WO2021223185A1

WO2021223185A1 - Camera and alignment method and apparatus therefor, and gimbal

Info

Publication number: WO2021223185A1
Application number: PCT/CN2020/089055
Authority: WO
Inventors: 严毅民; 陈安秋
Original assignee: 深圳市大疆创新科技有限公司
Priority date: 2020-05-07
Filing date: 2020-05-07
Publication date: 2021-11-11
Also published as: CN112655193A

Abstract

Disclosed are a camera alignment method and apparatus. The method comprises: determining a reference camera from a plurality of cameras; determining a white balance gain of the reference camera; determining the current color temperature corresponding to the white balance gain; for one or more target cameras of the other cameras of the plurality of cameras other than the reference camera, performing the following operation to realize white balance alignment between each target camera and the reference camera: determining a white balance gain mapping matrix corresponding to the current color temperature, wherein the white balance gain mapping matrix is used for characterizing a mapping relationship between the white balance gain of the reference camera and a white balance gain of the target camera under the current color temperature; and multiplying the white balance gain of the reference camera by the white balance gain mapping matrix to obtain a first white balance gain of the target camera. According to the present application, the white balance gains of the other cameras are obtained according to the the white balance gain of the reference camera and by means of direct mapping using the white balance gain mapping matrix, thus, the color difference between the other cameras and the reference camera depends on the error in precision during mapping, however, the error in precision during mapping is generally relatively small and relatively stable, and therefore, the camera alignment method used in the present application can ensure that the color difference between different cameras is relatively small and relatively stable.

Description

Camera, alignment method and device thereof, and platform

Technical field

This application relates to the field of camera technology, in particular to a camera alignment method, a camera alignment device, a camera, a pan-tilt, a computer-readable storage medium, and a computer program.

Background technique

In recent years, more and more camera systems support dual cameras or multiple cameras. Due to differences in the response of lenses, filters, and optical sensors between different cameras, the colors of photos or videos taken by different cameras are significantly different.

Summary of the invention

The embodiments of the present application provide a camera alignment method and device, which can solve the technical problem that the colors of photos or videos taken by different cameras in the same camera system have obvious differences.

In a first aspect, an embodiment of the present application provides a camera alignment method, the camera has multiple cameras, and the method includes: determining a reference camera from the multiple cameras; determining a white balance gain of the reference camera; Determine the current color temperature corresponding to the white balance gain; for one or more target cameras among the cameras other than the reference camera in the plurality of cameras, perform the following operations to make each target camera and all White balance alignment between the reference cameras: determining a white balance gain mapping matrix corresponding to the current color temperature, wherein the white balance gain mapping matrix is used to characterize the white balance gain of the reference camera at the current color temperature A mapping relationship with the white balance gain of the target camera; and multiplying the white balance gain of the reference camera by the white balance gain mapping matrix to obtain the first white balance gain of the target camera.

In a second aspect, an embodiment of the present application provides a camera alignment device. The camera has multiple cameras. The device includes a processor and a memory. The memory is used to store instructions. The instruction for implementing the following operations: determining a reference camera from the multiple cameras; determining the white balance gain of the reference camera; determining the current color temperature corresponding to the white balance gain; One or more target cameras in the cameras other than the reference camera, perform the following operations to achieve white balance alignment between each target camera and the reference camera: determine that it corresponds to the current color temperature The white balance gain mapping matrix is used to characterize the mapping relationship between the white balance gain of the reference camera and the white balance gain of the target camera at the current color temperature; and The white balance gain of the reference camera is multiplied by the white balance gain mapping matrix to obtain the first white balance gain of the target camera.

In a third aspect, an embodiment of the present application provides a camera, including: a plurality of cameras; and the camera alignment device according to any one of the above-mentioned second aspects.

In a fourth aspect, an embodiment of the present application provides a pan/tilt head, including: the camera according to any one of the foregoing third aspects.

In a fifth aspect, an embodiment of the present application provides a computer-readable storage medium, including instructions, which when run on a computer, cause the computer to execute the camera alignment method according to any one of the first aspects.

In a sixth aspect, an embodiment of the present application provides a computer program product containing instructions that, when the instructions are run on a computer, cause the computer to execute the camera alignment method described in any one of the first aspects above.

The embodiments of the present application provide a camera alignment method and device. According to the white balance gain of a reference camera, the white balance gain of other cameras is directly mapped through the white balance gain mapping matrix to obtain the white balance gains of other cameras. The reason for the color deviation of the photo or video is the accuracy error during the mapping, and the accuracy error during the mapping is usually relatively small and relatively stable. Therefore, the camera alignment method adopted in this application can ensure that the color deviation of the photos or videos taken by different cameras is small. And it's relatively stable.

Description of the drawings

FIG. 1A is a schematic diagram 1 of an application scenario of a camera alignment method provided by an embodiment of this application;

FIG. 1B is a second schematic diagram of an application scenario of the camera alignment method provided by an embodiment of this application;

2 is a schematic flowchart of a camera alignment method provided by an embodiment of the application;

3A and 3B are schematic diagrams of realizing white balance alignment provided by an embodiment of the application;

4 is a schematic diagram of realizing white balance alignment and color alignment provided by an embodiment of the application;

Fig. 5 is a block diagram of a camera 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.

The camera alignment method provided in the embodiments of the present application can be applied to any camera alignment process that requires white balance alignment and color alignment, and the camera alignment method can be specifically executed by a camera alignment device. The camera alignment device may be a device applied to a camera system with two cameras. Correspondingly, a schematic diagram of an application scenario of the camera alignment method provided in an embodiment of the present application may be as shown in FIG. 1A. Specifically, the processor of the camera alignment device may use one camera of the camera system as a reference camera, and use the camera alignment method provided in the embodiment of the present application to process the other camera of the camera system.

Alternatively, the camera alignment device may also be a device applied to a camera system with three or more cameras. Correspondingly, a schematic diagram of the application scene of the camera alignment method provided by the embodiment of the present application may be as shown in FIG. 1B (in the figure) The case of three cameras is shown). Specifically, the processor of the camera alignment device may use one camera of the camera system as a reference camera, and process other cameras of the camera system using the camera alignment method provided in the embodiments of the present application.

It should be noted that FIGS. 1A and 1B are only schematic diagrams, and do not limit the structure of the camera. For example, the camera may also include a memory for storing related parameters such as white balance gain.

