WO2023220866A1 - 云台的控制方法、装置、可移动平台和存储介质 - Google Patents

云台的控制方法、装置、可移动平台和存储介质 Download PDF

Info

Publication number
WO2023220866A1
WO2023220866A1 PCT/CN2022/093081 CN2022093081W WO2023220866A1 WO 2023220866 A1 WO2023220866 A1 WO 2023220866A1 CN 2022093081 W CN2022093081 W CN 2022093081W WO 2023220866 A1 WO2023220866 A1 WO 2023220866A1
Authority
WO
WIPO (PCT)
Prior art keywords
camera
target
preset condition
cameras
determined
Prior art date
Application number
PCT/CN2022/093081
Other languages
English (en)
French (fr)
Inventor
楼致远
张洪记
王映知
Original Assignee
深圳市大疆创新科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳市大疆创新科技有限公司 filed Critical 深圳市大疆创新科技有限公司
Priority to PCT/CN2022/093081 priority Critical patent/WO2023220866A1/zh
Priority to CN202280050281.4A priority patent/CN117730527A/zh
Publication of WO2023220866A1 publication Critical patent/WO2023220866A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00—Details of television systems
    • H04N5/222—Studio circuitry; Studio devices; Studio equipment
    • H04N5/262—Studio circuits, e.g. for mixing, switching-over, change of character of image, other special effects ; Cameras specially adapted for the electronic generation of special effects
    • H04N5/268—Signal distribution or switching
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/14—Systems for two-way working
    • H04N7/15—Conference systems
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast

Definitions

  • the present application relates to the field of control technology, and in particular to a cloud platform control method, device, removable platform and storage medium.
  • the intelligent following function is a very commonly used function in gimbal products. It can automatically identify the target to be followed through the machine learning module, and control the gimbal to make the target to be followed in the preset composition position, realizing intelligent following of the target to be followed. .
  • the display screen seen by the user is captured by the same camera and obtained through image processing.
  • this camera may be a telephoto lens, allowing users to capture details of distant objects and a shallower depth of field, but with a smaller field of view.
  • users will use the lens to take vertical shots to obtain a tall, three-dimensional picture, but the horizontal field of view will become smaller.
  • the image processing process will bring better stability to the picture and other effects, but it will also cause a significant time delay between the picture display and the actual shooting. In these cases, if you control the gimbal to do intelligent following, unexpected following problems will occur.
  • Embodiments of the present application provide a method, device, removable platform, and storage medium for controlling a pan/tilt to solve undesirable following problems that occur when controlling a pan/tilt to perform intelligent following in some cases.
  • embodiments of the present application provide a method for controlling a pan/tilt, where the pan/tilt is used to carry a shooting device, including:
  • the pan/tilt is controlled to rotate to follow the target to be followed.
  • embodiments of the present application provide a control device for a pan/tilt, the pan/tilt being used to carry a shooting device, and the control device includes: a memory and a processor;
  • the memory is used to store program code
  • the processor calls the program code, and when the program code is executed, is used to perform the following operations:
  • the pan/tilt is controlled to rotate to follow the target to be followed.
  • embodiments of the present application provide a PTZ system, including:
  • a pan/tilt, the pan/tilt is used to carry a shooting device;
  • control device of the pan/tilt controls the rotation of the pan/tilt to follow the target to be followed.
  • embodiments of the present application provide a movable platform, including:
  • control device is provided on the movable platform
  • control device controls the rotation of the pan/tilt to follow the target to be followed.
  • embodiments of the present application provide a computer-readable storage medium on which a computer program is stored. When the computer program is executed, the method as described in any one of the first aspects is implemented.
  • the opening operation for the smart function is obtained, and in response to the opening operation, one camera is automatically determined from the multiple cameras of the shooting device as the target camera corresponding to the intelligent follow function, and the target camera corresponding to the target camera is All images are used to control the rotation of the pan/tilt to follow the target to be followed. It is possible to determine a camera from multiple cameras and use the determined image corresponding to the camera to control the rotation of the pan/tilt. This can reduce the risk of directly using a certain camera by default. When the image corresponding to the camera controls the rotation of the gimbal, in some cases there may be following problems such as intelligent tracking that easily loses the target or has poor sensitivity.
  • Figure 1 is a schematic diagram of an application scenario of the PTZ control method provided by the embodiment of the present application.
  • Figure 2 is a schematic flowchart of a cloud platform control method provided by an embodiment of the present application
  • Figure 3 is a schematic flowchart of a cloud platform control method provided by another embodiment of the present application.
  • Figure 4 is a schematic flowchart of a cloud platform control method provided by yet another embodiment of the present application.
  • Figure 5 is a schematic flowchart of a cloud platform control method provided by yet another embodiment of the present application.
  • Figure 6 is a schematic flowchart of a cloud platform control method provided by yet another embodiment of the present application.
  • Figure 7 is a schematic flowchart of a cloud platform control method provided by yet another embodiment of the present application.
  • Figure 8 is a schematic flowchart of a cloud platform control method provided by yet another embodiment of the present application.
  • Figure 9 is a schematic structural diagram of a pan/tilt control device provided by an embodiment of the present application.
  • Figure 10 is a schematic structural diagram of a pan/tilt system provided by an embodiment of the present application.
  • Figure 11 is a schematic structural diagram of a movable platform provided by an embodiment of the present application.
  • FIG 1 is a schematic diagram of an application scenario of a pan/tilt control method provided by an embodiment of the present application.
  • the application scenario may include a pan/tilt 11 and its control device 12.
  • the control device 12 may communicate with the pan/tilt 11. Communication may be carried out by wireless communication or wired communication.
  • the pan/tilt 11 can be used to carry a shooting device, and the photography device can be fixedly or detachably connected to the pan/tilt.
  • the shooting device is equipped with multiple cameras, and the field of view (FOV) of the multiple cameras can be different.
  • the multiple cameras include at least two cameras with different field of view.
  • the plurality of cameras may also include at least one wide-angle camera, at least one ordinary camera, and at least one telephoto camera, wherein the angle of view of the ordinary camera is between the wide-angle camera and the telephoto camera.
  • the fields of view of multiple cameras can overlap so that the same subject can fall within the fields of view of different cameras.
  • the PTZ 11 may include a handheld PTZ or an airborne PTZ.
  • a handheld PTZ such as a handheld stabilizer, a PTZ camera, etc., and an airborne PTZ such as a PTZ mounted on a drone, an unmanned vehicle, etc.
  • the pan/tilt 11 can be a single-axis, two-axis, three-axis or four-axis pan/tilt, etc.
  • the multi-axis pan/tilt is used to improve the stability of the shooting image; taking the pan/tilt 11 as a three-axis pan/tilt as an example, the pan/tilt 11 can be deflected Pan, roll and pitch axis movements.
  • the method provided by the embodiments of the present application may be applied to a PTZ + mobile phone scenario.
  • control device 12 can be used to control the rotation of the pan/tilt 11 according to an image corresponding to a certain camera among the plurality of cameras, so as to follow the target to be followed. Specifically, the control device 12 can automatically identify the target to be followed in the image. And by controlling the pan/tilt 11, the target to be followed is in a preset composition position.
  • This function can be called an intelligent following function.
  • control device 12 can be provided on the pan/tilt 11, and the control device 12 can control the pan/tilt 11 to perform intelligent following; or, the control device 12 can be provided on the shooting device, and when the intelligent following function of the pan/tilt 11 is enabled, The control device 12 mounted on the shooting device can communicate with the pan/tilt 11 through wired or wireless means, and control the rotation of the pan/tilt 11 to achieve intelligent following.
  • the control device 12 can be disposed on a movable platform equipped with the pan/tilt 11. When the intelligent following function of the pan/tilt 11 is enabled, the control device 12 mounted on the movable platform can communicate with the pan/tilt 11 through wired or wireless means. Control the rotation of the pan/tilt 11 to achieve intelligent following.
  • the movable platform can be any type of equipment that can carry the pan/tilt 11, such as a drone, an unmanned vehicle, etc.
  • the images taken by a certain camera will undergo certain processing, for example, the images of the path will be processed based on the preset mode of the shooting device or the special needs of the user.
  • the image will be used as the current display screen.
  • the gimbal's intelligent following function when activated, the gimbal's control device will directly use the image of the channel for machine learning by default.
  • the target to be followed will be in a preset state. Set the composition position to achieve intelligent following of the target to be followed.
  • this method may cause problems with poor following effects. For example, the FOV of the lens corresponding to the image in this path is small, and the target is easily lost when the target is to be followed, or the processing of the image brings greater impact. Time delay leads to low sensitivity of following.
