CN109040729A - Image white balance correcting, device, storage medium and terminal - Google Patents

Abstract

The embodiment of the present application discloses a kind of image white balance correcting, device, storage medium and terminal and obtains multiple target subgraphs this method comprises: being split first to target image；Secondly, obtaining the corresponding light source direction of each target sub-picture content；Again, light mixing scene whether there is according to the multiple target subgraph corresponding light source walking direction；Finally, carrying out white balance correction when there are light mixing scene according to the corresponding light source of the light mixing scene, can be improved image white balance correction effect under light mixing scene.

Description

Image white balance correction method, device, storage medium and terminal

technical field

The embodiments of the present application relate to the technical field of image processing, and in particular, to an image white balance correction method, device, storage medium, and terminal.

Background technique

With the continuous development of mobile terminals, almost every mobile terminal is equipped with a camera function, and can take pictures based on the camera function. The mobile terminal tends to automate the photo-taking process, and can automatically perform exposure and white balance correction according to the photo-taking environment.

However, in use, it is found that if there are multiple light sources from different directions in the shooting screen, that is, when there is a scene with mixed light, the white balance effect of the captured photo will be poor.

Contents of the invention

The purpose of the embodiments of the present application is to provide an image white balance correction method, device, storage medium, and terminal, which can achieve white balance correction effects in mixed light scenes.

In the first aspect, the embodiment of the present application provides an image white balance correction method, including:

Segmenting the target image to obtain multiple target sub-images;

Obtain the light source direction corresponding to the content of each target sub-image;

judging whether there is a mixed light scene according to the direction of the light source corresponding to the plurality of target sub-images;

When there is a mixed light scene, the white balance correction is performed according to the light source corresponding to the mixed light scene.

In the second aspect, the embodiment of the present application provides an image white balance correction device, including:

A segmentation module is used to segment the target image to obtain multiple target sub-images;

An acquisition module, configured to acquire the light source direction corresponding to the content of each target sub-image obtained by the segmentation module;

A judging module, configured to judge whether there is a mixed light scene according to the direction of the light source corresponding to the plurality of target sub-images acquired by the acquiring module;

The white balance module is configured to perform white balance correction according to the light source corresponding to the mixed light scene when the judging module determines that there is a mixed light scene.

In a third aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the image white balance correction method as shown in the first aspect is implemented.

In a fourth aspect, an embodiment of the present application provides a terminal, including a memory, a processor, and a computer program stored in the memory and operable on the processor. When the processor executes the computer program, the computer program described in the first aspect is implemented. The image white balance correction method shown.

In the image white balance correction scheme provided in the embodiment of the present application, firstly, the target image is segmented to obtain multiple target sub-images; secondly, the direction of the light source corresponding to the content of each target sub-image is obtained; The image corresponds to the direction of the light source to determine whether there is a mixed light scene; finally, when there is a mixed light scene, the white balance correction is performed according to the light source corresponding to the mixed light scene, which can improve the image white balance correction effect in the mixed light scene.

Description of drawings

FIG. 1 is a schematic flow chart of an image white balance correction method provided in an embodiment of the present application;

FIG. 2 is a schematic flowchart of another image white balance correction method provided by the embodiment of the present application;

FIG. 3 is a schematic flowchart of another image white balance correction method provided by the embodiment of the present application;

FIG. 4 is a schematic flowchart of another image white balance correction method provided by the embodiment of the present application;

FIG. 5 is a schematic flowchart of another image white balance correction method provided by the embodiment of the present application;

FIG. 6 is a schematic flowchart of another image white balance correction method provided by the embodiment of the present application;

FIG. 7 is a schematic structural diagram of an image white balance correction device provided in an embodiment of the present application;

FIG. 8 is a schematic structural diagram of a mobile terminal provided by an embodiment of the present application.

Detailed ways

The technical solution of the present application will be further described below in conjunction with the accompanying drawings and through specific implementation methods. It should be understood that the specific embodiments described here are only used to explain the present application, but not to limit the present application. In addition, it should be noted that, for the convenience of description, only some structures related to the present application are shown in the drawings but not all structures.

Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the steps as sequential processing, many of the steps may be performed in parallel, concurrently, or simultaneously. Additionally, the order of steps may be rearranged. The process may be terminated when its operations are complete, but may also have additional steps not included in the figure. The processing may correspond to a method, function, procedure, subroutine, subroutine, or the like.

With the continuous development of mobile terminals, almost every mobile terminal is equipped with a camera function, and can take pictures based on the camera function. The mobile terminal tends to automate the photo-taking process, and can automatically perform exposure and white balance correction according to the photo-taking environment. However, in use, it is found that if there are multiple light sources from different directions in the shooting screen, that is, when there is a scene with mixed light, the white balance effect of the captured photo will be poor, and it will not be able to accurately reflect the color temperature of the subject being photographed. The mixed light scene in the embodiment of the present application includes a shooting scene in which there are multiple light sources in different directions. When the light emitted by multiple light sources from different angles enters the module through the reflection of the object to be photographed, the color temperature of the collected light will be abnormal. At present, in the automatic white balance program of the mobile terminal, there is a lack of a mechanism for identifying the directions of different light sources, resulting in the inability to identify light sources in different directions, and thus unable to perform effective white balance processing on mixed light scenes.

The embodiment of the present application provides an image white balance correction method, which can identify whether the shooting scene is a mixed light scene. When the shooting scene is a mixed light scene, it can perform white balance processing based on light sources in multiple light source directions to improve the mixed light scene. Excellent white balance processing efficiency, so that the white balance processed image has a more accurate color temperature and improves resource utilization. The specific plan is as follows:

Figure 1 is a schematic flow chart of an image white balance correction method provided by an embodiment of the present application. This method is used in the case of using a mobile terminal to take pictures, especially in the case of taking pictures in mixed light scenes. The method can be executed by the mobile terminal. The mobile The terminal can be a smart phone, a tablet computer, a wearable device, a notebook computer, etc., and the method specifically includes the following steps:

Step 110: Segment the target image to obtain multiple target sub-images.

The target image may be a photo taken by the user, or may be a preview image in a preview stage. After the user takes a photo and obtains a photo, the mobile terminal automatically performs white balance processing on the photo, or the user performs post-production based on the photo obtained by taking the photo for modification. When the user starts the camera application, the camera application displays the preview frame image acquired by the camera in real time on the preview interface. When the preview frame image in the preview interface meets the user's needs, the user clicks the camera button to take a photo and obtain the photo. Before displaying the preview image, the mobile terminal performs white balance processing on the preview image, so that the user can obtain an image with accurate color temperature during the preview stage. Considering that white balance processing will generate calculation time, the frame rate of the preview image can be reduced accordingly.

The area can be divided according to a fixed preset, for example, the target image can be divided with a fixed size of 50 pixels*50 pixels. Further, if the fixed-size segmentation method cannot match the target image, and if there is a sub-image on the side of the target image that is less than the preset segmentation area, the sub-image can be ignored. Alternatively, the target image can be memorized divided according to a fixed number of divisions, for example, the target image can be divided into 10*10, 100 target sub-images in total.

Step 120, acquire the light source direction corresponding to the content of each target sub-image.

In an implementation manner, the illumination of the light source has directionality, and by detecting the light source and reflected light, the direction of the light source can be determined by detecting the light reflection in the target sub-image. Exemplarily, if the light source directions in at least two target sub-images satisfy a reflection law, the reflection information may be determined according to the target sub-images. The reflection information may include a corresponding relationship of at least two detection areas that have a reflection relationship.

For example, for flickering light sources such as incandescent lamps, objects such as tabletops will reflect light when the incandescent lamp is lit. When the incandescent lamp is extinguished, objects such as the table top do not emit light. Therefore, detection areas where pixel or brightness changes occur in pairs can be determined according to the pixel change values. The light source and its irradiation area can be determined based on the detection area of synchronously changing brightness.

In another implementation, the direction of the light source in any target sub-image can be identified by establishing a preset machine learning model. If there is a light source in the target sub-image, the output is a direction vector representing the direction of the light source. If there is no light source in the target sub-image, the output result can be 0 or a default item, etc.