It should be noted that the type of the device including the camera alignment device may not be limited in the embodiment of the present application. The device may be, for example, a smart camera, a smart phone, a smart robot, a tablet computer, and the like.

The camera alignment method provided by the embodiment of the application can directly map the white balance gain of other cameras according to the white balance gain of the reference camera and directly map the white balance gain of other cameras, thereby making the photos taken by the other cameras and the reference camera or The reason for the difference in the color of the video is the accuracy error during the mapping, and the accuracy error during the mapping is usually relatively small and relatively stable. Therefore, the camera alignment method adopted in this application can ensure that the color deviation of photos or videos taken by different cameras is small and relatively stable.

It should be noted that the camera alignment method provided in the embodiments of the present application can be applied to the field of cameras, electronic devices with camera functions, and other fields, and the present application is not limited herein.

Hereinafter, some embodiments of the present application will be described in detail with reference to the accompanying drawings. In the case of no conflict, the following embodiments and features in the embodiments can be combined with each other.

FIG. 2 is a schematic flowchart of a camera alignment method provided by an embodiment of the application. The execution subject of this embodiment may be a camera alignment device, and specifically may be a processor of the camera alignment device. As shown in FIG. 2, the method of this embodiment may be applied to a camera with multiple cameras, for example, and the method may include S201 to S206, for example.

S201: Determine a reference camera from multiple cameras.

In an embodiment, any one of the cameras may be determined as the reference camera of the application, and the remaining cameras may be determined as the target camera of the application. For example, the main camera in the camera may be determined as the reference camera of the application, and the secondary camera may be determined as the target camera of the application.

In another embodiment, one camera designated by the user may be determined as the reference camera of the application, and the remaining cameras may be determined as the target camera of the application.

S202: Determine the white balance gain of the reference camera.

In the embodiments of the present disclosure, various white balance algorithms may be used to calculate the white balance gain (ie, white balance value) of the reference camera. In one embodiment, for example, an adaptive white balance algorithm may be used to calculate the white balance gain of the reference camera.

S203: Determine the current color temperature corresponding to the white balance gain.

In the embodiment of the present disclosure, in S203, various color temperature estimation algorithms may be used to calculate the current color temperature corresponding to the white balance gain of the reference camera (hereinafter referred to as the current color temperature CT). In one embodiment, for example, an adaptive color temperature estimation algorithm may be used to calculate the current color temperature.

S204: Perform the following operations (such as S205 and S206) for one or more target cameras among the cameras in the plurality of cameras other than the reference camera to achieve white balance alignment between each target camera and the reference camera.

S205: Determine a white balance gain mapping matrix corresponding to the current color temperature.

In the embodiment of the present disclosure, the white balance gain mapping matrix corresponding to the current color temperature is used to characterize the mapping relationship between the white balance gain of the reference camera and the white balance gain of the target camera at the current color temperature. It should be understood that under the same color temperature, the white balance gain mapping relationship between different target cameras and the same reference camera may be the same or may be different.

In the embodiment of the present disclosure, before S205, a series of white balance gain matrices used to characterize the mapping relationship between the white balance gain of the reference camera and the white balance gain of each target camera under different color temperatures may be pre-calibrated.

In S205, the current color temperature CT can be used to filter out the white balance gain mapping matrix corresponding to the color temperature adjacent or similar to the current color temperature CT from a series of pre-calibrated white balance gain mapping matrices, and then map the selected white balance gain The matrix is linearly interpolated to obtain the white balance gain mapping matrix corresponding to the current color temperature CT.

Specifically, as an optional embodiment, S205 may include the following operations, for example:

Determine the color temperature close to the current color temperature.

Determine the white balance gain mapping matrix calibrated for the adjacent color temperature.

The white balance gain mapping matrix corresponding to the current color temperature is determined based on the white balance gain mapping matrix calibrated for the adjacent color temperature.

For example, a camera system has two cameras: camera A and camera B. Camera A is the reference camera and camera B is the target camera. When white balance alignment is performed on camera A and camera B, adaptive white balance can be used first. The algorithm calculates the white balance gain of camera A, and then uses the adaptive color temperature estimation algorithm to calculate the current color temperature CT ₁ corresponding to the white balance gain of the camera A, and then selects the current color temperature from the pre-calibrated white balance gain mapping matrix The white balance gain mapping matrix corresponding to the adjacent or similar color temperature of CT _{1 is} then linearly interpolated on the selected white balance gain mapping matrix to obtain the white balance gain mapping matrix corresponding to the _{current color temperature CT 1.}

For example, the current color temperature CT ₁ is 4000K. The two color temperatures of 3000K and 5000K adjacent to 4000K have been calibrated with two corresponding white balance gain mapping matrices. The two white balance gain mapping matrices will be selected for calculation and The white balance gain mapping matrix corresponding to the current color temperature CT _1. And since the 4000K point is located at the center of the two points 3000K and 5000K, the white balance gain mapping matrix corresponding to 4000K may be, for example, the average of two white balance gain mapping matrices corresponding to 3000K and 5000K, respectively.

For another example, the current color temperature CT ₁ is 2000K, and the existing 3000K color temperature adjacent to 2000K has been calibrated with the corresponding white balance gain mapping matrix. The white balance gain mapping matrix will be selected for calculating the corresponding current color temperature CT ₁ White balance gain mapping matrix. Since 2000K points are located below 3000K points, the white balance gain mapping matrix corresponding to 3000K can be directly used as the white balance gain mapping matrix corresponding to 2000K.

For another example, the current color temperature CT ₁ is 6000K, and the existing 5000K color temperature adjacent to 6000K has been calibrated with the corresponding white balance gain mapping matrix. This white balance gain mapping matrix will be selected for calculating the corresponding current color temperature CT ₁ White balance gain mapping matrix. Since 6000K points are located above 5000K points, the white balance gain mapping matrix corresponding to 5000K can be directly used as the white balance gain mapping matrix corresponding to 6000K.

In the embodiment of the present disclosure, if there are multiple target cameras, the calculation method is similar, except that different target cameras and the reference camera have respective white balance gain mapping matrices, which will not be repeated in this application.

Further, in the embodiments of the present disclosure, calibrating the white balance gain mapping matrix between each target camera and the reference camera at different color temperatures may include, for example, performing the following operations at different color temperatures:

Control the reference camera and each target camera to collect images for the white balance card respectively.