  • the control device 12 directly uses an image corresponding to one of the multiple cameras to control the rotation of the pan/tilt 11 by default
  • the target is obtained
  • one camera is automatically determined from the multiple cameras of the shooting device as the target camera corresponding to the intelligent follow function, and the pan/tilt is controlled to rotate according to the image corresponding to the target camera.
  • the distance between the optical axes of the multiple cameras of the shooting device can be smaller, so that the position of the same shooting object in the images captured by multiple cameras can be smaller, so that when using different cameras corresponding to the image control
  • the pan/tilt rotates it is expected that the preset composition position of the target to be followed can be the same position, thus simplifying the implementation.
  • FIG. 2 is a schematic flowchart of a pan/tilt control method provided by an embodiment of the present application.
  • the execution subject of this embodiment may be the control device 12 in FIG. 1 , and specifically may be the processor of the control device 12 .
  • the method in this embodiment may include:
  • Step 20 Obtain the activation operation for the smart function.
  • the user's activation operation for the smart function can be obtained.
  • the activation operation includes various control methods such as voice control, button control, and touch screen.
  • Step 21 In response to the opening operation, automatically determine one camera from the multiple cameras of the shooting device as the target camera corresponding to the intelligent follow function.
  • the timing of determining the target camera can be flexibly implemented.
  • the activation operation for the intelligent follow function can be obtained, and in response to the activation operation, one camera is determined from multiple cameras of the shooting device as the target camera corresponding to the intelligent follow function, thereby determining when the user turns on the intelligent follow function.
  • the target camera corresponding to the intelligent following function can be flexibly implemented.
  • step 21 you can directly perform step 21 without considering whether there is a problem of poor following effect when using the image corresponding to the default camera to control the pan/tilt rotation to follow the target to be followed.
  • the default camera can be the camera in which the corresponding image of the multiple cameras is used for the current display screen (hereinafter it can be recorded as the first camera, that is, the image taken by the first camera is used for the current display screen).
  • the default camera is mainly used. Take the first camera as an example for detailed explanation.
  • any control system cannot completely control the error to 0, and its error band is a range value.
  • the control error is, for example, ⁇ 10°; and, since the control error is usually certain, the target to be followed moves at the same angle relative to the gimbal, the smaller the FOV, the better the shooting of the target to be followed. The greater the proportion of motion in the picture, the easier it is for the target to be followed to be lost in the picture.
  • the control error of the intelligent following system is ⁇ 10°
  • its FOV can be 74°
  • usually there will be no loss of target but if it is at a portrait focal length (for example, 50mm) or a telephoto range (for example, 70mm), its FOV becomes smaller.
  • a portrait focal length for example, 50mm
  • a telephoto range for example, 70mm
  • the target to be followed deviates from the preset composition position, resulting in a deviation. After this deviation is machine learned, the gimbal can start to rotate in response to the deviation so that the target to be followed is in the preset composition. Position, the shorter the time of this process, the less obvious the delay in following the picture, and the higher the sensitivity of the following. Therefore, if the image used for machine learning by the follow function itself has a certain delay, the delay of the follow image will be increased and the sensitivity of the follow will be reduced.
  • satisfying the preset condition may include the first camera satisfying the preset condition (hereinafter referred to as the first preset condition), and/or the current display screen satisfying the preset condition (hereinafter referred to as the first preset condition) recorded as the second preset condition).
  • the first camera meets the first preset condition, it means that the image corresponding to the first camera is used to control the pan/tilt rotation to follow the target to be followed. There is a problem that the target is easily lost due to the small FOV of the lens.
  • the first camera does not meet the first preset condition.
  • the preset condition can mean using the image corresponding to the first camera to control the pan/tilt rotation to follow the target to be followed. There is no problem of easily losing the target due to the small FOV of the lens.
  • the horizontal FOV of the first camera is larger.
  • the vertical FOV of the first camera is larger. Therefore, when the first camera is in vertical shooting mode, if you use the image corresponding to the first camera to control the gimbal rotation, it will increase the risk of losing the target due to the small FOV of the lens.
  • the target to be followed is more likely to be lost in the picture. Therefore, when there is a camera with a larger field of view among multiple cameras than the first camera, that is, when the first camera does not have the largest field of view among the multiple cameras, it can be considered that the camera corresponding to the first camera is used. Compared with using other cameras with a larger field of view, using one image to control the gimbal rotation will increase the risk of losing the target due to the small FOV of the lens.
  • the first preset condition may include that the first camera is in vertical shooting mode, and/or there is a camera with a larger field of view among the multiple cameras than the first camera. It should be understood that the first camera is in vertical shooting mode and there is a camera with a larger field of view among the multiple cameras than the first camera. The first camera is in vertical shooting mode or there is a camera with a larger field of view than the first camera among the multiple cameras. For larger cameras, it can be understood that the first camera meets the first preset condition.
  • the current display screen meets the second preset condition, it can mean that the image corresponding to the first camera is used to control the pan/tilt rotation. There is a problem of poor sensitivity due to the large time delay of the image. If the current display screen does not meet the second preset condition, it can mean using The image corresponding to the first camera controls the rotation of the pan/tilt, and there is no problem of poor sensitivity due to long image delays.
  • the second preset condition may include that the difference between the display time of the current display picture and the shooting time of the current display picture is greater than the preset threshold, that is, the current display picture is processed for display after shooting, and the processed The time is so long that when using the processed current display screen for intelligent following, a large delay can easily lead to poor sensitivity.
  • the second preset condition may include that the current display screen has undergone electronic image stabilization. deal with.
  • the target camera may be the first camera, or the target camera may be one camera among multiple cameras except the first camera.
  • the target camera in this application refers to the camera corresponding to the image used to control the rotation of the pan/tilt.
  • Step 22 Control the pan/tilt rotation according to the image corresponding to the target camera to follow the target to be followed.
  • the image corresponding to the target camera may be an image output by the target camera, or may be an image obtained after processing the image output by the target camera.
  • the image corresponding to the target camera may be the image to be displayed of the target camera; and/or, when the target camera is one of multiple cameras except the first camera.
  • the image corresponding to the target camera can be the image output by the target camera.
  • machine learning can be used to automatically identify the target to be followed in the image corresponding to the target camera, and the pan/tilt is controlled so that the target to be followed is in a preset composition position.
  • the implementation method of controlling the rotation of the pan/tilt based on the image to follow the target to be followed please refer to the specific description in the related art, and will not be described again here.
  • the control method of the pan/tilt determines one camera from multiple cameras of the shooting device as the target camera corresponding to the intelligent following function, and controls the pan/tilt rotation according to the image corresponding to the target camera to follow the target.
  • the image corresponding to a certain camera controls the rotation of the gimbal, it may easily lead to a problem of poor following effect.
  • FIG 3 is a schematic flow chart of a pan/tilt control method provided by another embodiment of the present application. Based on the embodiment shown in Figure 2, this embodiment mainly describes determining one camera from multiple cameras of the shooting device as an intelligent follower. An optional implementation of the target camera corresponding to the function. As shown in Figure 3, the method in this embodiment may include:
  • Step 31 When it is determined that the first camera satisfies the first preset condition or the current display screen satisfies the second preset condition, determine one camera from the plurality of cameras as the target camera corresponding to the smart follow function.
  • both the first camera satisfying the first preset condition and the current display screen satisfying the second preset condition can be used as trigger conditions for determining the target camera from multiple cameras.
  • the two situations in which the first camera satisfies the first preset condition as a trigger condition and the current display screen satisfies the second preset condition as respective trigger conditions are described in detail below.
  • the first camera meets the first preset condition.
  • step 31 may specifically include: when determining that the first camera meets the first preset condition, determining the camera with the largest field of view among the multiple cameras as the target camera corresponding to the smart follow function; If one camera does not meet the first preset condition, the first camera is determined as the target camera corresponding to the intelligent follow function.
  • the multiple cameras include at least one wide-angle camera and at least one ordinary camera.
  • the angle of view of the ordinary camera is smaller than that of the wide-angle camera.
  • the image corresponding to the camera is used to display the picture, and the image corresponding to the wide-angle camera is used to control the rotation of the pan/tilt.
  • the image corresponding to the wide-angle camera with the largest field of view is selected to control The pan/tilt rotates.
  • the vertical FOV of the same camera changes with its horizontal FOV in actual applications, there can be a fixed multiple relationship between the two. Therefore, when determining the target camera from multiple cameras, you only need to compare the horizontal FOV or vertical FOV of the camera. That is, for example, according to the horizontal FOV of the multiple cameras, the camera with the largest horizontal FOV among the multiple cameras can be determined as the target camera corresponding to the smart follow function.
  • step 31 may specifically include: when determining that a camera meets the first preset condition, if the currently displayed picture does not meet the second preset condition, determining the first camera as the camera corresponding to the smart follow function.