Step 130 , judging whether there is a mixed light scene according to the direction of the light source corresponding to the plurality of target sub-images.

If there are multiple light source directions, it may be determined that there is a mixed light scene, and step 140 is performed. If there is only one light source direction, it is determined that there is no mixed light scene, and step 150 is performed.

Step 140, when there is a mixed light scene, perform white balance correction according to the light source corresponding to the mixed light scene.

The light sources corresponding to the mixed light scene may be all light sources corresponding to the target sub-image. After obtaining the multiple light sources of the mixed light scene, the reference color for white balance correction is determined according to the light intensities of the multiple light sources.

Step 150, when there is no mixed light scene, perform white balance processing according to the whiteboard.

The whiteboard can obtain the ambient color temperature, and perform white balance processing according to the ambient color temperature. Optionally, set the blue, green, and red photosensitive balance under the condition of 3200K color temperature. When the ambient color temperature is 3200K, the camera color temperature filter is placed at 3200K, and the scene can get correct color reproduction; when the ambient color temperature is 5600K, the camera color temperature filter is placed at 5600K, and the scene can get correct color reproduction. When the ambient color temperature is within the range of 1000K around 3200K and 1000K around 5600K, the color reproduction acceptable to human eyes can be obtained by using the white balance preset function.

In the image white balance correction method provided in the embodiment of the present application, first, the target image is segmented to obtain multiple target sub-images. Secondly, the direction of the light source corresponding to the content of each target sub-image is acquired. Again, it is judged whether there is a mixed light scene according to the direction of the light source corresponding to the plurality of target sub-images. Finally, when there is a mixed light scene, the white balance correction is performed according to the light source corresponding to the mixed light scene, which can improve the image white balance correction effect in the mixed light scene. At present, it is impossible to identify light sources from different light source directions in the target image, resulting in inaccurate judgment of the overall color temperature of the target image and poor white balance effect. The embodiments of the present application can determine light sources in different light source directions based on the segmented target sub-images, and then perform white balance correction according to the light sources in different directions, thereby restoring the color temperature of the scene to be photographed with higher quality.

Fig. 2 is a schematic flowchart of an image white balance correction method provided by the embodiment of the present application, as a further description of the above embodiment, including:

Step 210, segment the target image to obtain multiple target sub-images.

Step 220, acquiring the light source direction corresponding to the content of each target sub-image.

Step 230, judging whether the light source directions corresponding to the multiple adjacent target sub-images are the same.

An important factor for judging whether there is a mixed light scene is whether there are multiple light sources, and one form of multiple light sources is that multiple adjacent target sub-images correspond to different light source directions. The light source direction of each target sub-image can be obtained by preset algorithms such as machine learning models. Whether there are different light sources can be determined by comparing whether the light source directions between adjacent target sub-images are the same. Further, when the angular difference of the light source directions between adjacent target sub-images is smaller than a preset angle, it may be determined that the light source directions between adjacent target sub-images are the same. The preset angle is 5 degrees and 10 degrees.

If the light source directions corresponding to multiple adjacent target sub-images are the same, step 240 is performed. If the light source directions corresponding to multiple adjacent target sub-images are not the same, step 260 is performed.

Step 240, if the directions of the light sources corresponding to the multiple adjacent target sub-images are not the same, then there is a mixed light scene.

Step 250, when there is a mixed light scene, perform white balance correction according to the light source corresponding to the mixed light scene.

Step 260, if the direction of the light source corresponding to the multiple adjacent target sub-images is the same, then there is no mixed light scene, and white balance processing is performed according to the whiteboard.

The image white balance correction method provided by the embodiment of the present application can determine whether there are multiple light sources based on whether the light source directions of adjacent target sub-images are the same, in other words, whether the light source directions of adjacent target sub-images are significantly different, and thus more accurate It can accurately determine mixed light scenes and improve resource utilization.

Fig. 3 is a schematic flowchart of an image white balance correction method provided by the embodiment of the present application, as a further description of the above embodiment, including:

Step 310: Segment the target image to obtain multiple target sub-images.