The first ratio of the average value of the three RGB channels in the central area of the white balance card is calculated based on the image data collected by the reference camera.

The second ratio of the average value of the three RGB channels in the central area of the white balance card is calculated based on the image data collected by each target camera.

Based on the mapping relationship between the first ratio and the second ratio corresponding to each target camera, a white balance gain mapping matrix used to characterize the white balance gain mapping relationship between the reference camera and the corresponding target camera at the corresponding color temperature is determined.

Taking a camera system with two main and sub-cameras as an example, the main camera and sub-camera can be used to shoot the white balance card separately under different light sources (such as D65, TL84, A and other common light sources and other light sources with adjustable temperature and color). , And calculate the ratio of the average value of the three RGB channels in the central area of the white balance card based on the pictures taken by the two cameras, namely, Rgain=G/R, Ggain=G/G, Bgain=G/B (or Rgain =R/G, Ggain=G/G, Bgain=B/G). Furthermore, the Rgain and Bgain at and near the corresponding color temperature are used to calculate the white balance gain mapping matrix used to characterize the white balance gain mapping relationship between the main camera and the secondary camera at the color temperature. It should be noted that when using Rgain and Bgain to calculate the white balance gain mapping matrix, the weights of Rgain and Bgain of common light sources can be configured to be higher. The method of calculating the white balance gain mapping matrix may include a polynomial fitting method, where the order of the polynomial is not limited in this application.

For example, a camera system has two cameras: camera A and camera B, where camera A is the reference camera, and camera B is the target camera. When calibrating the white balance gain mapping matrix for camera A and camera B, you can use different light sources. Calculate the Rgain _A and Bgain _A of camera A and the Rgain _B and Bgain _{B of} camera B, and use the Rgain and Bgain at and near the corresponding color temperature to calculate the color temperature to characterize the white balance gain between camera A and camera B The white balance gain mapping matrix of the mapping relationship (for example, to calculate the white balance gain mapping matrix corresponding to 4000K, you can use the Rgain and Bgain of camera A and camera B corresponding to the light source in the range of 3500K to 4500K to participate in the calculation. The closer to the weight assigned to 4000K Higher). The following formula 1 is a first-order polynomial fitting (a higher-order polynomial fitting can also be used), which can be used to calculate the white balance gain mapping matrix of the present application.

Formula 1:

Formula 1 can be written in matrix form:

Therefore, the white balance gain mapping matrix T of the present application can be obtained by the following formula, namely formula 2:

Formula 2:

Among them, W is the weight matrix. The weight matrix is a diagonal matrix, that is, each value on the diagonal of the W matrix is a weight value, and each weight value corresponds to a color temperature.

In the embodiments of the present disclosure, the current color temperature is estimated by referring to the white balance gain of the camera, and then the current color temperature and the pre-calibrated white balance gain mapping matrix are used to calculate the white balance gain mapping matrix corresponding to the current color temperature, and then the reference camera's white balance gain mapping matrix is calculated. The white balance gain and the calculated white balance gain mapping matrix map the white balance gain of the target camera. Because this kind of mapping scheme is not a simple linear mapping strongly dependent on color temperature, but on the mapping of the white balance gain mapping matrix, and the white balance gain mapping matrix is calibrated for multiple different light sources on the one hand, and on the other hand Various operations such as translation and rotation can also be performed. Therefore, the mapping that depends on the white balance gain mapping matrix is a complex polynomial fitting. Therefore, the white balance gain conversion accuracy of the present application is higher.

In addition, since the pre-calibrated white balance gain mapping matrix is calibrated under different light sources, the white balance of the camera alignment solution provided in this application is more stable. That is, no matter it is under the same temperature and different spectrum light source or under the same temperature and same spectrum light source, whether it is between cameras of different specifications or between cameras of the same specification, white balance is achieved through the camera alignment solution provided by this application. After alignment, the photos and videos taken by the multiple cameras of the camera system are not prone to color cast.

S206: Multiply the white balance gain of the reference camera by the white balance gain mapping matrix to obtain the first white balance gain of the target camera.

It should be noted that in the embodiments of the present disclosure, in addition to a path as shown in FIG. In addition to achieving white balance alignment with the reference camera, another path as shown in FIG. 3B (hereinafter referred to as the second path, that is, using the white balance statistical value mapping matrix) can also be used to calculate another white balance gain of the target camera, and then The white balance gains calculated through the two channels are merged to achieve white balance alignment between the target camera and the reference camera.

It should be noted that using the dual-channel fusion scheme, since the first channel can effectively enhance the stability of the white balance, while the second channel has a larger adjustable space, this application makes the white balance alignment between different cameras possible. At the same time, stability and adjustability are taken into account. In addition, the first channel and the second channel can be directly mapped on the white balance gain space and the white balance statistical value space respectively. The two mappings are not strongly dependent on the color temperature, so it can effectively improve the different specifications of cameras in the same temperature and different spectrum. The color cast problem under the light source.

In the embodiment of the present disclosure, the inventive idea of the first path is to directly map the white balance gain of the target camera based on the white balance gain of the reference camera. The main error lies in the accuracy error during the mapping. Therefore, the error is relatively stable and there is no The situation of sudden error. In addition, for the situation where the target camera changes the viewing angle of the screen due to zooming, which results in a change in the white balance statistical value, the white balance alignment result given by the first pass is not affected, and stability can be maintained. That is, this application can effectively solve the problem of color jumps caused by image changes before and after zooming when the same camera is zooming through the first path.

In the embodiments of the present disclosure, the inventive concept of the second path is to map the white balance statistics of the target camera based on the white balance statistics of the reference camera, and then use the white balance statistics of the target camera as the input of the adaptive white balance algorithm Calculate the white balance gain of the target camera. Since there are many adjustable parameters in the adaptive white balance algorithm, the second path has a large adjustable space. These parameters can be adjusted to improve the white balance gain conversion precision adjustment, thereby improving the accuracy of the final white balance alignment result sex.

The white balance alignment scheme of dual-path fusion will be described in detail below in conjunction with specific embodiments.