  • the target camera when it is determined that the first camera meets the first preset condition, and if the current display screen meets the second preset condition, the camera with the largest field of view among the multiple cameras except the first camera is determined as the camera corresponding to the intelligent follow function.
  • the multiple cameras include at least one wide-angle camera and at least one ordinary camera.
  • the angle of view of the ordinary camera is smaller than that of the wide-angle camera.
  • the first camera is an ordinary camera
  • the image corresponding to the ordinary camera is used as the current
  • the currently displayed picture satisfies the second preset condition.
  • step 31 may specifically include: when it is determined that the current display screen meets the second preset condition, determining the camera with the largest field of view among the multiple cameras except the first camera as the camera corresponding to the intelligent follow function.
  • Target camera For example, the current display screen has electronic stabilization turned on, indicating that the current display screen has a significant time delay. This image is used for intelligent following, which will cause the sensitivity of the following to decrease. Therefore, the image output by the current camera, that is, the first camera, is not selected for intelligent tracking. Follow, and select other cameras except the first camera to follow.
  • the camera with the largest field of view among the other cameras is selected as the target camera. That is to say, when it is determined that there is a problem of poor sensitivity in using the image corresponding to the first camera to control the movement of the pan/tilt, you can temporarily ignore the problem of using the image corresponding to the first camera to control the movement of the pan/tilt and there is no problem of easily losing track of the target, and select When using multiple cameras other than the first camera, the image corresponding to the camera with the largest field of view is considered to control the pan/tilt rotation.
  • step 31 may specifically include: when determining that the current display screen meets the second preset condition, if the first camera does not meet the first preset condition, determine the first camera as the camera corresponding to the smart follow function.
  • Target camera when it is determined that the current display screen meets the second preset condition, if the first camera meets the first preset condition, the camera with the largest field of view among the multiple cameras is determined as the target camera corresponding to the intelligent follow function.
  • the current display screen has electronic stabilization turned on, which indicates that the current display screen has a significant delay.
  • This image is used for intelligent following, which will cause the sensitivity of the following to be reduced. However, if the sensitivity of the following is not high, and in order to pursue the goal of prevention If the current camera, that is, the first camera is not in vertical shooting mode, or the first camera has the largest field of view compared to other cameras, the first camera will still be selected as the target camera, and the image output by the first camera will be selected for intelligent following.
  • the current camera that is, the first camera is in vertical shooting mode, or there is a camera with a larger field of view than the first camera among the multiple cameras
  • the camera with the largest field of view among the multiple cameras is selected as the target camera, and in the When the image corresponding to one camera is displayed on the current screen, select the image corresponding to the target camera for intelligent following.
  • the pan-tilt rotation can still be controlled based on the image corresponding to the first camera. If it exists, the pan-tilt rotation can be controlled based on the image corresponding to the camera with the largest field of view among the multiple cameras.
  • step 31 it may also include: in response to the opening operation, determining whether the first camera satisfies the first preset condition, and/or determining whether the currently displayed picture satisfies the second preset condition.
  • Step 32 Control the pan/tilt to rotate according to the image corresponding to the target camera to follow the target to be followed.
  • step 32 is similar to step 22 and will not be described again.
  • the control method of the pan/tilt provided by this embodiment automatically determines one camera from multiple cameras of the shooting device as the intelligent camera when it is determined that the first camera satisfies the first preset condition or the current display screen satisfies the second preset condition.
  • the gimbal rotation allows you to flexibly select the camera corresponding to the image corresponding to the image used to control the gimbal rotation, thereby reducing the problem of poor following effects due to the direct default use of the image corresponding to a certain camera to control the gimbal rotation.
  • Figure 4 is a schematic flow chart of a pan/tilt control method provided by yet another embodiment of the present application. Based on the embodiment shown in Figure 2, this embodiment mainly describes how to determine one camera from among multiple cameras of the shooting device as an intelligent follower. Another optional implementation of the target camera corresponding to the function. As shown in Figure 4, the method in this embodiment may include:
  • Step 41 If the first camera does not meet the first preset condition, determine the first camera as the target camera corresponding to the intelligent follow function;
  • Step 42 If the first camera meets the first preset condition, determine the camera with the largest field of view among the multiple cameras as the target camera corresponding to the intelligent following function;
  • Step 43 Control the pan/tilt to rotate according to the image corresponding to the target camera to follow the target to be followed.
  • the first camera is the camera whose corresponding image among the plurality of cameras is used for the current display screen.
  • the first preset condition may include that the first camera is in vertical shooting mode, and/or there is a camera with a larger field of view among the multiple cameras than the first camera.
  • a startup operation for the smart follow function can be obtained, and in response to the startup operation, it is determined whether the first camera meets the first preset condition.
  • the first camera does not meet the first preset condition, it can mean that using the image corresponding to the first camera to control the gimbal rotation does not cause the problem of easily losing the target due to the small FOV of the lens.
  • the pan-tilt rotation can be controlled based on the image corresponding to the first camera.
  • the first camera meets the first preset condition, which can mean that using the image corresponding to the first camera to control the gimbal rotation has the problem of easily losing the target due to the small FOV of the lens.
  • the control method of the pan/tilt can be as shown in Figure 5 . Show.
  • the FOV of each camera in the multiple cameras can be obtained in step 51; the opening operation for the intelligent follow function can be obtained in step 52; in step 53, in response to the opening operation, it can be determined Whether the first camera is for vertical shooting; if the first camera is not for vertical shooting, then in step 54 it can be determined whether there is a camera with a larger field of view than the first camera among the multiple cameras; if the first camera is for vertical shooting or there is a camera with a larger field of view than the first camera among the multiple cameras, then in step 55, the camera with the largest field of view among the multiple cameras can be determined as the target camera, so as to correspond to the camera with the largest field of view.
  • the first camera is determined as the target camera to control the pan/tilt rotation based on the image corresponding to the first camera.
  • steps 53 and 54 can be changed, that is, in step 53, in response to the opening operation, it can be determined whether there is a larger field of view angle than the first camera among the multiple cameras. If there is no camera with a larger field of view than the first camera, it may be determined in step 54 whether the first camera is a vertical camera.
  • step 53 in response to the opening operation, it can be determined whether the first camera is for vertical shooting. If the first camera is for vertical shooting, then in step 56, the camera with the largest field of view among the multiple cameras can be Determine it as the target camera to control the pan/tilt rotation based on the image corresponding to the camera with the largest field of view; if the first camera is not a vertical shot, determine the first camera as the target camera. For example, there is only step 54. In step 54, in response to the opening operation, it may be determined whether there is a camera with a larger field of view than the first camera among the multiple cameras.
  • step 56 the camera with the largest field of view among the multiple cameras can be determined as the target camera to control the pan/tilt rotation based on the image corresponding to the camera with the largest field of view; if among the multiple cameras, If there is no camera with a larger field of view than the first camera, that is, the first camera has the largest field of view among multiple cameras, then in step 56, the first camera can be determined as the target camera to determine the target camera according to the first camera. The image corresponding to the camera controls the rotation of the pan/tilt.
  • the pan/tilt Through the control method of the pan/tilt provided by this embodiment, it is possible to control the pan/tilt based on the image corresponding to the first camera without the problem of easily losing the target when using the image corresponding to the first camera to control the pan/tilt rotation.
  • Rotation when using the image corresponding to the first camera to control the pan/tilt rotation has the problem of easily losing the target, the pan/tilt rotation can be controlled based on the image corresponding to the camera with the largest field of view among multiple cameras, allowing flexible selection of control
  • the camera corresponding to one image corresponding to the gimbal rotation can reduce the problem of poor following effect caused by directly defaulting to using the image corresponding to a certain camera to control the gimbal rotation.
  • Figure 6 is a schematic flow chart of a pan/tilt control method provided by another embodiment of the present application. Based on the embodiment shown in Figure 2, this embodiment mainly describes determining one camera from multiple cameras of the shooting device as an intelligent follower. Another optional implementation of the target camera corresponding to the function. As shown in Figure 6, the method in this embodiment may include:
  • Step 61 If the currently displayed picture does not meet the second preset condition, determine the first camera as the target camera corresponding to the intelligent follow function;
  • Step 62 If the current display screen meets the second preset condition, determine whether there is a camera with a larger field of view than the first camera among the multiple cameras. If there is no camera with a larger field of view than the first camera among the multiple cameras, If there is a larger camera, the first camera is determined as the target camera. If there is a camera with a larger field of view among the multiple cameras than the first camera, the camera with the largest field of view among the multiple cameras is determined as the target camera.
  • Step 63 Control the pan/tilt to rotate according to the image corresponding to the target camera to follow the target to be followed.