Step 320, acquiring the light source direction corresponding to the content of each target sub-image.

Step 330 , judging whether there is a mixed light scene according to the direction of the light source corresponding to the plurality of target sub-images.

Step 340: When there is a mixed light scene, determine the main light source according to the multiple light sources corresponding to the multiple target sub-images.

Optionally, the light intensity of the light source may be determined according to the brightness of the target sub-image, and the light source with the highest light intensity is determined as the main light source. Perform white balance correction according to the color of the light source of the main light source.

Further, the weight of each light source may also be determined according to the light intensity of the light source. The higher the light source intensity is, the higher the weight is. Based on the weight of each light source and the color of each light source, the color of the main light source is obtained. Perform white balance correction according to the color of the light source of the main light source.

Step 350, performing white balance correction on the target image according to the main light source.

Step 360, when there is no mixed light scene, perform white balance processing according to the whitening board.

The image white balance correction method provided by the embodiment of the present application can determine the main light source from multiple light sources, and perform white balance processing according to the main light source, thereby more accurately determining the color of the main light source, and improving the white balance processing efficiency.

Fig. 4 is a schematic flowchart of an image white balance correction method provided by the embodiment of the present application, as a further description of the above embodiment, including:

Step 410: Segment the target image to obtain multiple target sub-images.

Step 420, acquiring the light source direction corresponding to the content of each target sub-image.

Step 430, judging whether there is a mixed light scene according to the direction of the light source corresponding to the plurality of target sub-images.

Step 440, when there is a mixed light scene, acquire multiple light source directions corresponding to multiple target sub-images.

After the mixed light scene is determined, multiple light source directions with strong light can be filtered out according to the light intensity. Optionally, the colors of each light source are obtained, and a light source with increased light intensity is selected from the same light source colors. Alternatively, all light source directions determined in step 420 can be read.

Step 450, count the number of target sub-images corresponding to each light source direction.

The number of target sub-images with the same light source direction is counted to obtain the number of target sub-images corresponding to each light source direction.

Step 460: Determine the main light source according to the counted number of target sub-images.

Determine the direction of the light source corresponding to the number of target sub-images with the highest value as the main light source.

Step 470, perform white balance correction on the target image according to the main light source.

Step 480, when there is no mixed light scene, perform white balance processing according to the whitening board.

The image white balance correction method provided in the embodiment of the present application can determine the light source with the highest light source ratio as the main light source from among multiple light sources, and then determine the color of the main light source more accurately, and improve the white balance processing efficiency.

Fig. 5 is a schematic flowchart of an image white balance correction method provided by the embodiment of the present application, as a further description of the above embodiment, including:

Step 510, acquiring multiple learning images with a single light source, and inputting the learning images into the convolutional neural network model to obtain a preset machine learning model.

When obtaining the light source direction of each target sub-image, it can be identified with the help of a preset machine learning model. The preset machine learning model may be a convolutional neural network model. Before the implementation of the embodiment of the present application, the convolutional neural network model is trained by means of machine learning, and the training sample may be a learning image with a single light source. The trained convolutional neural network model, that is, the preset machine learning model can identify the direction of the light source corresponding to any target sub-image.

Step 520: Segment the target image to obtain multiple target sub-images.

Step 530 : Preset the machine learning model with the input value of each target sub-image to obtain the light source direction corresponding to each target sub-image.

Step 540: Determine whether there is a mixed light scene according to the direction of the light source corresponding to the plurality of target sub-images.

Step 550, when there is a mixed light scene, perform white balance correction according to the light source corresponding to the mixed light scene.

Step 560, when there is no mixed light scene, perform white balance processing according to the whiteboard.

The image white balance correction method provided by the embodiment of the present application can obtain a preset machine learning model by inputting a learning image with a single light source for training, thereby avoiding the problem of inaccurate identification by a fixed algorithm and improving the usability of light source direction identification.

FIG. 6 is a schematic flowchart of an image white balance correction method provided by the embodiment of the present application, as a further description of the above embodiment, including:

Step 610, acquiring input specifications of a preset machine learning model.