As an optional embodiment, in addition to the above-mentioned steps (that is, the relevant steps of the first path), the method may also include the following operations (that is, the relevant steps of the second path), for example:

Determine the white balance statistical value mapping matrix corresponding to the current color temperature. Wherein, the white balance statistical value mapping matrix corresponding to the current color temperature is used to represent the mapping relationship between the white balance statistical value of the reference camera and the white balance statistical value of each target camera at the current color temperature. It should be understood that under the same color temperature, the white balance statistical value mapping relationship between different target cameras and the same reference camera may be the same or may be different.

Determine the white balance statistical value of the reference camera at the current color temperature.

The white balance statistical value of the reference camera is multiplied by the white balance statistical value mapping matrix corresponding to the current color temperature determined in the foregoing steps to obtain the second white balance gain of each target camera.

The white balance gain of each target camera at the current color temperature is determined based on the first white balance gain and the corresponding second white balance gain.

In the embodiments of the present disclosure, before the white balance statistical value mapping matrix corresponding to the current color temperature is determined, the white balance statistical value used to characterize the white balance of the reference camera at different color temperatures and the white balance statistical value of each target camera can be pre-calibrated. A series of white balance statistical value mapping matrix of the mapping relationship.

When determining the white balance statistical value mapping matrix corresponding to the current color temperature, you can first use the current color temperature CT to filter out the white balance corresponding to the color temperature adjacent or similar to the current color temperature CT from a series of pre-calibrated white balance statistical value mapping matrices The statistical value mapping matrix is then linearly interpolated on the selected white balance statistical value mapping matrix, so as to obtain the white balance statistical value mapping matrix corresponding to the current color temperature CT.

In the embodiment of the present disclosure, the white balance statistical value is based on color separation statistics. The white balance statistics may include, for example, Rgain=G/R, Bgain=G/B (or Rgain=R/G, Bgain=B/G; or the average value of three RGB channels, where the average value of the three RGB channels can be converted to Rgain and Bgain).

Taking the white balance statistics as Rgain=G/R and Bgain=G/B as an example, the white balance statistics corresponding to each color block of the reference camera are multiplied by the corresponding white balance statistics value mapping matrix to obtain the corresponding color blocks of the target camera. Statistic value of white balance. Then, an adaptive white balance algorithm is used to calculate the white balance gain of the target camera on the basis of the calibrated white balance statistical value of the target camera. It should be noted that in the adaptive white balance algorithm, the weight of the white balance statistical value corresponding to the color temperature closer to the current color temperature CT can be assigned a higher weight.

Specifically, as an optional embodiment, determining the white balance statistical value mapping matrix corresponding to the current color temperature may include, for example, the following operations:

Determine the color temperature close to the current color temperature.

Determine the white balance statistical value mapping matrix calibrated for the adjacent color temperature.

The white balance statistical value mapping matrix corresponding to the current color temperature is determined based on the white balance statistical value mapping matrix calibrated for the adjacent color temperature.

For example, a camera system has two cameras: camera A and camera B. Camera A is the reference camera and camera B is the target camera. When white balance alignment is performed on camera A and camera B, adaptive white balance can be used first. The algorithm calculates the white balance gain of camera A, and then uses the adaptive color temperature estimation algorithm to calculate the current color temperature CT ₂ corresponding to the white balance gain of the camera A, and then selects the current color temperature CT 2 from the pre-calibrated white balance statistical value mapping matrix. The white balance statistical value mapping matrix corresponding to the adjacent or similar color temperature of the color temperature CT _{2 is} then linearly interpolated on the selected white balance statistical value mapping matrix to obtain the white balance statistical value mapping matrix corresponding to the _{current color temperature CT 2.}

For example, the current color temperature CT ₂ is 4000K. The two color temperatures of 3000K and 5000K adjacent to 4000K have been calibrated with two corresponding white balance statistical value mapping matrices. The two white balance statistical value mapping matrices will be selected for use Calculate the white balance statistical value mapping matrix corresponding to the current color temperature CT _2. And since the 4000K point is located at the center of the two points 3000K and 5000K, the white balance statistical value mapping matrix corresponding to 4000K may be, for example, the average of two white balance statistical value mapping matrices corresponding to 3000K and 5000K, respectively.

For another example, the current color temperature CT ₂ is 2000K, and the existing 3000K color temperature adjacent to 2000K has been calibrated with the corresponding white balance statistical value mapping matrix. The white balance statistical value mapping matrix will be selected for calculation and the current color temperature CT ₂ Corresponding white balance statistical value mapping matrix. Since 2000K points are located below 3000K points, the white balance statistical value mapping matrix corresponding to 3000K can be directly used as the white balance statistical value mapping matrix corresponding to 2000K.

For another example, the current color temperature CT ₂ is 6000K, and the existing 5000K color temperature adjacent to 6000K has been calibrated with the corresponding white balance statistical value mapping matrix. The white balance statistical value mapping matrix will be selected for calculation and the current color temperature CT ₂ Corresponding white balance statistical value mapping matrix. Since 6000K points are located above 5000K points, the white balance statistical value mapping matrix corresponding to 5000K can be directly used as the white balance statistical value mapping matrix corresponding to 6000K.

In the embodiment of the present disclosure, if there are multiple target cameras, the calculation method is similar, except that different target cameras and the reference camera have their respective white balance statistical value mapping matrices, which will not be repeated in this application.

Further, when determining the white balance statistical value of the reference camera at the current color temperature (for example, assuming a total of 32*32 white balance statistical values, and each white balance statistical value includes G/R and G/B) and corresponding to the current color temperature After the white balance statistical value mapping matrix, multiply each white balance statistical value by the obtained white balance statistical value mapping matrix to obtain the calibrated 32*32 white balance statistical value of the target camera at the current color temperature . Then, these calibrated statistical values are calculated as the input of the adaptive white balance algorithm to obtain the white balance gain of the target camera. Among them, in the adaptive white balance algorithm for calculating the white balance gain of the target camera, the weight of each white balance statistical value can be assigned according to the current color temperature calculated by the reference camera. For example, the closer the color temperature corresponding to the white balance statistical value is to the current color temperature, the higher the weight assigned to the white balance statistical value.

Further, in the embodiments of the present disclosure, calibrating the white balance statistical value mapping matrix at different color temperatures may include, for example, performing the following operations at different color temperatures:

Control the reference camera and each target camera to collect images for the second target color card respectively.

Based on the image data collected by the reference camera, a third ratio of the average value of the three RGB channels in the central area of each color block on the second target color card is calculated.

Based on the image data collected by each target camera, a fourth ratio of the average value of the three RGB channels in the central area of each color block on the second target color card is calculated.