  • the first camera is the camera whose corresponding image among the plurality of cameras is used for the current display screen.
  • the second preset condition may include that the current display screen has undergone electronic anti-shake processing.
  • the current display screen does not meet the second preset condition, it can mean that using the image corresponding to the first camera to control the pan/tilt rotation does not have the problem of poor sensitivity due to large image delay.
  • the pan-tilt rotation can be controlled based on the image corresponding to the first camera.
  • the current display screen meets the second preset condition, which can mean that using the image corresponding to the first camera to control the pan/tilt rotation has the problem of poor sensitivity due to the large image delay.
  • the image delay can be prioritized. In the case of poor sensitivity, the impact of the lens' field of view on tracking and loss of the target should also be considered.
  • the gimbal control method can be as shown in FIG. 7 , for example.
  • the FOV of each camera in the multiple cameras can be obtained in step 71; in step 72, the opening operation for the intelligent follow function can be obtained; in step 73, in response to the opening operation, it can be determined Whether EIS is turned on; if it is turned on, in step 74, it is judged whether there is a camera with a larger field of view than the first camera among the multiple cameras, that is, it is judged whether other cameras among the multiple cameras except the first camera have a larger field of view than the first camera.
  • the camera with the largest field of view among the multiple cameras is determined as the target camera to control the pan/tilt rotation based on the image corresponding to the camera with the largest field of view. If it does not exist , then the first camera is determined as the target camera; if EIS is not turned on, then in step 75, the first camera can be determined as the target camera to control the pan/tilt rotation according to the image corresponding to the first camera.
  • EIS electronic anti-shake processing
  • the pan/tilt rotation can be controlled based on the image corresponding to the camera with the largest field of view among multiple cameras, so that the pan/tilt can be flexibly selected to be controlled. Rotate the camera corresponding to the image used, thereby reducing the problem of poor following effects caused by directly using the image corresponding to a certain camera to control the gimbal rotation.
  • FIG 8 is a schematic flow chart of a pan/tilt control method provided by yet another embodiment of the present application. Based on the embodiment shown in Figure 2, this embodiment mainly describes how to determine one camera from among multiple cameras of the shooting device as an intelligent follower. Another optional implementation of the target camera corresponding to the function. As shown in Figure 8, the method in this embodiment may include:
  • Step 81 If the first camera does not meet the first preset condition and the current display screen does not meet the second preset condition, determine the first camera as the target camera corresponding to the smart follow function;
  • Step 82 If the first camera meets the first preset condition or the current display screen meets the second preset condition, determine the camera with the largest field of view among the multiple cameras as the target camera corresponding to the intelligent follow function;
  • Step 83 Control the pan/tilt to rotate according to the image corresponding to the target camera to follow the target to be followed.
  • the first camera is the camera whose corresponding image among the plurality of cameras is used for the current display screen.
  • the first preset condition may include that the first camera is in vertical shooting mode, and/or the first camera does not have the largest field of view among multiple cameras.
  • the second preset condition may include that the current display screen has undergone electronic anti-shake processing.
  • the first camera does not meet the first preset condition, it can mean that the image corresponding to the first camera is used to control the gimbal rotation. There is no problem of easily losing the target due to the small FOV of the lens. It can mean that the current display screen does not meet the second preset condition. Using the image corresponding to the first camera to control the pan-tilt rotation does not cause the problem of poor sensitivity due to large image delay. Through the above step 81, it can be realized that the target is not easily lost when using the image corresponding to the first camera to control the pan-tilt rotation. And in the case of poor sensitivity, the pan/tilt rotation can be controlled based on the image corresponding to the first camera.
  • the first camera meets the first preset condition, it can mean that using the image corresponding to the first camera to control the gimbal rotation has the problem of easily losing the target due to the small FOV of the lens.
  • the current display screen satisfying the second preset condition can mean that using the first There is a problem of poor sensitivity due to the long image delay in the image control pan/tilt rotation corresponding to the camera.
  • the pan/tilt rotation can be controlled based on the image corresponding to the camera with the largest field of view among multiple cameras.
  • control the pan/tilt provided by this embodiment, it is possible to use the image corresponding to the first camera to control the pan/tilt rotation without problems of easy loss of the target and poor sensitivity. Control the pan/tilt rotation.
  • the image corresponding to the first camera to control the pan/tilt rotation has the problem of easily losing the target or having poor sensitivity, you can control the pan/tilt rotation based on the image corresponding to the camera with the largest field of view among multiple cameras.
  • the rotation of the pan/tilt can be controlled based on the image corresponding to the first camera, so that the camera corresponding to the image used by the control device to control the pan/tilt can be flexibly determined, thereby reducing the occurrence of direct use of the image corresponding to one camera to cause the pan/tilt to rotate. control, resulting in poor follow-up effect.
  • Figure 9 is a schematic structural diagram of a pan/tilt control device provided by an embodiment of the present application.
  • the device 90 may include a processor 91 and a memory 92.
  • the memory 92 is used to store program codes
  • the processor 91 calls the program code, and when the program code is executed, is used to perform the following operations:
  • the pan/tilt is controlled to rotate to follow the target to be followed.
  • the PTZ control device provided in this embodiment can be used to execute the technical solutions of the foregoing method embodiments. Its implementation principles and technical effects are similar to those of the method embodiments, and will not be described again here.
  • FIG. 10 is a schematic structural diagram of a pan/tilt system provided by an embodiment of the present application.
  • the pan/tilt system 100 may include: a pan/tilt 101 and a control device 102, where the pan/tilt 101 is used to carry a shooting device;
  • the control device 102 is used to control the rotation of the pan/tilt 101 to follow the target to be followed.
  • the specific implementation of the control device 102 reference may be made to the specific description of the embodiment shown in FIG. 9, which will not be described again here.
  • FIG 11 is a schematic structural diagram of a movable platform provided by an embodiment of the present application.
  • the movable platform 110 includes a movable platform body 111, a pan/tilt 112 and a control device 113.
  • the pan/tilt 112 is mounted on On the movable platform body 111, a control device 113 is provided on the movable platform.
  • the control device 113 is used to control the rotation of the pan/tilt to follow the target to be followed.
  • the specific implementation of the control device 102 reference may be made to the specific description of the embodiment shown in FIG. 9, which will not be described again here.
  • control device 113 is installed on the movable platform body 111 as an example.
  • Embodiments of the present application also provide a computer-readable storage medium on which a computer program is stored. When the computer program is executed, the method described in the above method embodiment is implemented.
  • the aforementioned program can be stored in a computer-readable storage medium.
