WO2021104352A1 - White balance correction method and apparatus, storage medium, and terminal device - Google Patents

Abstract

Disclosed are a white balance correction method and apparatus, a storage medium, and a terminal device. The method comprises: on the basis of image data acquired by a camera, determining an initial classification result and an initial confidence level of a light source; on the basis of data acquired by a color temperature sensor, determining an infrared band ratio and color temperature information of the light source; performing calculation according to the infrared band ratio, the color temperature information and the initial confidence level to obtain a target confidence level; when the target confidence level is less than a preset threshold value, obtaining a target classification result by adjusting the initial classification result; and determining a target white balance parameter according to the target classification result, and performing white balance correction on the image data according to the determined target white balance parameter.

Description

White balance correction method, device, storage medium and terminal equipment

Cross-references to related applications

This application is filed based on a Chinese patent application with application number 201911174624.4 and an application date of November 26, 2019, and claims the priority of the Chinese patent application. The entire content of the Chinese patent application is hereby incorporated into this application by way of introduction.

Technical field

This application relates to the field of image processing technology, and in particular to a white balance correction method, device, storage medium, and terminal equipment.

Background technique

White balance (White Balance) is an important concept in the field of photography. White balance can solve a series of problems such as color restoration and tone processing. For different light sources, the color of the light emitted by it is different, which causes the color of the light reflected by the object to be different. Then when the object is photographed through the terminal device, due to the influence of the light source, it is likely to cause the color cast of the photographed object. At this time, it is necessary to distinguish the indoor light source from the outdoor light source, and select the corresponding parameters according to the indoor/outdoor light source To adjust the white balance to improve color reproduction.

For indoor/outdoor judgments, there are already some solutions. However, these solutions are not considered comprehensively. For example, the similarity between tungsten filament lamps and daylight sources is not considered, and the judgment is too simple, which is prone to misjudgment, which makes the accuracy of indoor/outdoor judgments low.

Summary of the invention

The embodiments of the present application propose a white balance correction method, device, storage medium, and terminal equipment, which can improve the accuracy of indoor/outdoor judgment, and can better distinguish indoor/outdoor scenes, thereby achieving better and more stable White balance effect.

The technical solutions of the embodiments of the present application are implemented as follows:

In the first aspect, an embodiment of the present application provides a white balance correction method, which includes:

Based on the image data obtained by the camera, determine the initial classification result and initial confidence of the light source;

Based on the data obtained by the color temperature sensor, determine the infrared band ratio and color temperature information of the light source;

Calculate according to infrared band ratio, color temperature information and initial confidence to obtain target confidence;

When the target confidence is less than the preset threshold, the target classification result is obtained by adjusting the initial classification result;

Determine the target white balance parameter according to the target classification result, and perform white balance correction on the image data according to the determined target white balance parameter.

In a second aspect, an embodiment of the present application provides a white balance correction device, which includes a determination unit, a calculation unit, an adjustment unit, and a correction unit, wherein:

The determining unit is configured to determine the initial classification result and initial confidence of the light source based on the image data obtained by the camera; and is further configured to determine the infrared band ratio and color temperature information of the light source based on the data obtained by the color temperature sensor;

The calculation unit is configured to perform calculations based on the infrared band ratio, color temperature information and initial confidence to obtain the target confidence;

The adjustment unit is configured to obtain the target classification result by adjusting the initial classification result when the target confidence is less than the preset threshold;

The correction unit is configured to determine the target white balance parameter according to the target classification result, and perform white balance correction on the image data according to the determined target white balance parameter.

In a third aspect, an embodiment of the present application provides a white balance correction device, which includes a memory and a processor; wherein,

The memory is used to store computer programs that can run on the processor;

The processor is configured to execute the method described in the first aspect when the computer program is running.

In a fourth aspect, an embodiment of the present application provides a computer storage medium that stores a white balance correction program, which implements the method described in the first aspect when the white balance correction program is executed by at least one processor.

In a fifth aspect, an embodiment of the present application provides a terminal device, which at least includes the white balance correction device as described in the second aspect or the third aspect.

The white balance correction method, device, storage medium and terminal equipment provided by the embodiments of the present application determine the initial classification result and initial confidence of the light source based on the image data acquired by the camera; determine the light source's initial classification result and initial confidence based on the data acquired by the color temperature sensor Infrared band ratio and color temperature information; calculate according to the infrared band ratio, color temperature information and initial confidence to obtain the target confidence; when the target confidence is less than the preset threshold, the target classification result is obtained by adjusting the initial classification result; according to the target The classification result determines the target white balance parameters, and performs white balance correction on the image data according to the determined target white balance parameters; in this way, after the light source is judged once to obtain the initial classification result and initial confidence, the ratio of infrared bands and the The color temperature information performs a second judgment on the light source, which can effectively correct the misjudgment scenes that exist after the first judgment. It can not only better distinguish the indoor/outdoor scenes, but also improve the accuracy of indoor/outdoor judgments, so as to achieve better , A more stable white balance effect, so that the terminal equipment can better color reproduction.

Description of the drawings

FIG. 1 is a schematic diagram of a white balance correction process provided by related technical solutions;

2 is a schematic flowchart of a white balance correction method provided by an embodiment of the application;

3 is a schematic diagram 1 of the placement position of a color temperature sensor according to an embodiment of the application;

4 is a second schematic diagram of the placement position of a color temperature sensor according to an embodiment of the application;

FIG. 5 is a third schematic diagram of the placement position of a color temperature sensor according to an embodiment of the application;

FIG. 6 is a fourth schematic diagram of the placement position of a color temperature sensor according to an embodiment of the application;

FIG. 7 is a fifth schematic diagram of the placement position of a color temperature sensor according to an embodiment of the application; FIG.

FIG. 8 is a schematic flowchart of another white balance correction method provided by an embodiment of the application;

FIG. 9 is a schematic flowchart of another white balance correction method provided by an embodiment of the application;

FIG. 10 is a schematic diagram of a detailed process of obtaining a target confidence level according to an embodiment of this application;

FIG. 11 is a detailed flowchart of a white balance correction method provided by an embodiment of the application;

FIG. 12 is a schematic diagram of the composition structure of a white balance correction device provided by an embodiment of the application;

FIG. 13 is a schematic diagram of a specific hardware structure of a white balance correction device provided by an embodiment of the application;

FIG. 14 is a schematic diagram of the composition structure of a terminal device provided by an embodiment of the application.

Detailed ways

The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It is understandable that the specific embodiments described here are only used to explain the related application, but not to limit the application. In addition, it should be noted that, for ease of description, only the parts related to the relevant application are shown in the drawings.

White balance is an important concept in the field of photography. White balance can solve a series of problems such as color reproduction and tone processing. Among them, a color will appear different colors under the illumination of different light sources. For example, gray will appear yellow under the illumination of incandescent lamp, and will appear cyan under the illumination of cold light, which is the commonly referred to as color cast phenomenon. At this time, the color of the input original image needs to be processed to make it appear normal gray under the illumination of different light sources. At present, the common correction method is white balance correction, but the existing white balance is difficult to accurately distinguish the light source information (indoor/outdoor), resulting in the result of correction in some scenes, there is still color cast or white balance jump phenomenon .

There are already some solutions for indoor/outdoor judgments. However, most of the current solutions only make indoor/outdoor judgments based on infrared band information. The specific judgment process is that if the infrared band ratio (that is, the ratio of the infrared band to the full spectrum band) is less than 20%, then the current environment can be determined to be indoor; if the infrared band ratio is higher than 35%, or if the current environment If it is sunny and the infrared band ratio is higher than 45%, it can be determined that the current environment is outdoor.