If the input specification of the preset machine learning model is an image of a fixed size, before segmenting the target image, the input specification of the preset machine learning model is obtained, so as to segment the target image according to the input specification.

Step 620: Segment the target image according to the input specification to obtain multiple target sub-images.

Step 630: Preset the machine learning model with the input value of each target sub-image to obtain the light source direction corresponding to each target sub-image.

Step 640: Determine whether there is a mixed light scene according to the direction of the light source corresponding to the plurality of target sub-images.

Step 650: When there is a mixed light scene, perform white balance correction according to the light source corresponding to the mixed light scene.

Step 660, when there is no mixed light scene, perform white balance processing according to the whiteboard.

The image white balance correction method provided in the embodiment of the present application can divide the target image according to the input specifications of the preset machine learning model, so that the preset machine learning model can identify the light source direction of the target sub-image faster, and improve the processing efficiency. efficiency.

FIG. 7 is a schematic structural diagram of an image white balance correction device provided by an embodiment of the present application. As shown in FIG. 7 , the device includes: a segmentation module 710 , an acquisition module 720 , a judgment module 730 , a white balance module 740 and a learning module 750 .

A segmentation module 710, configured to segment the target image to obtain multiple target sub-images;

An acquisition module 720, configured to acquire the light source direction corresponding to each target sub-image content obtained by the segmentation module 710;

A judging module 730, configured to judge whether there is a mixed light scene according to the direction of the light source corresponding to the plurality of target sub-images acquired by the acquiring module 720;

The white balance module 740 is configured to perform white balance correction according to the light source corresponding to the mixed light scene when the judging module 730 determines that there is a mixed light scene.

Further, the judging module 730 is used for:

Judging whether the light source directions corresponding to multiple adjacent target sub-images are the same;

If the light source directions corresponding to multiple adjacent target sub-images are not the same, there is a mixed light scene.

Further, the white balance module 740 is used for:

determining a main light source according to multiple light sources corresponding to the multiple target sub-images;

Perform white balance correction on the target image according to the main light source.

Further, the white balance module 740 determines the main light source according to the multiple light sources corresponding to the multiple target sub-images, including:

Acquiring multiple light source directions corresponding to the multiple target sub-images;

Count the number of target sub-images corresponding to each light source direction;

The main light source is determined according to the counted number of target sub-images.

Further, the acquiring module 720 is used for:

Each target sub-image input value is preset to a machine learning model to obtain the light source direction corresponding to each target sub-image.

Further, a learning module 750 is also included, and the learning module 750 is used to: acquire multiple learning images with a single light source before each target sub-image input value is preset to the machine learning model;

Input the learning image into the convolutional neural network model to obtain a preset machine learning model, so that the judging module 730 judges the direction of the light source corresponding to the multiple target sub-images according to the preset machine learning model obtained by the learning module 750 and judges whether there is mixed light Scenes.

Further, the acquiring module 720 is used for:

Obtaining input specifications of the preset machine learning model;

Segmenting the target image according to the input specification to obtain multiple target sub-images.

In the image white balance correction device provided in the embodiment of the present application, the segmentation module 710 firstly segments the target image to obtain multiple target sub-images; secondly, the acquisition module 720 acquires the light source direction corresponding to the content of each target sub-image; thirdly, the judgment module 730 judges whether there is a mixed light scene according to the direction of the light source corresponding to the plurality of target sub-images; finally, when there is a mixed light scene, the white balance module 740 performs white balance correction according to the light source corresponding to the mixed light scene, which can improve the mixed light scene. Image white balance correction effect in the scene. At present, it is impossible to identify light sources from different light source directions in the target image, resulting in inaccurate judgment of the overall color temperature of the target image and poor white balance effect. The embodiments of the present application can determine light sources in different light source directions based on the segmented target sub-images, and then perform white balance correction according to the light sources in different directions, thereby restoring the color temperature of the scene to be photographed with higher quality.

The above-mentioned device can execute the methods provided by all the foregoing embodiments of the present application, and has corresponding functional modules and beneficial effects for executing the above-mentioned methods. For technical details not exhaustively described in this embodiment, reference may be made to the methods provided in all the foregoing embodiments of the present application.