  • the steps including the above method embodiments are executed; and the aforementioned storage media include: ROM, RAM, magnetic disks or optical disks and other media that can store program codes.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Studio Devices (AREA)
  • Accessories Of Cameras (AREA)

Abstract

一种云台（11，101，112）的控制方法、装置（12，90，102，113）、可移动平台（110）和存储介质。该方法包括：获取针对智能跟随功能的开启操作，响应于开启操作，自动地从拍摄装置的多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头；根据目标摄像头对应的一路图像，控制云台（11，101，112）转动，以对待跟随目标进行跟随。

Description

云台的控制方法、装置、可移动平台和存储介质 技术领域
本申请涉及控制技术领域，尤其涉及一种云台的控制方法、装置、可移动平台和存储介质。
背景技术
智能跟随功能是云台产品中一个十分常用的功能，它可以通过机器学习模块自动识别待跟随目标，通过云台的控制使得待跟随目标处于预设的构图位置，实现对于待跟随目标的智能跟随。
目前，云台搭载的拍摄装置上可以设置有多个摄像头，以适应不同的拍摄需求。通常，用户看到的显示画面是由同一个摄像头拍摄并经过图像处理得到的。在一些情况下，这个摄像头可能是长焦镜头，用户可以拍摄到远处景物的细节和更浅的景深，但是视场角更小。在一些情况下，用户会利用镜头做竖拍，以获得高大立体的画面，但是水平方向的视场角会变小。此外，图像处理的过程会给画面带来更好的稳定性等其他效果，但是也会导致画面显示与实际拍摄存在明显的时间延迟。在这些情况下，如果控制云台做智能跟随，会出现不期望的跟随问题。
发明内容
本申请实施例提供一种云台的控制方法、装置、可移动平台和存储介质，用以解决一些情况下控制云台做智能跟随不期望出现的跟随问题。
第一方面，本申请实施例提供一种云台的控制方法，所述云台用于搭载拍摄装置，包括：
获取针对智能功能的开启操作；
响应于所述开启操作，自动地从所述拍摄装置的多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头；
根据所述目标摄像头对应的一路图像，控制所述云台转动，以对待跟随目标进行跟随。
第二方面，本申请实施例提供一种云台的控制装置，所述云台用于搭载拍摄装置，所述控制装置包括：存储器和处理器；
所述存储器，用于存储程序代码；
所述处理器，调用所述程序代码，当程序代码被执行时，用于执行以下操作：
获取针对智能功能的开启操作；
响应于所述开启操作，自动地从所述拍摄装置的多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头；
根据所述目标摄像头对应的一路图像，控制所述云台转动，以对待跟随目标进行跟随。
第三方面，本申请实施例提供一种云台系统，包括：
云台，所述云台用于搭载拍摄装置；以及，
第二方面中任一项所述的云台的控制装置，所述控制装置控制所述云台转动，以对待跟随目标进行跟随。
第四方面，本申请实施例提供一种可移动平台，包括：
可移动平台本体；
云台，搭载于所述可移动平台本体上；
第二方面中任一项所述的控制装置，设置于所述可移动平台上；
其中，所述控制装置控制所述云台转动，以对待跟随目标进行跟随。
第五方面，本申请实施例提供一种计算机可读存储介质，其上存储有计算机程序，当所述计算机程序被执行时，实现如第一方面中任一项所述的方法。
在本申请实施例中，获取针对智能功能的开启操作，响应于所述开启操作，自动地从拍摄装置的多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，并根据目标摄像头对应的一路图像，控制云台转动，以对待跟随目标进行跟随，实现了从多个摄像头中确定一个摄像头，并使用确定的该摄像头对应的一路图像控制云台转动，从而能够减少由于直接默认使用某一个摄像头对应的一路图像控制云台转动时，一些情况下出现智能跟随容易丢失目标或灵敏度差的跟随问题。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作一简单地介绍，显而易见地，下面描述中的附图是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为本申请实施例提供的云台的控制方法的应用场景示意图；
图2为本申请一实施例提供的云台的控制方法的流程示意图；
图3为本申请另一实施例提供的云台的控制方法的流程示意图；
图4为本申请又一实施例提供的云台的控制方法的流程示意图；
图5为本申请又一实施例提供的云台的控制方法的流程示意图；
图6为本申请又一实施例提供的云台的控制方法的流程示意图；
图7为本申请又一实施例提供的云台的控制方法的流程示意图；
图8为本申请又一实施例提供的云台的控制方法的流程示意图；
图9为本申请一实施例提供的云台的控制装置的结构示意图；
图10为本申请一实施例提供的云台系统的结构示意图；
图11为本申请一实施例提供的可移动平台的结构示意图。
具体实施方式
为使本申请实施例的目的、技术方案和优点更加清楚，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
图1为本申请实施例提供的云台的控制方法的应用场景示意图，如图1所示，该应用场 景中可以包括云台11及其控制装置12，控制装置12可以与云台11之间进行通信，可以是采用无线通信方式进行通信，也可以是采用有线通信方式进行通信。
其中，云台11可以用于搭载拍摄装置，拍摄装置可以与云台固定连接或者可拆卸连接。拍摄装置上设置有多个摄像头，多个摄像头的视场角(field of view，FOV)可以不同，例如多个摄像头至少包括两个视场角不同的摄像头。具体地，多个摄像头也可以包括至少一个广角摄像头，至少一个普通摄像头和至少一个长焦摄像头，其中，普通摄像头的视场角介于广角摄像头和长焦摄像头之间。多个摄像头的视野范围可以重叠，以使得同一拍摄对象可以落入不同摄像头的视野范围内。
示例性的，云台11可以包括手持云台或机载云台，手持云台如手持稳定器、云台相机等，机载云台如无人机、无人车等搭载的云台；云台11可以为单轴、两轴、三轴或四轴云台等，多轴云台用来提高拍摄画面的稳定性；以云台11为三轴云台为例，云台11可以绕偏航轴、横滚轴和俯仰轴运动。与云台为手持云台且拍摄装置与云台可拆卸连接为例，本申请实施例提供的方法例如可以应用于云台+手机的场景。
其中，控制装置12可以用于根据多个摄像头中某一个摄像头对应的一路图像，控制云台11转动，以对待跟随目标进行跟随，具体的，控制装置12可以自动识别图像中的待跟随目标，并通过对云台11进行控制使得待跟随目标处于预设的构图位置，这一功能可以称为智能跟随功能。
示例性的，控制装置12可以设置在云台11上，控制装置12能够控制云台11进行智能跟随；或者，控制装置12可以设置在拍摄装置上，当启用云台11的智能跟随功能时，拍摄装置上搭载的控制装置12能够通过有线或无线方式与云台11通信，控制云台11转动以实现智能跟随。或者，控制装置12可以设置于搭载云台11的可移动平台上，当启用云台11的智能跟随功能时，可移动平台上搭载的控制装置12能够通过有线或无线方式与云台11通信，控制云台11转动以实现智能跟随。其中，可移动平台可以为能够搭载云台11的任意类型的设备，例如无人机、无人车等。
通常，对于具有多个摄像头的拍摄装置，在拍摄时，某一个摄像头拍摄的一路图像会经过一定的处理，例如基于拍摄装置的预设模式或者用户的特殊需求对该路图像进行处理，经过处理的图像会被用作当前显示画面。当云台搭载具有多个摄像头的拍摄装置时，在启动云台的智能跟随功能时，云台的控制装置会直接默认使用该路图像进行机器学习，通过云台的控制使得待跟随目标处于预设的构图位置，实现对于待跟随目标的智能跟随。然而，这样的方式在一些情况下会出现跟随效果较差的问题，例如该路图像对应的镜头FOV较小，在待跟随目标移动时容易丢失目标，或者，对图像的处理带来较大的时延导致跟随的灵敏度低。
为了解决由于控制装置12直接默认使用多个摄像头中某一个摄像头对应的一路图像控制云台11转动，导致在一些情况下会出现跟随效果较差的技术问题，在本申请实施例中，获取针对智能功能的开启操作，响应于所述开启操作，自动地从拍摄装置的多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，并根据目标摄像头对应的一路图像，控制云台转动，以对待跟随目标进行跟随，实现了从多个摄像头中确定一个摄像头，并使用确定的该摄像头对应的一路图像控制云台转动，使得能够灵活选择控制云台转动所使用的一路图像，从而能够减少由于直接默认使用某一个摄像头对应的一路图像控制云台转动时，容易导致跟随效果较差的问题。