However, in the current solution, the indoor/outdoor judgment method is too simple, and there are theoretically wrong logic. For example, when the infrared band ratio is lower than 20%, according to the current solution, it can be determined that the current environment is indoor ; But the light source spectrum of the indoor tungsten lamp is gradually rising, and its infrared band ratio is also very high. At this time, when the corresponding infrared band ratio of the tungsten lamp light source is higher than 45%, the current solution can be It is determined that the current environment is outdoor, which will cause misjudgment, that is, misjudge the tungsten light source as outdoor. Similarly, for indoor mixed light source scenes, such as a mixed light source of tungsten filament lamps and fluorescent lamps or a mixed light source of tungsten filament lamps and Light Emitting Diodes (LED), the current solutions may also cause misjudgments. It was misjudged as an outdoor situation. In addition, the current solution still has a defect that it does not consider the smoothing mechanism. For example, the infrared band ratio in the actual scene may fluctuate between 30% and 40%, while the current solution will have the infrared band ratio less than 35%. If the current environment is judged to be indoor, if the infrared band ratio is greater than 35%, judging that the current environment is outdoor, which may cause the indoor/outdoor judgment result to switch back and forth, resulting in a jump in the effect of the white balance algorithm, which is not stable enough. In other words, the current solution does not consider the similarity between tungsten filament lamps and daylight sources, nor does it consider adding other dimensions to further distinguish indoor/outdoor, resulting in low accuracy of indoor/outdoor judgment, and does not consider the time domain. Due to the stability of the indoor/outdoor judgment result, there may be an abnormal phenomenon of frequent switching, and the stability is poor.

Refer to FIG. 1, which shows a schematic flowchart of a white balance correction provided by related technical solutions. As shown in Figure 1, the process is mainly to use a two-classifier to make indoor/outdoor judgments, and perform white balance corrections based on the judgment results. Specifically, step S101 is first performed to input raw image data to be processed; after the raw image data is input to the second classifier, then step S102 is performed to judge by the second classifier; according to the judgment result of the second classifier , Execute step S103 to determine whether the judgment result is indoor; if it is determined that the judgment result is indoor, execute step S104 to call indoor parameters, and then execute step S105 to obtain the white balance correction result according to the indoor parameters; if it is determined that the judgment result is outdoor , Step S106 is executed to call the outdoor parameters, and then step S107 is executed to obtain the white balance correction result according to the outdoor parameters.

It should be noted that, based on the process shown in Figure 1, since indoors are basically artificial light sources, such as tungsten lamps (Tungsten Lamp), fluorescent lamps (Fluorescent Lamp), LEDs and other light sources; while outdoor light sources are basically daylight sources ( Day Light or Natural Light) mainly. After determining whether it is indoor or outdoor, the type of light source can basically be determined, so that some interference areas can be eliminated in the calculation of the white balance algorithm, so as to improve the effect of the white balance algorithm. For example, when it is judged that the current environment is outdoor, the light source landing area for fluorescent lamps and LEDs can be determined as the interference area, and at this time, the color of the light source is excluded from the statistics. However, the accuracy of the current two-classifiers cannot reach 100%, and there will be 5% to 10% probability of judgment errors during use, resulting in the accuracy of indoor/outdoor judgments being 90% to 95%. It is impossible to accurately determine the interference area, which makes the white balance algorithm less effective in misjudgment.

The embodiment of the application provides a white balance correction method, which determines the initial classification result and initial confidence of the light source based on the image data acquired by the camera; determines the infrared band ratio and color temperature information of the light source based on the data acquired by the color temperature sensor; The ratio, color temperature information and initial confidence are calculated to obtain the target confidence; when the target confidence is less than the preset threshold, the target classification result is obtained by adjusting the initial classification result; the target white balance parameter is determined according to the target classification result, and the target white balance parameter is determined according to the target classification result. The determined target white balance parameter performs white balance correction on the image data; in this way, after the current light source is judged once to obtain the initial classification result and initial confidence, the light source is performed according to the infrared band ratio and color temperature information determined by the color temperature sensor. The second judgment can effectively correct the misjudgment scenes that exist after the first judgment. It can not only better distinguish indoor/outdoor scenes, but also improve the accuracy of indoor/outdoor judgments, so as to achieve better and more stable whiteness. The balance effect enables better color reproduction of the terminal equipment.

Hereinafter, each embodiment of the present application will be described in detail with reference to the accompanying drawings.

In an embodiment of the present application, refer to FIG. 2, which shows a schematic flowchart of a white balance correction method provided by an embodiment of the present application. As shown in Figure 2, the method may include:

S201: Determine the initial classification result and initial confidence of the light source based on the image data acquired by the camera;

It should be noted that before performing the white balance correction method of the embodiment of the present application, the initial classification result and initial confidence of a light source can be obtained first. Here, it can be based on a preset classifier (such as a two-classifier) to obtain the initial classification result and initial confidence of the light source, or it can be based on other classification methods (such as only based on infrared band information) to obtain the initial classification of the light source The results and initial confidence are not specifically limited in the embodiments of the present application.

In addition, the initial classification results of the light source may generally include indoor and outdoor two types, and for each of the initial classification results, the initial confidence can be set to 50%, but the embodiment of the present application does not specifically limit it.

In some embodiments, when the initial classification result and initial confidence of the light source are obtained according to the two classifiers, for S201, the initial classification result and initial confidence of the light source are determined based on the image data obtained by the camera. include:

Input the image data obtained by the camera into the preset classifier;

Through the preset classifier, the initial classification result and initial confidence of the light source are output.

It should be noted that the preset classifier may be a two-classifier, that is, the current environment is classified as indoor or outdoor. Specifically, the acquired image data is input into a preset classifier, through which the initial classification result and initial confidence of the light source can be obtained. For example, when the initial classification result is indoor, the initial confidence can be set to 50%; when the initial classification result is outdoor, the initial confidence can also be set to 50%.

S202: Determine the infrared band ratio and color temperature information of the light source based on the data obtained by the color temperature sensor;

It should be noted that the infrared band ratio usually refers to the ratio of the infrared band to the full-spectrum band; the color temperature information usually refers to the relative color temperature (CCT), which specifically indicates that the color of the light emitted by the light source and the black body are in a certain range. When the radiant light color is the same at temperature, the temperature of the black body is the color temperature of the light source; that is, the correlated color temperature refers to the relative value closest to the radiant temperature of the black body.

Here, for S201 and S202, there is no priority between them. The embodiment of the present application can execute S201 and S202 in parallel to obtain the initial classification result, initial confidence, infrared band ratio and color temperature information of the light source.

In some embodiments, the color temperature sensor can be used to obtain the spectral channel values corresponding to each of the multiple spectral bands in the light source, so that the infrared band ratio and color temperature information of the light source can be determined. Therefore, for S202, the determination of the infrared band ratio and color temperature information of the light source based on the data obtained by the color temperature sensor may include:

Obtain the corresponding spectral channel values of multiple spectral bands in the light source through the color temperature sensor;

Based on the obtained spectrum channel value, the infrared band ratio and color temperature information of the light source are calculated.