FIG. 8 is a schematic structural diagram of a terminal device provided by an embodiment of the present application. As shown in FIG. 8 , the terminal may include: a housing (not shown in the figure), a memory 801, a central processing unit (Central Processing Unit, CPU) 802 (also known as a processor, hereinafter referred to as CPU), and stored in the memory 801. Computer programs running on the processor 802, a circuit board (not shown in the figure) and a power supply circuit (not shown in the figure). The circuit board is placed inside the space surrounded by the housing; the CPU 802 and the memory 801 are arranged on the circuit board; the power supply circuit is used to supply power to each circuit or device of the terminal; The memory 801 is used to store executable program codes; the CPU 802 runs programs corresponding to the executable program codes by reading the executable program codes stored in the memory 801 .

The terminal also includes: peripheral interface 803, RF (Radio Frequency, radio frequency) circuit 805, audio circuit 806, speaker 811, power management chip 808, input/output (I/O) subsystem 809, touch screen 812, other input /control device 810 and external port 804 , these components communicate via one or more communication buses or signal lines 807 .

It should be understood that the illustrated terminal device 800 is only an example of a terminal, and that the terminal device 800 may have more or fewer components than shown in the figure, two or more components may be combined, or Different component configurations are possible. The various components shown in the figures may be implemented in hardware, software, or a combination of hardware and software including one or more signal processing and/or application specific integrated circuits.

The following describes a terminal device provided in this embodiment in detail, and the terminal device takes a smart phone as an example.

Memory 801, the memory 801 can be accessed by the CPU 802, the peripheral interface 803, etc., the memory 801 can include a high-speed random access memory, and can also include a non-volatile memory, such as one or more disk storage devices, flash memory devices , or other volatile solid-state storage devices.

Peripheral interface 803 , which can connect the input and output peripherals of the device to CPU 802 and memory 801 .

The I/O subsystem 809 , the I/O subsystem 809 can connect input and output peripherals on the device, such as a touch screen 812 and other input/control devices 810 , to the peripheral interface 803 . I/O subsystem 809 may include a display controller 8091 and one or more input controllers 8092 for controlling other input/control devices 810 . Among them, one or more input controllers 8092 receive electrical signals from or send electrical signals to other input/control devices 810, which may include physical buttons (push buttons, rocker buttons, etc.) ), dials, slide switches, joysticks, click wheels. It is worth noting that the input controller 8092 can be connected to any of the following: a keyboard, an infrared port, a USB interface, and a pointing device such as a mouse.

Wherein, according to the working principle of the touch screen and the classification of the medium for transmitting information, the touch screen 812 can be a resistive type, a capacitive sensing type, an infrared type or a surface acoustic wave type. Classified according to the installation method, the touch screen 812 can be: plug-in type, built-in type or integral type. Classified according to technical principles, the touch screen 812 can be: a touch screen with vector pressure sensing technology, a touch screen with resistive technology, a touch screen with capacitive technology, a touch screen with infrared technology or a touch screen with surface acoustic wave technology.

A touch screen 812. The touch screen 812 is an input interface and an output interface between the user terminal and the user, and displays visual output to the user. The visual output may include graphics, text, icons, videos, and the like. Optionally, the touch screen 812 sends an electrical signal triggered by the user on the touch screen (such as an electrical signal on the contact surface) to the processor 802 .

The display controller 8091 in the I/O subsystem 809 receives electrical signals from the touch screen 812 or sends electrical signals to the touch screen 812 . The touch screen 812 detects the contact on the touch screen, and the display controller 8091 converts the detected contact into an interaction with the user interface object displayed on the touch screen 812, that is, realizes human-computer interaction, and the user interface object displayed on the touch screen 812 can be a running Icons for games, icons for networking to appropriate networks, etc. It is worth noting that the device may also include an optical mouse, which is a touch-sensitive surface that does not display visual output, or that is an extension of a touch-sensitive surface formed by a touch screen.