本申请中拍摄装置的多个摄像头的光轴之间的距离可以较小，使得同一拍摄对象在多个摄像头拍摄的图像中的位置可以相差较小，以使得在使用不同摄像头对应的一路图像控制云台转动时，期望待跟随目标所处的预设的构图位置可以是同一位置，从而有利于简化实现。
下面结合附图，对本申请的一些实施方式作详细说明。在不冲突的情况下，下述的实施例及实施例中的特征可以相互组合。
图2为本申请一实施例提供的云台的控制方法的流程示意图，本实施例的执行主体可以为图1中的控制装置12，具体可以为控制装置12的处理器。如图2所示，本实施例的方法可以包括：
步骤20，获取针对智能功能的开启操作。
在本步骤中，可以获取用户针对智能功能的开启操作，示例性的，开启操作包括语音控制、按键控制、触屏等各种控制方式。
步骤21，响应于所述开启操作，自动地从拍摄装置的多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头。
本步骤中，确定目标摄像头的时机可以灵活实现。示例性的，可以获取针对智能跟随功能的开启操作，并响应于开启操作，从拍摄装置的多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，从而实现在用户打开智能跟随功能时确定智能跟随功能对应的目标摄像头。
可选的，可以不考虑使用默认摄像头对应的一路图像控制云台转动以对待跟随目标进行跟随是否存在跟随效果较差的问题，直接执行步骤21。或者可选的，在执行步骤21之前，可以先确定是否满足预设条件，其中，满足预设条件可以表示使用默认摄像头对应的一路图像控制云台转动容易存在跟随效果较差的问题，不满足预设条件可以表示使用默认摄像头对应的一路图像控制云台转动不存在跟随效果较差的问题。
其中，默认摄像头可以为多个摄像头中对应的一路图像用于当前显示画面的摄像头(以下可以记为第一摄像头，即第一摄像头拍摄的一路图像用于当前显示画面)，以下主要以默认摄像头为第一摄像头为例进行具体说明。
发明人发现：智能跟随功能的效果好坏主要与待跟随目标是否不容易跟丢以及跟随是否灵敏有关。
关于待跟随目标是否不容易跟丢：应当理解的是，任何的控制系统都不可能将误差完全控制为0，其误差带是一个范围值。对于云台的智能跟随系统，其控制误差例如为±10°；并且，由于控制误差通常是一定的，因此待跟随目标相对于云台运动同样大小的角度，FOV越小，待跟随目标在拍摄画面中运动的比例就会越大，待跟随目标就越容易丢失在画面中。例如，假设智能跟随系统的控制误差为±10°，如果在正常焦段(例如24mm)，其FOV可以为74°，通常不会出现跟丢目标，但是如果在人像焦段(例如50mm)或者长焦段(例如70mm)，其FOV变小，对于同样的控制误差，拍摄过程中便很容易丢失待跟随目标，跟随效果较差。
关于跟随是否灵敏：在智能跟随功能中，待跟随目标偏离预设的构图位置而产生偏差，该偏差经过机器学习后，云台能够响应于该偏差开始转动从而使得待跟随目标处于预设的构图位置，这一过程的时间越短代表跟随画面的时延越不明显，跟随的灵敏度越高。因此，如果用于跟随功能做机器学习的图像本身具有一定的时延，会增加跟随画面的时延，跟随 的灵敏度降低。
由于镜头的FOV可以与摄像头本身有关，跟随画面的时延可以与摄像头输出的图像经过的处理有关，且在第一摄像头为默认摄像头的情况下，通常是使用第一摄像头输出的待显示的一路图像控制云台转动，因此一个实施例中，满足预设条件可以包括第一摄像头满足预设条件(以下可以记为第一预设条件)，和/或当前显示画面满足预设条件(以下可以记为第二预设条件)。
第一摄像头满足第一预设条件可以表示使用第一摄像头对应的一路图像控制云台转动以对待跟随目标进行跟随，存在由于镜头的FOV小导致容易丢失目标的问题，第一摄像头不满足第一预设条件可以表示使用第一摄像头对应的一路图像控制云台转动以对待跟随目标进行跟随，不存在由于镜头的FOV小导致容易丢失目标的问题。
由于待跟随目标的移动通常是水平方向上的移动，在第一摄像头为横拍模式时，第一摄像头的水平FOV较大，然而在第一摄像头为竖拍模式时，第一摄像头的垂直FOV较小，因此在第一摄像头为竖拍模式时，如果使用第一摄像头对应的一路图像控制云台转动，会增加由于镜头的FOV小导致容易丢失目标的风险。
由于摄像头的视场角越小，待跟随目标就越容易丢失在画面中。因此，在多个摄像头中存在比第一摄像头的视场角更大的摄像头时，即第一摄像头在多个摄像头中并不具有最大的视场角时，即可以认为使用第一摄像头对应的一路图像控制云台转动，相比使用视场角更大的其他摄像头，会增加由于镜头的FOV小导致容易丢失目标的风险。
在一个实施例中，第一预设条件可以包括第一摄像头为竖拍模式，和/或，多个摄像头中存在比第一摄像头的视场角更大的摄像头。应理解，第一摄像头为竖拍模式且多个摄像头中存在比第一摄像头的视场角更大的摄像头，第一摄像头为竖拍模式或多个摄像头中存在比第一摄像头的视场角更大的摄像头，均可以理解为第一摄像头满足第一预设条件。
当前显示画面满足第二预设条件可以表示使用第一摄像头对应的一路图像控制云台转动，存在由于图像的时延大导致灵敏度差的问题，当前显示画面不满足第二预设条件可以表示使用第一摄像头对应的一路图像控制云台转动，不存在由于图像的时延大导致灵敏度差的问题。
在一个实施例中，第二预设条件可以包括当前显示画面的显示时间与当前显示画面的拍摄时间的差值大于预设阈值，即当前显示画面在拍摄后经过处理后用于显示，处理的时间较长，以至于利用处理后的当前显示画面做智能跟随时，较大的时延容易导致灵敏度差的问题。
通常，由于为了使得图像更加稳定所加入的电子防抖处理(Electric Image Stabilization，EIS)会加大图像时延，因此一个实施例中，第二预设条件可以包括当前显示画面经过了电子防抖处理。
需要说明的是，目标摄像头可以为第一摄像头，或者，目标摄像头可以为多个摄像头中除第一摄像头之外的一个摄像头。本申请中的目标摄像头指的是用于控制云台转动的一路图像所对应的摄像头。
步骤22，根据目标摄像头对应的一路图像，控制云台转动，以对待跟随目标进行跟随。
本步骤中，目标摄像头对应的一路图像可以为目标摄像头输出的一路图像，或者，可以为目标摄像头输出的一路图像经过处理后所得到的图像。一个实施例中，在目标摄像头为第一摄像头时，目标摄像头对应的一路图像可以为目标摄像头的待显示的一路图像；和 /或，在目标摄像头为多个摄像头除第一摄像头之外的一个摄像头时，目标摄像头对应的一路图像可以为目标摄像头输出的一路图像。
例如，可以采用机器学习的方式自动识别目标摄像头对应的一路图像中的待跟随目标，并通过对云台进行控制使得待跟随目标处于预设的构图位置。关于根据图像控制云台转动，以对待跟随目标进行跟随的实现方式，可以参考相关技术中的具体描述，在此不再赘述。
本实施例提供的云台的控制方法，通过从拍摄装置的多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，并根据目标摄像头对应的一路图像，控制云台转动，以对待跟随目标进行跟随，实现了从多个摄像头中确定一个摄像头，并使用确定的该摄像头对应的一路图像控制云台转动，使得能够灵活选择控制云台转动所使用的一路图像，从而能够减少由于直接默认使用某一个摄像头对应的一路图像控制云台转动时，容易导致跟随效果较差的问题。
图3为本申请另一实施例提供的云台的控制方法的流程示意图，本实施例在图2所示实施例的基础上主要描述了从拍摄装置的多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头的一种可选实现方式。如图3所示，本实施例的方法可以包括：
步骤31，在确定第一摄像头满足第一预设条件或当前显示画面满足第二预设条件时，从多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头。
本步骤中，第一摄像头满足第一预设条件和当前显示画面满足第二预设条件，均可以作为从多个摄像头中确定目标摄像头的触发条件。以下对第一摄像头满足第一预设条件作为触发条件和当前显示画面满足第二预设条件作为分别触发条件的两种情况分别进行具体说明。
一种情况，第一摄像头满足第一预设条件。
一个实施例中，在第一摄像头满足第一预设条件时可以采用优先考虑镜头的FOV对跟随效果影响的方式，从多个摄像头中确定目标摄像头。基于此，一个实施例中，步骤31具体可以包括：在确定第一摄像头满足第一预设条件时，将多个摄像头中视场角最大的摄像头确定为智能跟随功能对应的目标摄像头；在确定第一摄像头不满足第一预设条件，将第一摄像头确定为智能跟随功能对应的目标摄像头。从而实现在确定使用第一摄像头对应的一路图像控制云台转动存在目标容易跟丢的问题时，可以不考虑使用第一摄像头对应的一路图像控制云台转动是否存在灵敏度差的问题，根据多个摄像头中视场角最大的摄像头对应的一路图像控制云台转动。例如，多个摄像头包括至少一个广角摄像头和至少一个普通摄像头，普通摄像头的视场角均小于广角摄像头的视场角，当第一摄像头为普通摄像头时，选择广角摄像头作为目标摄像头，即利用普通摄像头对应的一路图像用来显示画面，利用广角摄像头对应的一路图像用来控制云台转动，优选地，当具有多个广角摄像头时，选择视场角最大的广角摄像头对应的一路图像用来控制云台转动。