It should be noted that the color temperature is a characteristic that visible light has important applications in photography, video and other fields. Theoretically, color temperature refers to the color that an absolute black body will appear after being heated from absolute zero (-273°C). After the black body is heated, it gradually turns from black to red, turns yellow, turns white, and finally emits blue light. When heated to a certain temperature, the spectral component of the light emitted by the black body is called the color temperature at this temperature, and the unit of measurement is Kelvin (indicated by K). Here, CCT refers to the relative value closest to the black body radiation temperature.

It should also be noted that for the color temperature sensor set on the terminal device, specifically, the color temperature sensor can be set on the side of the front camera of the terminal device. As shown in Figure 3, it is configured on the left side of the front camera of the terminal device. There is a color temperature sensor; it can also be set on the side of the rear camera of the terminal device. As shown in Figure 4, a color temperature sensor is configured on the lower side of the rear camera of the terminal device.

The color temperature sensor can also be set in the fringe area of the full screen. Specifically, Fig. 5 is a schematic diagram of setting the color temperature sensor. As shown in Fig. 5, the terminal device places the color temperature sensor under the ink in the fringe area.

The terminal device can also set the color temperature sensor in the slit on the top. Specifically, FIG. 6 is a schematic diagram of the placement of the color temperature sensor on the side of the display screen of the terminal device; FIG. 7 is a schematic diagram of the placement of the color temperature sensor on the side of the rear camera of the terminal device.

In the embodiment of the present application, the color temperature sensor can determine the infrared band ratio and color temperature information of the current light source. Specifically, the color temperature sensor can usually output five channel values. The five channels are red light band (represented by R), green light band (represented by G), blue light band (represented by B), and infrared band (represented by IR). Shows), visible light and infrared full-range bands (indicated by C) and the corresponding channels of the five spectrum bands. In other words, through the color temperature sensor, the corresponding spectral channel values of the five spectral bands in the light source can be obtained; then based on the five spectral channel values, the infrared band ratio (that is, the infrared band relative to the full-spectrum band) can be calculated Ratio, expressed in IR%) and the color temperature information of the light source (expressed in CCT). Among artificial light sources and daylight sources, tungsten filament lamps and low color temperature daylight sources (around 3000K) have the highest IR%, while high color temperature daylight sources (between 4000K and 7500K) have higher IR%. Fluorescent lamps and LEDs The light source has almost no IR% value. For details, see Table 1, which shows an example of the IR% value range under different light sources (the specific value varies with the response curves of the five channels of the color temperature sensor).

Table 1

In Table 1, the value range of IR% corresponding to daylight light source is 49%～70%, the value range of IR% corresponding to fluorescent light source is 8%～21%, and the value range of IR% corresponding to LED light source is 12%-18%; it can be seen that the IR% corresponding to fluorescent lamps and LED light sources is very low, and there is almost no infrared band. Therefore, light sources can be distinguished according to different IR% values; and for mixed light sources, the infrared band ratio and color temperature information output by the color temperature sensor can also be combined for auxiliary judgment, thereby improving the accuracy of indoor/outdoor judgment.

S203: Calculate according to the infrared band ratio, the color temperature information, and the initial confidence to obtain the target confidence;

It should be noted that for the initial classification result, the obtained infrared band ratio can be compared with a preset infrared band threshold, and then the initial confidence level can be correlated according to the infrared band ratio and color temperature information to obtain the target confidence level.

Specifically, the preset infrared band threshold may include a first threshold and a second threshold, where the first threshold represents a preset infrared band threshold when the initial classification result is indoor, and the second threshold represents when the initial classification result is outdoor. The preset infrared band threshold; here, the first threshold and the second threshold are both adjustable parameters, and the specific values are set according to actual application scenarios, which are not specifically limited in the embodiment of the present application.

Optionally, in some embodiments, when the initial classification result is indoors, as shown in FIG. 8, for S203, according to the infrared band ratio, the color temperature information, and the initial confidence level Perform calculations to obtain target confidence, which can include:

S301: Compare the infrared band ratio with a first threshold;

S302: When the infrared waveband ratio is greater than or equal to the first threshold, judge the light source according to the infrared waveband ratio and the color temperature information;

S303: If it is determined that the light source is a daylight light source, perform a subtraction operation on the initial confidence level and a first preset value to obtain the target confidence level;

S304: If it is determined that the light source is a non-daylight light source, perform an addition operation on the initial confidence level and a second preset value to obtain the target confidence level.

Further, after S301, the method may further include:

S305: When the infrared band ratio is less than the first threshold, add the initial confidence level and a third preset value to obtain the target confidence level.

It should be noted that the first preset value, the second preset value, and the third preset value are all adjustable parameters, and the specific values are set according to actual application scenarios, which are not specifically limited in the embodiment of the present application.

It should also be noted that after comparing the infrared band ratio with the first threshold, if the infrared band ratio is greater than or equal to the first threshold, it indicates that the current light source has a certain IR component, that is, the light source has three possibilities , It may be a daylight light source, it may be a tungsten light source, or a mixed light source (such as a mixed light source of a tungsten lamp and a fluorescent lamp or a mixed light source of a tungsten lamp and LED); among them, the daylight source belongs to the outdoor, and the tungsten lamp And the mixed light source belongs to indoors. At this time, step S302 needs to be performed to judge the light source according to the infrared band ratio and the color temperature information to determine whether the light source is a daylight light source; if the light source is a daylight light source, the current environment may be outdoor. At this time, step S303 needs to be performed, that is, the initial confidence level is subtracted from the first preset value to obtain the target confidence level; if the light source is a non-daylight light source, the current environment may be indoors. At this time, step S304 needs to be performed, that is, the initial confidence level is added. Set the second preset value to obtain the target confidence.

In addition, if the infrared band ratio is less than the first threshold, it means that there are few IR components in the current light source, and it can be determined with a high probability that the judgment result is indoor. At this time, step S305 needs to be performed, that is, the initial confidence level is added to the third preset value. To get the target confidence.

Optionally, in some embodiments, when the initial classification result is outdoor, as shown in FIG. 9, for S203, according to the infrared band ratio, the color temperature information, and the initial confidence level Perform calculations to obtain target confidence, which can include:

S401: Compare the infrared band ratio with a second threshold;

S402: When the infrared waveband ratio is greater than or equal to the second threshold, judge the light source according to the infrared waveband ratio and the color temperature information;

S403: If it is determined that the light source is a daylight source, perform an addition operation on the initial confidence level and a fourth preset value to obtain the target confidence level;

S404: If it is determined that the light source is a non-daylight light source, perform a subtraction operation on the initial confidence level and a fifth preset value to obtain the target confidence level.

Further, after S401, the method may further include:

S405: When the infrared band ratio is less than the second threshold, perform a subtraction calculation on the initial confidence level and a sixth preset value to obtain the target confidence level.

It should be noted that the fourth preset value, the fifth preset value, and the sixth preset value are all adjustable parameters, and the specific values are set according to actual application scenarios, which are not specifically limited in the embodiment of the present application.

It should also be noted that after comparing the infrared band ratio with the second threshold, if the infrared band ratio is greater than or equal to the second threshold, it indicates that the current light source has a certain IR component, that is, the light source has three possibilities , It may be a daylight light source, it may be a tungsten light source, or a mixed light source (such as a mixed light source of a tungsten lamp and a fluorescent lamp or a mixed light source of a tungsten lamp and LED); among them, the daylight source belongs to the outdoor, and the tungsten lamp And the mixed light source belongs to indoors. At this time, step S402 needs to be executed to judge the light source according to the infrared band ratio and the color temperature information to determine whether the light source is a daylight light source; if the light source is a daylight light source, the current environment may be outdoor. At this time, step S403 needs to be performed, that is, the initial confidence level is added to the fourth preset value to obtain the target confidence level; if the light source is a non-daylight light source, the current environment may be indoors. At this time, step S404 needs to be performed, that is, the initial confidence level is reduced. Go to the fifth preset value to get the target confidence.