Based on the mapping relationship between the second ratio and the fourth ratio, a white balance statistical value mapping matrix at the corresponding color temperature is determined.

Take a camera system with a main and a sub-camera as an example. The main camera and the sub-camera can be used to shoot Classic Color Checker under different light sources (such as D65, TL84, A and other common light sources and other light sources with adjustable temperature and color). Color card or Digital SG Color Checker color card (or other color card), and calculate the ratio of the average value of the three RGB channels in the central area of each color block on the color card based on the pictures taken by the two cameras, that is, Rgain=G/R, Bgain=G/B (or Rgain=R/G, Bgain=B/G). Furthermore, the Rgain and Bgain at and near the corresponding color temperature are used to calculate a white balance statistical value mapping matrix used to characterize the white balance statistical value mapping relationship between the main camera and the secondary camera at the color temperature. It should be noted that when Rgain and Bgain are used to calculate the white balance statistical value mapping matrix, the weights of Rgain and Bgain of common light sources can be configured higher. The method of calculating the white balance statistical value mapping matrix may include a polynomial fitting method, wherein the order of the polynomial is not limited in this application.

It should be understood that the "calibration method of the white balance statistical value mapping matrix" is similar to the "calibration method of the white balance gain mapping matrix", the difference is that the white balance card data is used when calibrating the "white balance gain mapping matrix", and When calibrating the "white balance statistical value mapping matrix", the color card data containing multiple color blocks (such as Classic Color Checker color card or Digital SG Color Checker color card) is used. More specifically, the calibration of the "white balance statistical value mapping matrix" uses the data of different color patches on the color card.

In addition, as an optional embodiment, determining the white balance gain of the target camera at the current color temperature based on the first white balance gain and the second white balance gain may include, for example, the following operations:

Based on the current color temperature and the Euclidean distance between the first white balance gain and the second white balance gain, the first weight of the first white balance gain and the second weight of the second white balance gain are determined.

Based on the first white balance gain and the second white balance gain, and the first weight and the second weight, the white balance gain of the target camera at the current color temperature is determined.

In the embodiment of the present disclosure, after the two white balance gains are obtained through the above two channels, the current color temperature and the Euclidean distance between the two white balance gains can be adjusted to blend the two white balance gains. Weight, and finally get the white balance gain of each target camera.

For example, the greater the Euclidean distance between the two white balance gains, the higher the weight of the first white balance gain can be configured. The first white balance gain is the white balance gain of the target camera obtained by directly mapping the white balance gain of the reference camera. The main error is the accuracy error during the mapping. Although there is an error, it is relatively stable, so it can be used for white balance alignment. Initial positioning. The second white balance gain is to first map the white balance statistics of the reference camera to obtain the white balance statistics of the target camera, and then use the white balance statistics of the target camera as the input of the adaptive white balance algorithm to calculate the target camera. White balance gain value. Since there are many adjustable parameters in the adaptive white balance algorithm, the adjustment space of the second path is relatively large, and the accuracy of the final white balance gain value of the target camera can be improved through fine adjustment. After the weights of the first white balance gain and the second white balance gain are obtained, the fusion result of the two white balance gains can be obtained by weighted average.

Further, in the embodiments of the present disclosure, in addition to realizing white balance alignment, color alignment can also be realized on this basis.

Specifically, in an embodiment of the present disclosure, as shown in FIG. 4, the method may further include, for example, after achieving white balance alignment between each target camera and the reference camera, for each target camera, executing the target camera and the reference camera. The color alignment operation of the camera.

More specifically, in the embodiment of the present disclosure, the color alignment operation may include the following operations, for example:

Obtain the first color lookup table (such as 3DLUT) calibrated to the reference camera under different color temperatures.

Obtain a second color lookup table calibrated for the target camera at different color temperatures, where the first color lookup table calibrated for the target camera at different color temperatures is aligned with the second color lookup table calibrated for the reference camera at different color temperatures.

Based on the acquired first color lookup table, determine the color lookup table corresponding to the current color temperature for the reference camera (hereinafter referred to as 3DLUT ₁ ), and based on the acquired second color lookup table, determine the color lookup corresponding to the current color temperature for the target camera The table (hereinafter referred to as 3DLUT ₂ ) makes the target camera and the reference camera realize color alignment.

It should be noted that, in the embodiments of the present disclosure, after obtaining 3DLUT ₁ and 3DLUT ₂ , the color of the reference camera based on _{the photos and recorded video of 3DLUT 1} and the color of the target camera based on the photos and recorded video of _{3DLUT 2} Alignment can be achieved.

Further, in the embodiments of the present disclosure, performing color lookup table calibration on the target camera at different color temperatures includes performing the following operations at different color temperatures:

Control the reference camera and the target camera to perform image collection for the first target color card respectively in the same environment.

Based on the image data collected by the reference camera, the reference camera is first calibrated with the CCM matrix (color correction matrix), and then the color look-up table is calibrated. It should be noted that when calibrating the color lookup table for the reference camera, the target value can be a target value specified by an international standard, or a user-specified target value (for example, a target value that has been applied to a certain product).

Based on the image data collected by the target camera and the CCM matrix calibrated to the reference camera, the target camera is calibrated with the CCM matrix to align the color gamut of the target camera with the reference camera.

After aligning the color gamut of the target camera with the reference camera, based on the CCM matrix calibrated for the target camera and the color lookup table calibrated for the reference camera, the target camera is calibrated with a color lookup table, so that the color of the reference camera is calibrated at the same temperature The look-up table is aligned with the color look-up table calibrated to the target camera. It should be noted that when calibrating the color lookup table for the target camera, the output value of the reference camera based on its calibrated color lookup table under the same color temperature can be used as the target value.

It should be noted that, in the embodiments of the present disclosure, after the reference camera based on the output value of the calibrated color lookup table as the target value, the target camera is automatically calibrated with the color lookup table at the same color temperature, the target camera and the reference camera The degree of color alignment can reach a higher level, for example, the degree of color alignment can reach 70 or more.

In order to achieve a higher level of color alignment between the reference camera and the target camera, in another optional embodiment, after the color look-up table of the target camera is automatically calibrated based on the above operation, the color lookup table of the target camera can also be checked for the reference camera and the target camera. The calibrated color look-up table can be adjusted manually. For example, after manual adjustment, the color alignment can reach more than 90.