The RF circuit 805 is mainly used to establish communication between the smart speaker and the wireless network (that is, the network side), and realize data reception and transmission between the smart speaker and the wireless network. Such as sending and receiving short messages, e-mails, etc.

The audio circuit 806 is mainly used to receive audio data from the peripheral interface 803 , convert the audio data into electrical signals, and send the electrical signals to the speaker 811 .

The speaker 811 is used to restore the voice signal received by the smart speaker from the wireless network through the RF circuit 805 into sound and play the sound to the user.

The power management chip 808 is used for power supply and power management for the hardware connected to the CPU 802 , the I/O subsystem and the peripheral interface.

In this embodiment, the central processing unit 802 is used for:

Segmenting the target image to obtain multiple target sub-images;

Obtain the light source direction corresponding to the content of each target sub-image;

judging whether there is a mixed light scene according to the direction of the light source corresponding to the plurality of target sub-images;

When there is a mixed light scene, the white balance correction is performed according to the light source corresponding to the mixed light scene.

Further, the judging whether there is a mixed light scene according to the direction of the light source corresponding to the plurality of target sub-images includes:

Judging whether the light source directions corresponding to multiple adjacent target sub-images are the same;

If the light source directions corresponding to multiple adjacent target sub-images are not the same, there is a mixed light scene.

Further, the performing white balance correction according to the light source corresponding to the mixed light scene includes:

determining a main light source according to multiple light sources corresponding to the multiple target sub-images;

Perform white balance correction on the target image according to the main light source.

Further, the determining the main light source according to the multiple light sources corresponding to the multiple target sub-images includes:

Acquiring multiple light source directions corresponding to the multiple target sub-images;

Count the number of target sub-images corresponding to each light source direction;

The main light source is determined according to the counted number of target sub-images.

Further, the acquisition of the light source direction corresponding to the content of each target sub-image includes:

Each target sub-image input value is preset to a machine learning model to obtain the light source direction corresponding to each target sub-image.

Further, before each target sub-image is input into a value preset machine learning model, it also includes:

Acquire multiple learning images with a single light source;

The learning image is input into the convolutional neural network model to obtain a preset machine learning model.

Further, the target image is segmented to obtain multiple target sub-images, including:

Obtaining input specifications of the preset machine learning model;

Segmenting the target image according to the input specification to obtain multiple target sub-images.

The embodiment of the present application also provides a storage medium containing executable instructions of a terminal device. The executable instructions of the terminal device are used to execute an image white balance correction method when executed by a processor of the terminal device. The method includes:

Segmenting the target image to obtain multiple target sub-images;

Obtain the light source direction corresponding to the content of each target sub-image;

judging whether there is a mixed light scene according to the direction of the light source corresponding to the plurality of target sub-images;

When there is a mixed light scene, the white balance correction is performed according to the light source corresponding to the mixed light scene.

Further, the judging whether there is a mixed light scene according to the direction of the light source corresponding to the plurality of target sub-images includes:

Judging whether the light source directions corresponding to multiple adjacent target sub-images are the same;

If the light source directions corresponding to multiple adjacent target sub-images are not the same, there is a mixed light scene.

Further, the performing white balance correction according to the light source corresponding to the mixed light scene includes:

determining a main light source according to multiple light sources corresponding to the multiple target sub-images;

Perform white balance correction on the target image according to the main light source.

Further, the determining the main light source according to the multiple light sources corresponding to the multiple target sub-images includes:

Acquiring multiple light source directions corresponding to the multiple target sub-images;

Count the number of target sub-images corresponding to each light source direction;

The main light source is determined according to the counted number of target sub-images.

Further, the acquisition of the light source direction corresponding to the content of each target sub-image includes:

Each target sub-image input value is preset to a machine learning model to obtain the light source direction corresponding to each target sub-image.

Further, before each target sub-image is input into a value preset machine learning model, it also includes:

Acquire multiple learning images with a single light source;

The learning image is input into the convolutional neural network model to obtain a preset machine learning model.

Further, the target image is segmented to obtain multiple target sub-images, including:

Obtaining input specifications of the preset machine learning model;

Segmenting the target image according to the input specification to obtain multiple target sub-images.