In addition, if the infrared band ratio is less than the second threshold, it means that there are few IR components in the current light source. It can be determined with a high probability that the judgment result is indoor, and the probability that the judgment result is outdoor is very low. In this case, step S405 is required, namely The sixth preset value is subtracted from the initial confidence level to obtain the target confidence level.

In this way, after the target confidence level is obtained, the target confidence level can be compared with a preset threshold, and based on the result of the comparison, it is determined whether the initial classification result needs to be changed, that is, whether the initial classification result needs to be adjusted.

S204: When the target confidence is less than a preset threshold, obtain a target classification result by adjusting the initial classification result;

It should be noted that the preset threshold is a preset judgment value used to judge whether the initial classification result is reliable. Here, the preset threshold is an adjustable parameter, and the specific value is set according to actual application scenarios, which is not specifically limited in the embodiment of the present application. Generally speaking, the preset threshold value needs to be set relatively small, for example, the preset threshold value is set to 30%; in this way, when the target confidence is less than 30%, it can be judged that the initial classification result is very unreliable, and there is a misjudgment at this time In the situation, the classification result needs to be changed, that is, the initial classification result is adjusted to obtain the target classification result.

In some embodiments, for S204, when the target confidence is less than a preset threshold, obtaining a target classification result by adjusting the initial classification result may include:

When the target confidence is less than the preset threshold, the initial classification result is adjusted to determine the target classification result; wherein,

If the initial classification result is indoor, it is determined that the target classification result is outdoor;

If the initial classification result is outdoor, it is determined that the target classification result is indoor.

Further, in some embodiments, after S203, the method may further include:

When the target confidence is greater than or equal to the preset threshold, it is determined that the target classification result is to maintain the initial classification result; wherein,

If the initial classification result is indoor, it is determined that the target classification result is indoor;

If the initial classification result is outdoor, it is determined that the target classification result is outdoor.

It should be noted that the target confidence is compared with the preset threshold. If the target confidence is less than the preset threshold, it indicates that the current initial classification result is unreliable. At this time, the initial classification result needs to be adjusted, for example, if the initial classification If the result is indoor, the target classification result obtained after adjustment is outdoor; if the initial classification result is outdoor, the target classification result obtained after adjustment is indoor. In addition, if the target confidence is greater than or equal to the preset threshold, it indicates that the current initial classification result is reliable. At this time, the initial classification result needs to be maintained, that is, if the initial classification result is indoor, the target classification result is also indoor; If the classification result is outdoor, the target classification result is also outdoor; in this way, the target classification result obtained is reliable, which improves the accuracy of indoor/outdoor judgment; and the smaller preset threshold also ensures that the probability of misjudgment is relatively high. Adjusting the classification result only when it is large can also be regarded as considering the smoothing mechanism in the time domain, which helps maintain the stability of the white balance algorithm, thereby avoiding the abnormal phenomenon of frequent switching of the white balance algorithm.

S205: Determine the target white balance parameter according to the target classification result, and perform white balance correction on the image data according to the determined target white balance parameter.

It should be noted that since the obtained target classification result takes into account the infrared band ratio and color temperature information at the same time, the target classification result is reliable. At this time, the white balance correction of the image data to be processed can achieve a better white balance effect. .

Further, in some embodiments, determining the target white balance parameter according to the target classification result, and performing white balance correction on the image data according to the determined target white balance parameter may include:

If the target classification result is indoor, call indoor parameters, obtain target white balance parameters from the indoor parameters, and perform white balance correction on the image data according to the target white balance parameters;

If the target classification result is outdoor, call outdoor parameters, obtain target white balance parameters from the outdoor parameters, and perform white balance correction on the image data according to the target white balance parameters.

It should be noted that if the target classification result is indoor, you can call the indoor parameters at this time, and then obtain the target white balance parameters from the indoor parameters, and perform white balance correction on the image data according to the target white balance parameters; if the target classification result is outdoor, At this time, you can call the outdoor parameters, and then obtain the target white balance parameters from the outdoor parameters, and perform white balance correction on the image data according to the target white balance parameters. This can improve the effect of the white balance algorithm and eliminate the color cast in the original image. Can better achieve color reproduction.

In addition, when the obtained infrared band ratio is greater than or equal to the preset infrared band threshold (such as the first threshold or the second threshold), there are three possibilities for the current light source, which may be daylight or tungsten. The filament light source may also be a mixed light source (such as a mixed light source of a tungsten lamp and a fluorescent lamp or a mixed light source of a tungsten lamp and LED); among them, the daylight light source belongs to the outdoor, and the tungsten lamp and the mixed light source belong to the indoor. Further distinguish the light source. Therefore, in some embodiments, the judging the light source based on the infrared band ratio and the color temperature information may include:

According to the infrared band ratio, query the first color temperature information corresponding to the infrared band ratio from a pre-established comparison table;

The light source is judged according to the color temperature information and the first color temperature information obtained by the query.

Further, the judging the light source according to the color temperature information and the first color temperature information obtained by the query may include:

Calculating the difference between the color temperature information and the first color temperature information;

When the difference is less than a preset difference threshold, selecting the smallest color temperature information from the color temperature information and the first color temperature information;

Compare the minimum color temperature information with the preset color temperature threshold;

If the minimum color temperature information is greater than the preset color temperature threshold, it is determined that the light source is a daylight light source;

If the minimum color temperature information is less than or equal to the preset color temperature threshold, it is determined that the light source is a non-daylight light source.

It should be noted that under the premise that the infrared band ratio is greater than or equal to the preset infrared band threshold (such as the first threshold or the second threshold), there are three possibilities for the light source. At this time, the range of the infrared band ratio (IR%) They are relatively similar and cannot be distinguished only by IR%. At this time, the color temperature information output by the color temperature sensor needs to be considered, and the first color temperature information corresponding to the IR% can be queried from the pre-established comparison table. Multiple dimensions to distinguish the light source, which can improve the accuracy.

Specifically, after obtaining the color temperature information and the first color temperature information, calculate the difference between the color temperature information and the first color temperature information, and then compare the difference with a preset difference threshold. If the difference is greater than or equal to the preset The difference threshold indicates that the light source is a mixed light source, such as a mixed light source of a tungsten lamp and a fluorescent lamp or a mixed light source of a tungsten lamp and LED; if the difference is less than the preset difference threshold, it indicates that the light source is a pure color light source, such as Daylight light source or tungsten light source; At this time, it is also necessary to select the smallest color temperature information from the color temperature information and the first color temperature information, and then compare the smallest color temperature information with the preset color temperature threshold; if the smallest color temperature information is greater than the preset color temperature Threshold (for example, 3000K), the light source is determined to be a daylight source; if the minimum color temperature information is less than or equal to the preset color temperature threshold, the light source is determined to be a tungsten light source, that is, the light source is a non-daylight light source.

Exemplarily, there are currently two typical scenes that are easy to be misjudged: The first typical scene is an indoor scene with high brightness and high CCT. Because the color feature is close to outdoor, it is easy to be wrongly judged as outdoor; The light source is a fluorescent lamp or an LED, and its IR component is very small. The method of the embodiment of the present application can greatly correct the misjudgment, and ensure that the accuracy of the indoor judgment can reach about 99%. The second typical scene is an outdoor scene with low brightness, such as the outdoor near the evening, it is easy to be mistakenly judged as indoor at this time; but the light source of this scene is sunlight, and its IR component is very high. It can better correct misjudgments, and ensure that the correct rate of judgement outdoor can reach about 97%.

