth vectors according to a fourth gain value of each fourth white balance gain value on each color component;

Calculating a vector distance between the third vector and each of the fourth vectors; the vector distance comprising a Euclidean distance;

And selecting, according to the vector distance, a target white balance gain value that is the smallest vector distance from the plurality of fourth white balance gain values.
The white balance adjustment method according to any one of claims 9 to 11, wherein before the third white balance gain value of the image is calculated by using the face white balance algorithm, the method further includes:

Performing face recognition on the image to determine that the image includes a face region;

Or determining that the image is obtained by using a front camera;

Alternatively, it is determined that the image is imaged using a portrait mode of a rear camera.
The white balance adjustment method according to any one of claims 9 to 12, wherein the light source comprises one or more combinations of a daylight source, a fluorescent source, a tungsten filament source, and an F-A-H source.
The white balance adjustment method according to any one of claims 9 to 13, wherein the image is white balance adjusted according to the target white balance gain value and the fifth white balance gain value, include:

Calculating a weighted average of the target white balance gain value and the fifth white balance gain value;

The image is subjected to white balance adjustment based on the weighted average.
The white balance adjustment method according to any one of claims 9-14, wherein the calculating a weighted average of the target white balance gain value and the fifth white balance gain value comprises:

Determining a weight of the target white balance gain value and a weight of the fifth white balance gain value according to an area ratio of a face area in the image; wherein a weight of the target white balance gain value and the area The ratio between the ratios is positive;

And calculating the weighted average according to the weight of the target white balance gain value and the weight of the fifth white balance gain value.
A white balance adjusting device, comprising:

a third calculating module, configured to calculate a third white balance gain value of the image by using a face white balance algorithm;

a fourth calculating module, configured to calculate a plurality of fourth white balance gain values corresponding to the image when the image is obtained by imaging under a plurality of light sources respectively;

a second selecting module, configured to select, according to the third white balance gain value, a target white balance gain value that is close to the third white balance gain value from the plurality of fourth white balance gain values;

a fifth calculating module, configured to calculate a fifth white balance gain value of the image by using a simple gray world algorithm;

And a second adjusting module, configured to perform white balance adjustment on the image according to the target white balance gain value and the fifth white balance gain value.
A computer device, comprising: a memory, a processor, and a computer program stored on the memory and operable on the processor, wherein when the processor executes the program, implementing any one of claims 1-5 The white balance adjustment method.
A computer readable storage medium having stored thereon a computer program, wherein the program is executed by a processor to implement the white balance adjustment method according to any one of claims 1-5.
A computer device, comprising: a memory, a processor, and a computer program stored on the memory and operable on the processor, the processor executing the program, implementing any one of claims 9-15 The white balance adjustment method.
A computer readable storage medium having stored thereon a computer program, characterized in that the program is executed by a processor to implement the white balance adjustment method according to any one of claims 9-15.