In addition, in the embodiments of the present disclosure, in order to achieve a better color alignment effect, when the output value of the reference camera is used as the target value of the target camera, the parameter range of the output value can be limited, for example, the obtained color search The diagonal of the table is a straight line, so that the displacement between the actual value and the target value does not exceed the displacement threshold, so that the overall brightness of the color does not change, and the mapping distance of each point in the color lookup table does not exceed the preset value, etc. Wait.

Specifically, on the basis of white balance alignment, the reference camera and the target camera may be used to collect the color card images for the same color card under the same environment and illumination. Then, based on the image collected by the reference camera, the CCM matrix calibration of the reference camera is performed and the color gamut is determined. Then for the difference between the actual value and the target value of each color block on the color card (such as on the 24 color card), first convert the RGB space to the LAB space, and then use the optimization algorithm (such as formula 3) in the LAB space to get 3x3 Matrix makes the actual value of each color block on the color card approach the target value.

Formula 3:

After obtaining the 3x3 matrix of the reference camera, the target camera is calibrated with the same CCM matrix based on the image collected by the target camera, and the same processing is performed on the color card (such as the 24-color card). The target value at this time is replaced with the output value of the aforementioned reference camera based on its calibrated fixed color look-up table under the same color temperature.

The 3DLUT calibration of the reference camera includes shooting 24 color cards and 144 color cards, and optimizing the raw data of the shooting through the optimization algorithm to obtain a 16x16x16 RGB domain mapping 3DLUT, that is, the final color of the reference camera and the reference camera's 3DLUT calibration data.

After applying 3x3 matrix and 3DLUT on the reference camera, get the reference camera color, extract its RGB value, use the target camera to shoot 24 color cards and 144 color cards, use the 3x3 matrix to process the original data shot by the target camera to get the intermediate color, and then The intermediate color is optimized to obtain the 16x16x16 RGB domain mapping 3DLUT, that is, the final color of the target camera is finally obtained, and the 3DLUT calibration data of the target camera is obtained at the same time.

Then the 3DLUT is linearly interpolated according to different color temperatures. At this time, since the white balance of the reference camera and the target camera have been aligned, the color temperature calculated based on the white balance gain can be directly used to interpolate the 3DLUT (such as formula 4). Get the same color as the reference camera.

Formula 4:

result _3dlut =(hig htemp-ct)* _{hightemp 3diut} +(ct-lowtemp)*lowtemp _3dlut

FIG. 5 is a block diagram of a camera provided by an embodiment of the application. As shown in FIG. 5, the camera 500 may include, for example, multiple cameras 510 and a camera alignment device 520. Further, the camera alignment device 520 includes a processor 521 and a memory 522.

Specifically, the memory 522 is used to store instructions, and the processor 521 calls the instructions stored in the memory 522 to perform the following operations: determine a reference camera from a plurality of cameras; determine the white balance gain of the reference camera; The current color temperature corresponding to the balance gain; for one or more target cameras among multiple cameras other than the reference camera, perform the following operations to achieve white balance alignment between each target camera and the reference camera: Determine and current The white balance gain mapping matrix corresponding to the color temperature, where the white balance gain mapping matrix is used to characterize the mapping relationship between the white balance gain of the reference camera and the white balance gain of the target camera at the current color temperature; and the white balance gain of the reference camera Multiply by the white balance gain mapping matrix to obtain the first white balance gain of the target camera.

As an optional embodiment, the processor is further configured to: determine a white balance statistical value mapping matrix corresponding to the current color temperature, where the white balance statistical value mapping matrix is used to characterize the white balance statistical value of the reference camera at the current color temperature The mapping relationship with the white balance statistics of the target camera; determine the white balance statistics of the reference camera at the current color temperature; multiply the white balance statistics of the reference camera with the white balance statistics mapping matrix to obtain the target camera's white balance statistics A second white balance gain; and determining the white balance gain of the target camera at the current color temperature based on the first white balance gain and the second white balance gain.

As an optional embodiment, the processor is further configured to: determine the first weight and the second weight of the first white balance gain based on the current color temperature and the Euclidean distance between the first white balance gain and the second white balance gain. The second weight of the white balance gain; and based on the first white balance gain and the second white balance gain, and the first weight and the second weight, the white balance gain of the target camera at the current color temperature is determined.

As an optional embodiment, the processor is further configured to: after achieving white balance alignment between each target camera and the reference camera, for each target camera, perform a color alignment operation between the target camera and the reference camera.

As an optional embodiment, the color alignment operation includes: obtaining a first color lookup table calibrated to the reference camera at different color temperatures; obtaining a second color lookup table calibrated to the target camera at different color temperatures, where The first color lookup table calibrated for the target camera and the second color lookup table calibrated for the reference camera under different color temperatures are aligned; based on the obtained first color lookup table, the color lookup table corresponding to the current color temperature is determined for the reference camera , And based on the acquired second color lookup table, determine the color lookup table corresponding to the current color temperature for the target camera, so that the target camera and the reference camera realize color alignment.

As an optional embodiment, the processor is also used for different color temperatures: control the reference camera and the target camera to perform image acquisition for the first target color card in the same environment; based on the image data collected by the reference camera, the reference camera The camera is first calibrated by the CCM matrix, and then the color look-up table is calibrated; based on the image data collected by the target camera and the CCM matrix calibrated to the reference camera, the target camera is calibrated with the CCM matrix to align the color gamut of the target camera with the reference camera; and After aligning the color gamut of the target camera with the reference camera, based on the CCM matrix calibrated for the target camera and the color lookup table calibrated for the reference camera, the target camera is calibrated with a color lookup table, so that the color of the reference camera is calibrated at the same temperature The look-up table is aligned with the color look-up table calibrated to the target camera.

As an optional embodiment, the processor is further configured to: determine the color temperature adjacent to the current color temperature; determine the white balance gain mapping matrix calibrated for the adjacent color temperature; and determine based on the white balance gain mapping matrix calibrated for the adjacent color temperature The white balance gain mapping matrix corresponding to the current color temperature.

As an optional embodiment, the processor is also used under different color temperatures: control the reference camera and the target camera to capture images for the white balance card respectively; calculate the three RGB channels in the central area of the white balance card based on the image data collected by the reference camera Calculate the second ratio of the average value of the three RGB channels in the central area of the white balance card based on the image data collected by the target camera; and determine the second ratio based on the mapping relationship between the first ratio and the second ratio Corresponding to the white balance gain mapping matrix at the color temperature.