That is to say, the method of the embodiment of the present application can effectively correct the scene of indoor/outdoor misjudgment in the preset classifier (such as the two classifier) in the original white balance algorithm, and improve the accuracy of indoor/outdoor classification; especially When the indoor light source is a fluorescent lamp or LED, or there is only an outdoor daylight light source, the accuracy of the judgment can reach about 99%. In this way, not only the correct and appropriate white balance parameters are called (assuming that the white balance parameters include two sets of parameters such as indoor parameters and outdoor parameters), but also the scene can be better distinguished in advance, so that the debugger can eliminate the problem without adding additional parameters. For some misjudged scenes, there is no need to add additional parameters to distinguish between confusing scenes or confusing colors, which can achieve the effect of a better, more stable, and more convenient debugging white balance algorithm.

This embodiment provides a white balance correction method, which determines the initial classification result and initial confidence of the light source based on the image data acquired by the camera; determines the infrared band ratio and color temperature information of the light source based on the data acquired by the color temperature sensor; The ratio, color temperature information and initial confidence are calculated to obtain the target confidence; when the target confidence is less than the preset threshold, the target classification result is obtained by adjusting the initial classification result; the target white balance parameter is determined according to the target classification result, and the target white balance parameter is determined according to the target classification result. The determined target white balance parameter performs white balance correction on the image data; in this way, after the light source is judged once to obtain the initial classification result and initial confidence, the light source can also be judged twice based on the infrared band ratio and color temperature information. Effectively correct the misjudgment scene that exists after a judgment, not only can better distinguish indoor/outdoor scenes, but also improve the accuracy of indoor/outdoor judgments, so as to achieve a better and more stable white balance effect, and make the terminal The equipment is capable of better color reproduction.

In another embodiment of the present application, refer to FIG. 10, which shows a detailed flowchart of a target confidence acquisition provided by an embodiment of the present application. As shown in Figure 10, the method may include:

S501: Obtain the initial classification result and initial confidence of the preset classifier;

It should be noted that the preset classifier may be a two-class classifier or a Support Vector Machine (SVM) classifier; according to the preset classifier, the initial classification result and initial confidence of the light source can be obtained. Generally speaking, the initial confidence value is 50%, but the embodiment of the present application does not specifically limit it.

S502: If the initial classification result is indoor, perform step S503;

S503: Compare the acquired infrared band ratio with the first threshold;

S504: If the infrared band ratio is not less than the first threshold, judge the light source according to the infrared band ratio and the color temperature information;

S505: If it is determined that the light source is a daylight source, the target confidence is equal to the initial confidence minus the first preset value;

S506: If it is determined that the light source is a non-daylight source, the target confidence is equal to the initial confidence plus the second preset value;

S507: If the infrared band ratio is less than the first threshold, the target confidence is equal to the initial confidence plus the third preset value;

S508: Obtain the target confidence when the initial classification result is indoor;

It should be noted that if the initial classification result is Indoor, it is necessary to judge whether the infrared band ratio (expressed in IR%) is less than the first threshold (expressed in THR1); if the IR% is not less than THR1, that is, the IR% is greater than Or equal to THR1, indicating that the current light source has a certain IR component. At this time, there are three possibilities for the light source. It may be a daylight source, it may be a tungsten light source, or a mixed light source (such as a mixed light source of a tungsten lamp and a fluorescent lamp) Or a mixed light source of a tungsten lamp and an LED lamp); among them, the daylight light source belongs to the outdoor, and the tungsten lamp and the mixed light source belong to the indoor. At this time, it is necessary to distinguish the three possibilities. Because the IR% of the tungsten light source is the highest, and its corresponding CCT is about 2850K, it is easy to distinguish the tungsten light source; however, the IR% value range of the daylight source and the mixed light source are similar, and it is impossible to pass only IR. % To distinguish, the CCT value needs to be added at this time; that is to say, the light source is judged according to the infrared band ratio and the color temperature information. If it is judged that the light source is a daylight source, the current environment may be outdoor. At this time, you need to perform steps S505: The initial confidence level such as the target confidence level is subtracted from the first preset value (indicated by X1%), that is, the target confidence level can be (50-X1)%; if it is determined that the light source is a non-sunlight light source, the current environment may be Indoors, step S506 needs to be performed at this time, the initial confidence level such as the target confidence level plus the second preset value (indicated by X2%), that is, the target confidence level can be (50+X2)%; in addition, if the IR% is less than THR1 , Which shows that the IR component of the current light source is very small, so it can be judged that the light source is most likely to be fluorescent or LED, that is, the most likely to be an indoor scene. At this time, execute S507, and the target confidence is equal to the initial confidence plus the third preset value (Expressed by X3%), that is, the target confidence can be (50+X3)%; and the closer the target confidence is to 100%, the more it can be determined as an indoor scene. In this way, after these steps, the target confidence level (CL) when the initial classification result is indoor can be calculated.

S509: If the initial classification result is outdoor, perform step S510;

S510: Compare the acquired infrared band ratio with a second threshold;

S511: If the infrared band ratio is not less than the second threshold, judge the light source according to the infrared band ratio and the color temperature information;

S512: If it is determined that the light source is a daylight source, the target confidence is equal to the initial confidence plus the fourth preset value;

S513: If it is determined that the light source is a non-daylight source, the target confidence is equal to the initial confidence minus the fifth preset value;

S514: If the infrared band ratio is less than the second threshold, the target confidence is equal to the initial confidence minus the sixth preset value;

S515: Obtain the target confidence when the initial classification result is outdoor.

It should be noted that if the initial classification result is Outdoor, it is necessary to judge whether the infrared band ratio (represented by IR%) is less than the second threshold (represented by THR2); if the IR% is not less than THR2, that is, the IR% is greater than Or equal to THR2, indicating that the current light source has a certain IR component. At this time, the light source has three possibilities. It may be a daylight light source, a tungsten light source, or a mixed light source (such as a mixed light source of a tungsten lamp and a fluorescent lamp). Or a mixed light source of a tungsten lamp and an LED lamp); among them, the daylight source belongs to the outdoor, and the tungsten lamp and the mixed light source belong to the indoor. At this time, it is necessary to distinguish the three possibilities. The light source is judged according to the infrared band ratio and the color temperature information. If it is judged that the light source is a daylight light source, the current environment may be outdoor. At this time, step S512 needs to be performed, and the initial confidence level such as the target confidence level plus the fourth preset value (Expressed by X4%), that is, the target confidence level can be (50+X4)%; if it is determined that the light source is a non-daylight light source, the current environment may be indoors. At this time, step S513 needs to be performed, the target confidence level and other initial confidence levels Subtract the fifth preset value (indicated by X5%), that is, the target confidence can be (50-X5)%; in addition, if the IR% is less than THR2, it means that the current light source has very little IR component, then the light source can be judged The high probability is a fluorescent lamp or LED light, that is, the high probability is an indoor scene, and the probability that the judgment result is outdoor is very low. At this time, execute S514, and the target confidence is equal to the initial confidence minus the sixth preset value (indicated by X6%) , That is, the target confidence level can be (50-X6)%; the closer the target confidence level obtained at this time is to 0%, the more it can be determined as an indoor scene. In this way, after these steps, the target confidence level (CL) when the initial classification result is outdoor can be calculated.