由于实际应用中同一摄像头的垂直FOV是随其水平FOV变化的，两者之间可以是固定的倍数关系，因此在从多个摄像头中确定目标摄像头时，只需要对比摄像头的水平FOV或垂直FOV即可，示例性的，可以根据多个摄像头的水平FOV，将多个摄像头中水平FOV最大的摄像头确定为智能跟随功能对应的目标摄像头。
另一个实施例中，在第一摄像头满足第一预设条件时可以采用优先考虑图像的时延对跟随效果影响的方式，从多个摄像头中确定目标摄像头。基于此，一个实施例中，步骤31 具体可以包括：在确定一摄像头满足第一预设条件时，如果当前显示画面不满足第二预设条件，则将第一摄像头确定为智能跟随功能对应的目标摄像头，在确定第一摄像头满足第一预设条件时，如果当前显示画面满足第二预设条件，将多个摄像头中除第一摄像头外视场角最大的摄像头确定为智能跟随功能对应的目标摄像头。例如，多个摄像头包括至少一个广角摄像头和至少一个普通摄像头，普通摄像头的视场角均小于广角摄像头，当第一摄像头为某个普通摄像头时，如果该普通摄像头对应的一路图像用来作为当前显示画面时并没有被时延处理，为了追求跟随的灵敏度，可以直接选择该普通摄像头对应的一路图像用来控制云台转动；而如果该普通摄像头对应的一路图像用来作为当前显示画面时经过了时延处理如电子增稳，为了追求跟随的灵敏度，可以选择其他摄像头对应的一路图像用来控制云台转动，优选地，选择多个摄像头中除第一摄像头外视场角最大的摄像头对应的一路图像用来控制云台转动。
另一种情况，当前显示画面满足第二预设条件。
一个实施例中，在当前显示画面满足第二预设条件时可以采用优先考虑图像的时延对跟随效果影响的方式，从多个摄像头中确定目标摄像头。基于此，一个实施例中，步骤31具体可以包括：在确定当前显示画面满足第二预设条件时，将多个摄像头中除第一摄像头外视场角最大的摄像头确定为智能跟随功能对应的目标摄像头。例如，当前显示画面开启了电子增稳，表明当前显示画面具有明显的时延，该图像用于智能跟随，会导致跟随的灵敏度降低，因此，不选择当前摄像头即第一摄像头输出的图像做智能跟随，而选择除第一摄像头外的其他摄像头做跟随，优选地，选择其他摄像头中视场角最大的摄像头作为目标摄像头。即在确定使用第一摄像头对应的一路图像控制云台转动存在灵敏度差的问题时，可以暂不考虑使用第一摄像头对应的一路图像控制云台运动不存在容易跟丢目标的问题，而在选择多个摄像头中除第一摄像头外的其他摄像头时，考虑视场角最大的摄像头对应的一路图像控制云台转动。
另一个实施例中，在当前显示画面满足第二预设条件时可以采用优先考虑镜头的FOV对跟随效果影响的方式，从多个摄像头中确定目标摄像头。基于此，一个实施例中，步骤31具体可以包括：在确定当前显示画面满足第二预设条件时，如果第一摄像头不满足第一预设条件，将第一摄像头确定为智能跟随功能对应的目标摄像头；在确定当前显示画面满足第二预设条件时，如果第一摄像头满足第一预设条件，将多个摄像头中视场角最大的摄像头确定为智能跟随功能对应的目标摄像头。例如，当前显示画面开启了电子增稳，表明当前显示画面具有明显的时延，该图像用于智能跟随，会导致跟随的灵敏度降低，但是如果对跟随的灵敏度要求不高，而为了追求防止目标跟丢，如果当前摄像头即第一摄像头不处于竖拍模式，或者第一摄像头相比其他摄像头具有最大的视场角，仍选择第一摄像头作为目标摄像头，选择第一摄像头输出的图像做智能跟随；如果当前摄像头即第一摄像头处于竖拍模式，或者多个摄像头中还存在比第一摄像头的视场角更大的摄像头，那么选择多个摄像头中视场角最大的摄像头作为目标摄像头，在第一摄像头对应的一路图像做当前画面显示的情况下，选择目标摄像头对应的一路图像做智能跟随。
从而实现在确定使用第一摄像头对应的一路图像控制云台转动存在灵敏度差的问题时，还可以进一步考虑使用第一摄像头对应的一路图像控制云台运动是否存在目标容易跟丢的问题，如果不存在，则还是可以根据第一摄像头对应的一路图像控制云台转动，如果存在，可以根据多个摄像头中视场角最大的摄像头对应的一路图像控制云台转动。
可选的，在步骤31之前还可以包括：响应于开启操作，确定第一摄像头是否满足第一预设条件，和/或，确定当前显示画面是否满足第二预设条件。
步骤32，根据目标摄像头对应的一路图像，控制云台转动，以对待跟随目标进行跟随。
需要说明的是，步骤32与步骤22类似，在此不再赘述。
本实施例提供的云台的控制方法，通过在确定第一摄像头满足第一预设条件或当前显示画面满足第二预设条件时，自动地从拍摄装置的多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，并根据目标摄像头对应的一路图像，控制云台转动，以对待跟随目标进行跟随，实现了从多个摄像头中确定一个摄像头，并使用确定的该摄像头对应的一路图像控制云台转动，使得能够灵活选择控制云台转动所使用的一路图像对应的摄像头，从而能够减少出现由于直接默认使用某一个摄像头对应的一路图像控制云台转动，导致跟随效果较差的问题。
图4为本申请又一实施例提供的云台的控制方法的流程示意图，本实施例在图2所示实施例的基础上主要描述了从拍摄装置的多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头的另一种可选实现方式。如图4所示，本实施例的方法可以包括：
步骤41，如果第一摄像头不满足第一预设条件，将第一摄像头确定为智能跟随功能对应的目标摄像头；
步骤42，如果第一摄像头满足第一预设条件，将多个摄像头中视场角最大的摄像头确定为智能跟随功能对应的目标摄像头；
步骤43，根据目标摄像头对应的一路图像，控制云台转动，以对待跟随目标进行跟随。
其中，第一摄像头为多个摄像头中对应的一路图像用于当前显示画面的摄像头。示例性的，第一预设条件可以包括第一摄像头为竖拍模式，和/或，多个摄像头中存在比第一摄像头的视场角更大的摄像头。
示例性的，可以获取针对智能跟随功能的开启操作，并响应于开启操作确定第一摄像头是否满足第一预设条件。
第一摄像头不满足第一预设条件，可以表示使用第一摄像头对应的一路图像控制云台转动不存在由于镜头的FOV小导致容易丢失目标的问题，通过上述步骤41可以实现在使用第一摄像头对应的一路图像控制云台转动不存在容易丢失目标的问题的情况下，可以根据第一摄像头对应的一路图像控制云台转动。
第一摄像头满足第一预设条件，可以表示使用第一摄像头对应的一路图像控制云台转动存在由于镜头的FOV小导致容易丢失目标的问题，通过上述步骤42可以实现在使用第一摄像头对应的一路图像控制云台转动存在容易丢失目标的问题的情况下，可以根据多个摄像头中视场角最大的摄像头对应的一路图像控制云台转动，以尽量选择FOV大的摄像头对应的一路图像控制云台转动，从而减少出现丢失目标的情况。
以第一预设条件包括第一摄像头为竖拍模式，和/或，多个摄像头中存在比第一摄像头的视场角更大的摄像头为例，云台的控制方法例如可以如图5所示。
参考图5，云台开机后，在步骤51中可以获取多个摄像头中每个摄像头的FOV；在步骤52中可以获取针对智能跟随功能的开启操作；在步骤53中，响应于开启操作可以判断第一摄像头是否为竖拍；如果第一摄像头不为竖拍，则在步骤54中可以判断在多个摄像头中是否存在比第一摄像头的视场角更大的摄像头；如果第一摄像头为竖拍或者多个摄像头中存在比第一摄像头的视场角更大的摄像头，则在步骤55中可以将多个摄像头中视场 角最大的摄像头确定为目标摄像头，以根据视场角最大的摄像头对应的一路图像控制云台转动；如果多个摄像头中不存在比第一摄像头的视场角更大的摄像头，即第一摄像头在多个摄像头中具有最大的视场角，则在步骤56中可以将第一摄像头确定为目标摄像头，以根据第一摄像头对应的一路图像控制云台转动。
需要说明的是，图5所示的云台的控制方法的流程仅为举例。
应当理解的是，在一些实施例中，步骤53和步骤54的顺序可以改变，即在步骤53中，响应于开启操作可以判断在多个摄像头中是否存在比第一摄像头的视场角更大的摄像头，如果不存在比第一摄像头的视场角更大的摄像头，则在步骤54中可以判断第一摄像头是否为竖拍。
应当理解的是，在一些实施例中，步骤53和步骤54可能仅存在一个。例如，仅存在步骤53，在步骤53中，响应于开启操作可以判断第一摄像头是否为竖拍，如果第一摄像头为竖拍，则在步骤56中可以将多个摄像头中视场角最大的摄像头确定为目标摄像头，以根据视场角最大的摄像头对应的一路图像控制云台转动；如果第一摄像头不为竖拍，则将第一摄像头确定为目标摄像头。例如，仅存在步骤54，在步骤54中，响应于开启操作可以判断在多个摄像头中是否存在比第一摄像头的视场角更大的摄像头，如果多个摄像头中存在比第一摄像头的视场角更大的摄像头，则在步骤56中可以将多个摄像头中视场角最大的摄像头确定为目标摄像头，以根据视场角最大的摄像头对应的一路图像控制云台转动；如果多个摄像头中不存在比第一摄像头的视场角更大的摄像头，即第一摄像头在多个摄像头中具有最大的视场角，则在步骤56中可以将第一摄像头确定为目标摄像头，以根据第一摄像头对应的一路图像控制云台转动。