The computer storage medium in the embodiments of the present application may use any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (non-exhaustive list) of computer readable storage media include: electrical connections with one or more leads, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), Erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

A computer readable signal medium may include a data signal carrying computer readable program code in baseband or as part of a carrier wave. Such propagated data signals may take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. .

Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including - but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for performing the operations of the present application may be written in one or more programming languages or combinations thereof, including object-oriented programming languages—such as Java, Smalltalk, C++, and conventional procedural Programming language - such as "C" or a similar programming language. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In cases involving a remote computer, the remote computer can be connected to the user computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (such as through an Internet service provider). Internet connection).

Certainly, a storage medium containing computer-executable instructions provided in the embodiments of the present application, the computer-executable instructions are not limited to the above-mentioned image white balance correction operation, and may also execute the image white balance correction operation provided in any embodiment of the present application. Related operations in the balance correction method.

Note that the above are only preferred embodiments and technical principles used in this application. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein, and various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the protection scope of the present application. Therefore, although the present application has been described in detail through the above embodiments, the present application is not limited to the above embodiments, and can also include more other equivalent embodiments without departing from the concept of the present application, and the present application The scope is determined by the scope of the appended claims.

Claims (10)

1. A method for image white balance correction, characterized in that, comprising: Segmenting the target image to obtain multiple target sub-images; Obtain the light source direction corresponding to the content of each target sub-image; judging whether there is a mixed light scene according to the direction of the light source corresponding to the plurality of target sub-images; When there is a mixed light scene, the white balance correction is performed according to the light source corresponding to the mixed light scene. 2. The image white balance correction method according to claim 1, wherein said judging whether there is a mixed light scene according to the direction of the light source corresponding to the plurality of target sub-images comprises: Judging whether the light source directions corresponding to multiple adjacent target sub-images are the same; If the light source directions corresponding to multiple adjacent target sub-images are not the same, there is a mixed light scene. 3. The image white balance correction method according to claim 1, wherein the white balance correction is performed according to the light source corresponding to the mixed light scene, comprising: determining a main light source according to multiple light sources corresponding to the multiple target sub-images; Perform white balance correction on the target image according to the main light source. 4. The image white balance correction method according to claim 3, wherein the determining the main light source according to the plurality of light sources corresponding to the plurality of target sub-images comprises: Acquiring multiple light source directions corresponding to the multiple target sub-images; Count the number of target sub-images corresponding to each light source direction; The main light source is determined according to the counted number of target sub-images. 5. The image white balance correction method according to claim 1, wherein said obtaining the light source direction corresponding to each target sub-image content comprises: Each target sub-image input value is preset to a machine learning model to obtain the light source direction corresponding to each target sub-image. 6. the image white balance correction method according to claim 5, is characterized in that, before each target sub-image input value presets machine learning model respectively, also comprises: Acquire multiple learning images with a single light source; The learning image is input into the convolutional neural network model to obtain a preset machine learning model. 7. The image white balance correction method according to claim 5, wherein said segmenting the target image to obtain a plurality of target sub-images comprises: Obtaining input specifications of the preset machine learning model; Segmenting the target image according to the input specification to obtain multiple target sub-images. 8. An image white balance correction device, characterized in that, comprising: A segmentation module is used to segment the target image to obtain multiple target sub-images; An acquisition module, configured to acquire the light source direction corresponding to the content of each target sub-image obtained by the segmentation module; A judging module, configured to judge whether there is a mixed light scene according to the direction of the light source corresponding to the plurality of target sub-images acquired by the acquiring module; The white balance module is configured to perform white balance correction according to the light source corresponding to the mixed light scene when the judging module determines that there is a mixed light scene. 9. A computer-readable storage medium, on which a computer program is stored, wherein when the program is executed by a processor, the image white balance correction method according to any one of claims 1-7 is realized. 10. A terminal, comprising a memory, a processor, and a computer program stored on the memory and operable by the processor, characterized in that, when the processor executes the computer program, it implements any of claims 1-7. The image white balance correction method described above.