As an optional embodiment, the processor is further configured to: determine a color temperature adjacent to the current color temperature; determine a white balance statistical value mapping matrix calibrated for the adjacent color temperature; and a white balance statistical value mapping based on the adjacent color temperature calibration The matrix determines the white balance statistical value mapping matrix corresponding to the current color temperature.

As an optional embodiment, the processor is also used under different color temperatures: control the reference camera and the target camera to capture images for the second target color card respectively; calculate the color blocks on the second target color card based on the image data collected by the reference camera The third ratio of the average value of the three RGB channels in the central area; the fourth ratio of the average value of the three RGB channels in the central area of each color patch on the second target color card is calculated based on the image data collected by the target camera; and the fourth ratio is based on the second The mapping relationship between the ratio and the fourth ratio determines the white balance statistical value mapping matrix at the corresponding color temperature.

It should be understood that the camera alignment device provided in this embodiment can be used to implement the technical solutions of the foregoing method embodiments, and its implementation principles and technical effects are similar to those of the method embodiments, and will not be repeated here.

An embodiment of the present application also provides a pan/tilt head, which includes the camera described in the foregoing embodiment.

In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented by software, it can be implemented in the form of a computer program product in whole or in part. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the processes or functions according to the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted through a computer-readable storage medium. Computer instructions can be sent from one website site, computer, server, or data center to another website site, computer via wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) , Server or data center for transmission. The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or a data center integrated with one or more available media. The usable medium may be a magnetic medium (for example, a floppy disk, a hard disk, and a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid state disk (SSD)).

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit it; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: The technical solutions described in the foregoing embodiments can still be modified, or some or all of the technical features can be equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the technical solutions of the embodiments of the present application. Scope.

Claims

A camera alignment method, the camera having multiple cameras, characterized in that the method includes:

Determining a reference camera from the plurality of cameras;

Determining the white balance gain of the reference camera;

Determining the current color temperature corresponding to the white balance gain;

For one or more target cameras among the cameras other than the reference camera among the plurality of cameras, perform the following operations to achieve white balance alignment between each of the target cameras and the reference camera:

Determine the white balance gain mapping matrix corresponding to the current color temperature, where the white balance gain mapping matrix is used to characterize the white balance gain of the reference camera and the white balance gain of the target camera at the current color temperature. The mapping relationship between; and

The white balance gain of the reference camera is multiplied by the white balance gain mapping matrix to obtain the first white balance gain of the target camera.
The method according to claim 1, further comprising:

Determine a white balance statistical value mapping matrix corresponding to the current color temperature, wherein the white balance statistical value mapping matrix is used to characterize the white balance statistical value of the reference camera and the white balance of the target camera at the current color temperature. Balance the mapping relationship between statistical values;

Determining the white balance statistical value of the reference camera at the current color temperature;

Multiplying the white balance statistical value of the reference camera by the white balance statistical value mapping matrix to obtain the second white balance gain of the target camera; and

Determine the white balance gain of the target camera at the current color temperature based on the first white balance gain and the second white balance gain.
The method according to claim 2, wherein the determining the white balance gain of the target camera at the current color temperature based on the first white balance gain and the second white balance gain comprises:

Based on the current color temperature and the Euclidean distance between the first white balance gain and the second white balance gain, the first weight of the first white balance gain and the second weight of the second white balance gain are determined. Two weights; and

Based on the first white balance gain and the second white balance gain, and the first weight and the second weight, the white balance gain of the target camera at the current color temperature is determined.
The method according to any one of claims 1 to 3, further comprising: after achieving white balance alignment between each of the target cameras and the reference camera, for each of the target cameras,

Perform a color alignment operation between the target camera and the reference camera.
The method according to claim 4, wherein the color alignment operation comprises:

Acquiring a first color lookup table calibrated to the reference camera under different color temperatures;

Acquire a second color lookup table calibrated for the target camera at different color temperatures, wherein the first color lookup table calibrated for the target camera at different color temperatures and the second color lookup table calibrated for the reference camera at different color temperatures The color lookup table is aligned;

Based on the acquired first color lookup table, determine the color lookup table corresponding to the current color temperature for the reference camera, and determine the color corresponding to the current color temperature for the target camera based on the acquired second color lookup table The lookup table enables the target camera and the reference camera to achieve color alignment.
The method of claim 5, wherein the calibration of the color lookup table on the target camera under different color temperatures comprises: under different color temperatures,

Controlling the reference camera and the target camera to respectively perform image collection for the first target color card in the same environment;

Based on the image data collected by the reference camera, first perform CCM matrix calibration on the reference camera, and then perform color lookup table calibration;

Performing CCM matrix calibration on the target camera based on the image data collected by the target camera and the CCM matrix calibrated for the reference camera, so that the target camera is aligned with the color gamut of the reference camera; and

After aligning the color gamut of the target camera with the reference camera, based on the CCM matrix calibrated for the target camera and the color lookup table calibrated for the reference camera, the target camera is calibrated with a color lookup table, so that The color lookup table calibrated for the reference camera at the same temperature is aligned with the color lookup table calibrated for the target camera.
The method according to any one of claims 1 to 6, wherein the determining a white balance gain mapping matrix corresponding to the current color temperature comprises:

Determining a color temperature close to the current color temperature;

Determining a white balance gain mapping matrix calibrated for the adjacent color temperature; and

The white balance gain mapping matrix corresponding to the current color temperature is determined based on the white balance gain mapping matrix calibrated for the adjacent color temperature.
The method according to claim 7, wherein calibrating the white balance gain mapping matrix under different color temperatures comprises: under different color temperatures,

Controlling the reference camera and the target camera to collect images for the white balance card respectively;

Calculating a first ratio of the average value of the three RGB channels in the central area of the white balance card based on the image data collected by the reference camera;

Calculating the second ratio of the average value of the three RGB channels in the central area of the white balance card based on the image data collected by the target camera; and

Based on the mapping relationship between the first ratio and the second ratio, a white balance gain mapping matrix at the corresponding color temperature is determined.
The method according to any one of claims 2 to 8, wherein the determining a white balance statistical value mapping matrix corresponding to the current color temperature comprises:

Determining a color temperature close to the current color temperature;