After the target confidence level (represented by CL) is determined, the target confidence level can be used to determine whether the current judgment result needs to be changed. Specifically, the target confidence level can be compared with a preset threshold. The preset threshold is a predetermined judgment value used to determine whether the initial classification result is reliable. The preset threshold is set to be relatively small, which may be the lowest confidence level. Threshold (represented by Y%); in the embodiment of the present application, the preset threshold may be set to 30%, but the embodiment of the present application does not specifically limit it.

Among them, if the CL is less than 30%, then the initial classification result is not reliable. At this time, the initial classification result can be adjusted, that is, if the initial classification result is indoor, the adjusted target classification result is outdoor; if the initial classification result is outdoor , The adjusted target classification result is indoor. If CL is greater than or equal to 30%, then the initial classification result is reliable, and the target classification result at this time can maintain the initial classification result, that is, if the initial classification result is indoor, the target classification result is also indoor; if the initial classification result is outdoor , The target classification result is also outdoor.

It is worth noting that the preset threshold usually needs to be set relatively small, so that when the target confidence is lower than the preset threshold, it can be determined that the current initial classification result is very unreliable, that is, the initial classification result is determined to be incorrect. In the case of judgment, the initial classification result needs to be changed at this time. That is to say, because changing the initial classification result is a risky operation, it is necessary to change the initial classification result only when the misjudgment is determined with a high probability. In the medium confidence interval, if it is not certain whether it is a misjudgment, you can choose to maintain the original initial classification result at this time. This process can be regarded as a smoothing mechanism in the time domain, which helps maintain the stability of the white balance algorithm, avoids the abnormal phenomenon of frequent switching of the white balance algorithm, and also avoids the instability of the white balance effect in the same scene. .

Refer to FIG. 11, which shows a detailed flowchart of a white balance correction method provided by an embodiment of the present application. On the basis of Figure 1, the detailed process adds four steps:

S601: Calculate the initial confidence level according to the infrared band ratio and color temperature information to obtain the target confidence level;

S602: Determine whether the target confidence level is less than a preset threshold;

S603: If the target confidence is greater than or equal to the preset threshold, the target classification result is to maintain the initial classification result;

S604: If the target confidence is less than the preset threshold, the target classification result is the adjusted initial classification result.

In other words, after adding the color temperature sensor, the IR% and CCT can be obtained through the color temperature sensor; the target confidence of the current indoor/outdoor initial classification results can be calculated according to the IR% and CCT; the target confidence is low (ie, the target confidence When the degree is less than the preset threshold), change the initial classification result, that is, correct the misjudgment result under lower confidence; in this way, by combining the color temperature sensor for auxiliary judgment, not only can the accuracy of indoor/outdoor judgment be improved, but also Achieve a better white balance effect.

Through the foregoing embodiments, the specific implementation of the foregoing embodiments is described in detail. It can be seen from the technical solutions of the foregoing embodiments that after the light source is judged once to obtain the initial classification result and the initial confidence, it can also be determined according to the Infrared band ratio and color temperature information for the second judgment of the light source, which can effectively correct the misjudgment scene after the first judgment, not only can better distinguish indoor/outdoor scenes, but also improve the accuracy of indoor/outdoor judgments. So as to achieve a better and more stable white balance effect, so that the terminal device can better color reproduction.

Based on the same inventive concept as the foregoing embodiment, refer to FIG. 12, which shows a schematic diagram of the composition structure of a white balance correction device 70 provided by an embodiment of the present application. As shown in FIG. 12, the white balance correction device 70 may include: a determination unit 701, a calculation unit 702, an adjustment unit 703, and a correction unit 704, where

The determining unit 701 is configured to determine the initial classification result and initial confidence of the light source based on the image data obtained by the camera; and is also configured to determine the infrared band ratio and color temperature information of the light source based on the data obtained by the color temperature sensor;

The calculation unit 702 is configured to perform calculation according to the infrared band ratio, the color temperature information, and the initial confidence to obtain the target confidence;

The adjustment unit 703 is configured to obtain a target classification result by adjusting the initial classification result when the target confidence is less than a preset threshold;

The correction unit 704 is configured to determine a target white balance parameter according to the target classification result, and perform white balance correction on the image data according to the determined target white balance parameter.

In the above solution, the determining unit 701 is specifically configured to input the image data obtained by the camera into a preset classifier; and output the initial classification result and the initial confidence of the light source through the preset classifier.

In the above solution, the determining unit 701 is specifically configured to obtain the spectral channel value corresponding to each of the multiple spectral bands in the light source through the color temperature sensor; and calculate the infrared band ratio and color temperature information of the light source based on the obtained spectral channel value .

In the above solution, referring to FIG. 12, the white balance correction device 70 may further include a comparison unit 705 and a judgment unit 706, wherein,

The comparing unit 705 is configured to compare the infrared band ratio with a first threshold when the initial classification result is indoor;

The judging unit 706 is configured to judge the light source according to the infrared waveband ratio and the color temperature information when the infrared waveband ratio is greater than or equal to the first threshold;

The calculation unit 702 is specifically configured to, if it is determined that the light source is a daylight light source, perform a subtraction operation between the initial confidence level and a first preset value to obtain the target confidence level; and if it is determined that the light source is a non-daylight light source, Then, the initial confidence level and the second preset value are added together to obtain the target confidence level.

In the above solution, the calculation unit 702 is further configured to add the initial confidence level and a third preset value to obtain the target confidence when the infrared band ratio is less than the first threshold. degree.

In the above solution, the comparing unit 705 is further configured to compare the infrared band ratio with a second threshold when the initial classification result is outdoor;

The judging unit 706 is further configured to judge the light source according to the infrared waveband ratio and the color temperature information when the infrared waveband ratio is greater than or equal to the second threshold;

The calculation unit 702 is specifically configured to, if it is determined that the light source is a daylight light source, add the initial confidence level and a fourth preset value to obtain the target confidence level; and if it is determined that the light source is a non-daylight light source, Then, the initial confidence level is subtracted from the fifth preset value to obtain the target confidence level.

In the above solution, the calculation unit 702 is further configured to perform a subtraction calculation between the initial confidence level and a sixth preset value when the infrared band ratio is less than the second threshold value to obtain the target confidence degree.

In the above solution, the adjustment unit 703 is specifically configured to adjust the initial classification result to determine the target classification result when the target confidence is less than the preset threshold; If the initial classification result is indoor, it is determined that the target classification result is outdoor; if the initial classification result is outdoor, it is determined that the target classification result is indoor.

In the above solution, referring to FIG. 12, the white balance correction device 70 may further include a maintaining unit 707 configured to determine that the target classification result is to maintain the initial value when the target confidence is greater than or equal to the preset threshold. Classification result; wherein, if the initial classification result is indoor, it is determined that the target classification result is indoor; if the initial classification result is outdoor, it is determined that the target classification result is outdoor.

In the above solution, the correction unit 704 is specifically configured to call indoor parameters if the target classification result is indoor, and obtain the target white balance parameters from the indoor parameters, and compare the image according to the target white balance parameters. White balance correction is performed on the data; and if the target classification result is outdoor, outdoor parameters are called, the target white balance parameters are obtained from the outdoor parameters, and the image data is white balance corrected according to the target white balance parameters.