Determining a white balance statistical value mapping matrix calibrated for the adjacent color temperature; and

A white balance statistical value mapping matrix corresponding to the current color temperature is determined based on the white balance statistical value mapping matrix calibrated for the adjacent color temperature.
The method according to claim 9, wherein calibrating the white balance statistical value mapping matrix under different color temperatures comprises: under different color temperatures,

Controlling the reference camera and the target camera to respectively capture images for a second target color card;

Calculating a third ratio of the average value of the three RGB channels of the central area of each color block on the second target color card based on the image data collected by the reference camera;

Calculating the fourth ratio of the average value of the three RGB channels in the central area of each color block on the second target color card based on the image data collected by the target camera; and

Based on the mapping relationship between the second ratio and the fourth ratio, a white balance statistical value mapping matrix at the corresponding color temperature is determined.
A camera alignment device, the camera has a plurality of cameras, wherein the device includes a processor and a memory, the memory is used for storing instructions, and the processor calls the instructions stored in the memory, Used to achieve the following operations:

Determining a reference camera from the plurality of cameras;

Determining the white balance gain of the reference camera;

Determining the current color temperature corresponding to the white balance gain;

For one or more target cameras among the cameras other than the reference camera among the plurality of cameras, perform the following operations to achieve white balance alignment between each of the target cameras and the reference camera:

Determine the white balance gain mapping matrix corresponding to the current color temperature, where the white balance gain mapping matrix is used to characterize the white balance gain of the reference camera and the white balance gain of the target camera at the current color temperature. The mapping relationship between; and

The white balance gain of the reference camera is multiplied by the white balance gain mapping matrix to obtain the first white balance gain of the target camera.
The device according to claim 11, wherein the processor is further configured to:

Determine a white balance statistical value mapping matrix corresponding to the current color temperature, wherein the white balance statistical value mapping matrix is used to characterize the white balance statistical value of the reference camera and the white balance of the target camera at the current color temperature. Balance the mapping relationship between statistical values;

Determining the white balance statistical value of the reference camera at the current color temperature;

Multiplying the white balance statistical value of the reference camera by the white balance statistical value mapping matrix to obtain the second white balance gain of the target camera; and

Determine the white balance gain of the target camera at the current color temperature based on the first white balance gain and the second white balance gain.
The device according to claim 12, wherein the processor is further configured to:

Based on the current color temperature and the Euclidean distance between the first white balance gain and the second white balance gain, the first weight of the first white balance gain and the second weight of the second white balance gain are determined. Two weights; and

Based on the first white balance gain and the second white balance gain, and the first weight and the second weight, the white balance gain of the target camera at the current color temperature is determined.
The device according to any one of claims 11 to 13, wherein the processor is further configured to: after achieving white balance alignment between each of the target camera and the reference camera, for each The target camera performs a color alignment operation between the target camera and the reference camera.
The device of claim 14, wherein the color alignment operation comprises:

Acquiring a first color lookup table calibrated to the reference camera under different color temperatures;

Acquire a second color lookup table calibrated for the target camera at different color temperatures, wherein the first color lookup table calibrated for the target camera at different color temperatures and the second color lookup table calibrated for the reference camera at different color temperatures The color lookup table is aligned;

Based on the acquired first color lookup table, determine the color lookup table corresponding to the current color temperature for the reference camera, and determine the color corresponding to the current color temperature for the target camera based on the acquired second color lookup table The lookup table enables the target camera and the reference camera to achieve color alignment.
The device according to claim 15, wherein the processor is further configured to: under different color temperatures,

Controlling the reference camera and the target camera to respectively perform image collection for the first target color card in the same environment;

Based on the image data collected by the reference camera, first perform CCM matrix calibration on the reference camera, and then perform color lookup table calibration;

Performing CCM matrix calibration on the target camera based on the image data collected by the target camera and the CCM matrix calibrated for the reference camera, so that the target camera is aligned with the color gamut of the reference camera; and

After aligning the color gamut of the target camera with the reference camera, based on the CCM matrix calibrated for the target camera and the color lookup table calibrated for the reference camera, the target camera is calibrated with a color lookup table, so that The color lookup table calibrated for the reference camera at the same temperature is aligned with the color lookup table calibrated for the target camera.
The device according to any one of claims 11 to 16, wherein the processor is further configured to:

Determining a color temperature close to the current color temperature;

Determining a white balance gain mapping matrix calibrated for the adjacent color temperature; and

The white balance gain mapping matrix corresponding to the current color temperature is determined based on the white balance gain mapping matrix calibrated for the adjacent color temperature.
The device according to claim 17, wherein the processor is further configured to: under different color temperatures,

Controlling the reference camera and the target camera to collect images for the white balance card respectively;

Calculating a first ratio of the average value of the three RGB channels in the central area of the white balance card based on the image data collected by the reference camera;

Calculating the second ratio of the average value of the three RGB channels in the central area of the white balance card based on the image data collected by the target camera; and

Based on the mapping relationship between the first ratio and the second ratio, a white balance gain mapping matrix at the corresponding color temperature is determined.
The device according to any one of claims 12 to 18, wherein the processor is further configured to:

Determining a color temperature close to the current color temperature;

Determining a white balance statistical value mapping matrix calibrated for the adjacent color temperature; and

A white balance statistical value mapping matrix corresponding to the current color temperature is determined based on the white balance statistical value mapping matrix calibrated for the adjacent color temperature.
The device according to claim 19, wherein the processor is further configured to: under different color temperatures,

Controlling the reference camera and the target camera to respectively capture images for a second target color card;

Calculating a third ratio of the average value of the three RGB channels of the central area of each color block on the second target color card based on the image data collected by the reference camera;

Calculating the fourth ratio of the average value of the three RGB channels in the central area of each color block on the second target color card based on the image data collected by the target camera; and

Based on the mapping relationship between the second ratio and the fourth ratio, a white balance statistical value mapping matrix at the corresponding color temperature is determined.
A camera including:

Multiple cameras; and

The camera alignment device according to any one of claims 11 to 20.
A pan-tilt, including:

The camera of claim 21.
A computer-readable storage medium, characterized by comprising instructions, which when run on a computer, causes the computer to execute the camera alignment method according to any one of claims 1-10.
A computer program product containing instructions, characterized in that, when the instructions are run on a computer, the computer is caused to execute the camera alignment method according to any one of claims 1-10.