In the above solution, referring to FIG. 12, the white balance correction device 70 may further include a query unit 708, configured to query the first color temperature corresponding to the infrared band ratio from a pre-established comparison table according to the infrared band ratio. information;

The judging unit 706 is specifically configured to judge the light source according to the color temperature information and the first color temperature information obtained by the query.

In the above solution, the determining unit 706 is specifically configured to calculate the difference between the color temperature information and the first color temperature information; and when the difference is less than a preset difference threshold, from the color temperature Selecting the minimum color temperature information from the information and the first color temperature information; and comparing the minimum color temperature information with a preset color temperature threshold; and if the minimum color temperature information is greater than the preset color temperature threshold, determining that the light source is sunlight Light source; if the minimum color temperature information is less than or equal to the preset color temperature threshold, it is determined that the light source is a non-daylight light source.

It can be understood that, in this embodiment, a "unit" may be a part of a circuit, a part of a processor, a part of a program or software, etc., of course, it may also be a module, or it may also be non-modular. Moreover, the various components in this embodiment may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit. The above-mentioned integrated unit can be realized in the form of hardware or software function module.

If the integrated unit is implemented in the form of a software function module and is not sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solution of this embodiment is essentially or It is said that the part that contributes to the existing technology or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can It is a personal computer, a server, or a network device, etc.) or a processor (processor) that executes all or part of the steps of the method described in this embodiment. The aforementioned storage media include: U disk, mobile hard disk, read only memory (Read Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program codes.

Therefore, this embodiment provides a computer storage medium that stores a white balance correction program that implements the method described in any one of the foregoing embodiments when the white balance correction program is executed by at least one processor.

Based on the above composition of the white balance correction device 70 and the computer storage medium, refer to FIG. 13, which shows the specific hardware structure of the white balance correction device 70 provided by the embodiment of the present application, which may include: a communication interface 801, a memory 802, and a processor 803: The components are coupled together through the bus system 804. It can be understood that the bus system 804 is used to implement connection and communication between these components. In addition to the data bus, the bus system 804 also includes a power bus, a control bus, and a status signal bus. However, for clear description, various buses are marked as the bus system 804 in FIG. 13. Among them, the communication interface 801 is used for receiving and sending signals in the process of sending and receiving information with other external network elements;

The memory 802 is configured to store a computer program that can run on the processor 803;

The processor 803 is configured to execute: when the computer program is running:

Based on the image data obtained by the camera, determine the initial classification result and initial confidence of the light source;

Based on the data obtained by the color temperature sensor, determine the infrared band ratio and color temperature information of the light source;

Calculate according to the infrared band ratio, the color temperature information and the initial confidence to obtain the target confidence;

When the target confidence is less than a preset threshold, the target classification result is obtained by adjusting the initial classification result;

Determine the target white balance parameter according to the target classification result, and perform white balance correction on the image data according to the determined target white balance parameter.

It can be understood that the memory 802 in the embodiment of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memory. Among them, the non-volatile memory can be read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable ROM, PROM), erasable programmable read-only memory (Erasable PROM, EPROM), and electrically available Erase programmable read-only memory (Electrically EPROM, EEPROM) or flash memory. The volatile memory may be a random access memory (Random Access Memory, RAM), which is used as an external cache. By way of exemplary but not restrictive description, many forms of RAM are available, such as static random access memory (Static RAM, SRAM), dynamic random access memory (Dynamic RAM, DRAM), synchronous dynamic random access memory (Synchronous DRAM, SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (Double Data Rate SDRAM, DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (Enhanced SDRAM, ESDRAM), Synchronous Link Dynamic Random Access Memory (Synchlink DRAM, SLDRAM) And Direct Rambus RAM (DRRAM). The memory 802 of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

The processor 803 may be an integrated circuit chip with signal processing capability. In the implementation process, the steps of the foregoing method can be completed by an integrated logic circuit of hardware in the processor 803 or instructions in the form of software. The aforementioned processor 803 may be a general-purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (Field Programmable Gate Array, FPGA) or other Programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps, and logical block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor or the like. The steps of the method disclosed in the embodiments of the present application may be directly embodied as being executed and completed by a hardware decoding processor, or executed and completed by a combination of hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the field, such as random access memory, flash memory, read-only memory, programmable read-only memory, or electrically erasable programmable memory, registers. The storage medium is located in the memory 802, and the processor 803 reads the information in the memory 802, and completes the steps of the foregoing method in combination with its hardware.

It can be understood that the embodiments described in this application can be implemented by hardware, software, firmware, middleware, microcode, or a combination thereof. For hardware implementation, the processing unit can be implemented in one or more application specific integrated circuits (ASICs), digital signal processors (Digital Signal Processing, DSP), digital signal processing devices (DSP Device, DSPD), programmable Logic device (Programmable Logic Device, PLD), Field-Programmable Gate Array (Field-Programmable Gate Array, FPGA), general-purpose processors, controllers, microcontrollers, microprocessors, and others for performing the functions described in this application Electronic unit or its combination. For software implementation, the technology described in this application can be implemented through modules (for example, procedures, functions, etc.) that perform the functions described in this application. The software codes can be stored in the memory and executed by the processor. The memory can be implemented in the processor or external to the processor.

Optionally, as another embodiment, the processor 803 is further configured to execute the steps of the method described in any one of the foregoing embodiments when the computer program is running.

The terminal device in the embodiments of the present application may be implemented in various forms. For example, the terminal equipment may include smart phones, tablet computers, notebook computers, handheld computers, personal digital assistants (Personal Digital Assistant, PDA), portable media players (Portable Media Player, PMP), navigation devices, and wearable devices. , Digital cameras, video cameras and other mobile terminals, and fixed terminals such as digital TVs and desktop computers. Refer to FIG. 14, which shows a schematic diagram of the composition structure of a terminal device provided by an embodiment of the present application. As shown in FIG. 14, the terminal device 90 may at least include the white balance correction device 70 described in any one of the foregoing embodiments.

It should be noted that in this application, the terms "including", "including" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements , And also include other elements not explicitly listed, or elements inherent to the process, method, article, or device. If there are no more restrictions, the element defined by the sentence "including a..." does not exclude the existence of other identical elements in the process, method, article, or device that includes the element.

The serial numbers of the foregoing embodiments of the present application are for description only, and do not represent the superiority or inferiority of the embodiments.

The methods disclosed in the several method embodiments provided in this application can be combined arbitrarily without conflict to obtain new method embodiments.

The features disclosed in the several product embodiments provided in this application can be combined arbitrarily without conflict to obtain new product embodiments.

The features disclosed in the several method or device embodiments provided in this application can be combined arbitrarily without conflict to obtain a new method embodiment or device embodiment.

The above are only specific implementations of this application, but the scope of protection of this application is not limited to this. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application. Should be covered within the scope of protection of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Industrial applicability

In the embodiment of this application, the initial classification result and initial confidence of the light source are determined based on the image data acquired by the camera; the infrared band ratio and color temperature information of the light source are determined based on the data acquired by the color temperature sensor; according to the infrared band ratio, color temperature information and initial The confidence is calculated to obtain the target confidence; when the target confidence is less than the preset threshold, the target classification result is obtained by adjusting the initial classification result; the target white balance parameter is determined according to the target classification result, and the target white balance is determined according to the target white balance Parameters perform white balance correction on the image data; in this way, after the light source is judged once to obtain the initial classification result and initial confidence, the light source is judged twice according to the determined infrared band ratio and color temperature information, which can effectively correct the result after the first judgment. The existence of misjudgment scenes can not only better distinguish indoor/outdoor scenes, but also improve the accuracy of indoor/outdoor judgments, so as to achieve a better and more stable white balance effect, so that the terminal device can have better colors reduction.

Claims (20)

A white balance correction method, the method includes: Based on the image data obtained by the camera, determine the initial classification result and initial confidence of the light source; Based on the data obtained by the color temperature sensor, determine the infrared band ratio and color temperature information of the light source; Calculate according to the infrared band ratio, the color temperature information and the initial confidence to obtain the target confidence; When the target confidence is less than a preset threshold, the target classification result is obtained by adjusting the initial classification result; Determine the target white balance parameter according to the target classification result, and perform white balance correction on the image data according to the determined target white balance parameter. The method according to claim 1, wherein the determining the initial classification result and initial confidence of the light source based on the image data obtained by the camera comprises: Input the image data obtained by the camera into the preset classifier; Through the preset classifier, the initial classification result and initial confidence of the light source are output. The method according to claim 1, wherein the determining the infrared band ratio and color temperature information of the light source based on the data obtained by the color temperature sensor comprises: Obtain the corresponding spectral channel values of multiple spectral bands in the light source through the color temperature sensor; Based on the obtained spectrum channel value, the infrared band ratio and color temperature information of the light source are calculated. The method according to claim 1, wherein when the initial classification result is indoor, the calculation based on the infrared band ratio, the color temperature information and the initial confidence to obtain the target confidence includes: Comparing the infrared band ratio with a first threshold; When the infrared band ratio is greater than or equal to the first threshold, judging the light source according to the infrared band ratio and the color temperature information; If it is determined that the light source is a daylight light source, perform a subtraction operation on the initial confidence level and the first preset value to obtain the target confidence level; If it is determined that the light source is a non-daylight light source, the initial confidence level and the second preset value are added together to obtain the target confidence level. The method according to claim 4, wherein, after the comparing the infrared band ratio with a first threshold, the method further comprises: When the infrared band ratio is less than the first threshold, the initial confidence level and a third preset value are added together to obtain the target confidence level. The method according to claim 1, wherein when the initial classification result is outdoor, the calculation based on the infrared band ratio, the color temperature information and the initial confidence to obtain the target confidence includes: Comparing the infrared band ratio with a second threshold; When the infrared band ratio is greater than or equal to the second threshold, judging the light source according to the infrared band ratio and the color temperature information; If it is determined that the light source is a daylight light source, performing an addition operation on the initial confidence level and a fourth preset value to obtain the target confidence level; If it is determined that the light source is a non-daylight light source, the initial confidence level is subtracted from the fifth preset value to obtain the target confidence level. The method according to claim 6, wherein, after the comparing the infrared band ratio with a second threshold, the method further comprises: When the infrared band ratio is less than the second threshold, the initial confidence level is subtracted from the sixth preset value to obtain the target confidence level. The method according to claim 1, wherein said obtaining a target classification result by adjusting said initial classification result when said target confidence is less than a preset threshold comprises: When the target confidence is less than the preset threshold, the initial classification result is adjusted to determine the target classification result; wherein, If the initial classification result is indoor, it is determined that the target classification result is outdoor; If the initial classification result is outdoor, it is determined that the target classification result is indoor. The method according to claim 1, wherein the method further comprises: When the target confidence is greater than or equal to the preset threshold, it is determined that the target classification result is to maintain the initial classification result; wherein, If the initial classification result is indoor, it is determined that the target classification result is indoor; If the initial classification result is outdoor, it is determined that the target classification result is outdoor. The method according to claim 1, wherein the determining a target white balance parameter according to the target classification result, and performing white balance correction on the image data according to the determined target white balance parameter, comprises: If the target classification result is indoor, call indoor parameters, obtain target white balance parameters from the indoor parameters, and perform white balance correction on the image data according to the target white balance parameters; If the target classification result is outdoor, call outdoor parameters, obtain target white balance parameters from the outdoor parameters, and perform white balance correction on the image data according to the target white balance parameters. The method according to claim 4 or 6, wherein the judging the light source according to the infrared band ratio and the color temperature information comprises: According to the infrared band ratio, query the first color temperature information corresponding to the infrared band ratio from a pre-established comparison table; The light source is judged according to the color temperature information and the first color temperature information obtained by the query. The method according to claim 11, wherein the judging the light source according to the color temperature information and the first color temperature information obtained by the query comprises: Calculating the difference between the color temperature information and the first color temperature information; When the difference is less than a preset difference threshold, selecting the smallest color temperature information from the color temperature information and the first color temperature information; Comparing the minimum color temperature information with a preset color temperature threshold; If the minimum color temperature information is greater than the preset color temperature threshold, determining that the light source is a daylight light source; If the minimum color temperature information is less than or equal to the preset color temperature threshold, it is determined that the light source is a non-daylight light source. A white balance correction device, which includes a determination unit, a calculation unit, an adjustment unit, and a correction unit, wherein: The determining unit is configured to determine the initial classification result and initial confidence of the light source based on the image data obtained by the camera; and is further configured to determine the infrared band ratio and color temperature information of the light source based on the data obtained by the color temperature sensor; The calculation unit is configured to perform calculation according to the infrared band ratio, the color temperature information, and the initial confidence to obtain the target confidence; The adjustment unit is configured to obtain a target classification result by adjusting the initial classification result when the target confidence is less than a preset threshold; The correction unit is configured to determine a target white balance parameter according to the target classification result, and perform white balance correction on the image data according to the determined target white balance parameter. The white balance correction device according to claim 13, wherein the white balance correction device further comprises a comparison unit and a judgment unit; The comparing unit is configured to compare the infrared band ratio with a first threshold when the initial classification result is indoor; The judging unit is configured to judge the light source according to the infrared waveband ratio and the color temperature information when the infrared waveband ratio is greater than or equal to the first threshold; The calculation unit is specifically configured to perform a subtraction operation between the initial confidence level and a first preset value if the light source is determined to be a daylight light source to obtain the target confidence level; and if it is determined that the light source is a non-daylight light source, then The initial confidence level and the second preset value are added together to obtain the target confidence level. The white balance correction device according to claim 14, wherein: The calculation unit is further configured to perform an addition operation on the initial confidence level and a third preset value when the infrared band ratio is less than the first threshold value to obtain the target confidence level. The white balance correction device according to claim 14, wherein: The comparing unit is further configured to compare the infrared band ratio with a second threshold when the initial classification result is outdoor; The judgment unit is further configured to judge the light source according to the infrared waveband ratio and the color temperature information when the infrared waveband ratio is greater than or equal to the second threshold; The calculation unit is specifically configured to, if it is determined that the light source is a daylight light source, add the initial confidence level and a fourth preset value to obtain the target confidence level; and if it is determined that the light source is a non-daylight light source, then Perform a subtraction operation on the initial confidence level and a fifth preset value to obtain the target confidence level. The white balance correction device according to claim 16, wherein: The calculation unit is further configured to perform a subtraction calculation between the initial confidence level and a sixth preset value when the infrared band ratio is less than the second threshold value to obtain the target confidence level. A white balance correction device, the white balance correction device includes a memory and a processor; wherein, The memory is used to store a computer program that can run on the processor; The processor is configured to execute the method according to any one of claims 1 to 12 when running the computer program. A computer storage medium, wherein the computer storage medium stores a white balance correction program, and when the white balance correction program is executed by at least one processor, the method according to any one of claims 1 to 12 is implemented. A terminal device, wherein the terminal device at least comprises the white balance correction device according to any one of claims 13 to 18.

Images