通过本实施例提供的云台的控制方法，实现了在使用第一摄像头对应的一路图像控制云台转动不存在容易丢失目标的问题的情况下，可以根据第一摄像头对应的一路图像控制云台转动，在使用第一摄像头对应的一路图像控制云台转动存在容易丢失目标的问题的情况下，可以根据多个摄像头中视场角最大的摄像头对应的一路图像控制云台转动，使得能够灵活选择控制云台转动所使用的一路图像对应的摄像头，从而能够减少出现由于直接默认使用某一个摄像头对应的一路图像控制云台转动，导致跟随效果较差的问题。
图6为本申请又一实施例提供的云台的控制方法的流程示意图，本实施例在图2所示实施例的基础上主要描述了从拍摄装置的多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头的又一种可选实现方式。如图6所示，本实施例的方法可以包括：
步骤61，如果当前显示画面不满足第二预设条件，将第一摄像头确定为智能跟随功能对应的目标摄像头；
步骤62，如果当前显示画面满足第二预设条件，则判断多个摄像头中是否存在比第一摄像头的视场角更大的摄像头，如果多个摄像头中不存在比第一摄像头的视场角更大的摄像头，则确定第一摄像头为目标摄像头，如果多个摄像头中存在比第一摄像头的视场角更大的摄像头，则将多个摄像头中视场角最大的摄像头确定为目标摄像头。
步骤63，根据目标摄像头对应的一路图像，控制云台转动，以对待跟随目标进行跟随。
其中，第一摄像头为多个摄像头中对应的一路图像用于当前显示画面的摄像头。示例性的，第二预设条件可以包括当前显示画面经过了电子防抖处理。
当前显示画面不满足第二预设条件，可以表示使用第一摄像头对应的一路图像控制云台转动不存在由于图像的延迟大导致灵敏度差的问题，通过上述步骤61可以实现在使用 第一摄像头对应的一路图像控制云台转动不存在灵敏度低的问题的情况下，可以根据第一摄像头对应的一路图像控制云台转动。
当前显示画面满足第二预设条件，可以表示使用第一摄像头对应的一路图像控制云台转动存在由于图像的时延大导致灵敏度差的问题，通过上述步骤62可以在优先考虑图像的时延导致灵敏度差的问题的情况下，兼顾考虑镜头的视场角对待跟随目标跟丢的影响。
以第二预设条件包括当前显示画面经过了电子防抖处理为例，云台的控制方法例如可以如图7所示。
参考图7，云台开机后，在步骤71中可以获取多个摄像头中每个摄像头的FOV；在步骤72中可以获取针对智能跟随功能的开启操作；在步骤73中，响应于开启操作可以判断是否开启EIS；如果开启，则在步骤74中判断多个摄像头中是否存在比第一摄像头的视场角更大的摄像头，即判断多个摄像头中除第一摄像头外的其他摄像头是否存在比第一摄像头的视场角更大的摄像头，如果存在，则将多个摄像头中视场角最大的摄像头确定为目标摄像头，以根据视场角最大的摄像头对应的一路图像控制云台转动，如果不存在，则将第一摄像头确定为目标摄像头；如果未开启EIS，则在步骤75中可以将第一摄像头确定为目标摄像头，以根据第一摄像头对应的一路图像控制云台转动。
应理解，图7中，如果开启EIS，则可以表示当前显示画面经过了电子防抖处理，如果未开启EIS，则可以表示当前显示画面未经过电子防抖处理。
需要说明的是，图7所示的云台的控制方法的流程仅为举例。
通过本实施例提供的云台的控制方法，实现了在使用第一摄像头对应的一路图像控制云台转动不存在灵敏度差的问题的情况下，可以根据第一摄像头对应的一路图像控制云台转动，在使用第一摄像头对应的一路图像控制云台转动存在灵敏度差的问题的情况下，可以根据多个摄像头中视场角最大的摄像头对应的一路图像控制云台转动，使得能够灵活选择控制云台转动所使用的一路图像对应的摄像头，从而能够减少出现由于直接使用某一个摄像头对应的一路图像控制云台转动，导致跟随效果较差的问题。
图8为本申请又一实施例提供的云台的控制方法的流程示意图，本实施例在图2所示实施例的基础上主要描述了从拍摄装置的多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头的又一种可选实现方式。如图8所示，本实施例的方法可以包括：
步骤81，如果第一摄像头不满足第一预设条件且当前显示画面不满足第二预设条件，将第一摄像头确定为智能跟随功能对应的目标摄像头；
步骤82，如果第一摄像头满足第一预设条件或当前显示画面满足第二预设条件，将多个摄像头中视场角最大的摄像头确定为智能跟随功能对应的目标摄像头；
步骤83，根据目标摄像头对应的一路图像，控制云台转动，以对待跟随目标进行跟随。
其中，第一摄像头为多个摄像头中对应的一路图像用于当前显示画面的摄像头。示例性的，第一预设条件可以包括第一摄像头为竖拍模式，和/或，第一摄像头在多个摄像头中并不具有最大的视场角。示例性的，第二预设条件可以包括当前显示画面经过了电子防抖处理。
第一摄像头不满足第一预设条件可以表示使用第一摄像头对应的一路图像控制云台转动不存在由于镜头的FOV小导致容易丢失目标的问题，当前显示画面不满足第二预设条件可以表示使用第一摄像头对应的一路图像控制云台转动也不存在由于图像的延迟大导致灵敏度差的问题，通过上述步骤81可以实现在使用第一摄像头对应的一路图像控制云 台转动不存在容易丢失目标以及灵敏度差的问题的情况下，可以根据第一摄像头对应的一路图像控制云台转动。
第一摄像头满足第一预设条件可以表示使用第一摄像头对应的一路图像控制云台转动存在由于镜头的FOV小导致容易丢失目标的问题，当前显示画面满足第二预设条件可以表示使用第一摄像头对应的一路图像控制云台转动存在由于图像的时延大导致灵敏度差的问题，通过上述步骤82可以实现在使用第一摄像头对应的一路图像控制云台转动存在容易丢失目标的问题或存在灵敏度差的问题的情况下，可以根据多个摄像头中视场角最大的摄像头对应的一路图像控制云台转动。
通过本实施例提供的云台的控制方法，实现了在使用第一摄像头对应的一路图像控制云台转动不存在容易丢失目标以及灵敏度差的问题的情况下，可以根据第一摄像头对应的一路图像控制云台转动，在使用第一摄像头对应的一路图像控制云台转动存在容易丢失目标或灵敏度差的问题的情况下，可以根据多个摄像头中视场角最大的摄像头对应的一路图像控制云台转动可以根据第一摄像头对应的一路图像控制云台转动，使得能够灵活确定控制装置对云台进行控制所使用的一路图像对应的摄像头，从而能够减少出现由于直接使用一个摄像头对应的一路图像对云台进行控制，导致跟随效果较差的问题。
图9为本申请一实施例提供的云台的控制装置的结构示意图，如图9所示，该装置90可以包括：处理器91和存储器92。
所述存储器92，用于存储程序代码；
所述处理器91，调用所述程序代码，当程序代码被执行时，用于执行以下操作：
获取针对智能跟随功能的开启操作；
响应于所述开启操作，自动地从拍摄装置的多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头；
根据目标摄像头对应的一路图像，控制云台转动，以对待跟随目标进行跟随。
本实施例提供的云台的控制装置，可以用于执行前述方法实施例的技术方案，其实现原理和技术效果与方法实施例类似，在此不再赘述。
图10为本申请一实施例提供的云台系统的结构示意图，如图10所示，该云台系统100可以包括：云台101和控制装置102，其中，云台101用于搭载拍摄装置；控制装置102用于控制云台101转动，以对待跟随目标进行跟随。关于控制装置102的具体实现方式可以参考图9所示实施例的具体描述，在此不再赘述。
图11为本申请一实施例提供的可移动平台的结构示意图，如图11所示，该可移动平台110包括可移动平台本体111、云台112和控制装置113，其中，云台112搭载于可移动平台本体111上，控制装置113设置于可移动平台上，控制装置113用于控制云台转动，以对待跟随目标进行跟随。关于控制装置102的具体实现方式可以参考图9所示实施例的具体描述，在此不再赘述。
需要说明的是，图11中是以控制装置113设置于可移动平台本体111上为例
本申请实施例还提供一种计算机可读存储介质，其上存储有计算机程序，当所述计算机程序被执行时，实现如上述方法实施例所述的方法。
本领域普通技术人员可以理解：实现上述各方法实施例的全部或部分步骤可以通过程序指令相关的硬件来完成。前述的程序可以存储于一计算机可读取存储介质中。该程序在执行时，执行包括上述各方法实施例的步骤；而前述的存储介质包括：ROM、RAM、磁碟或 者光盘等各种可以存储程序代码的介质。
最后应说明的是：以上各实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述各实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分或者全部技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的范围。

Claims (40)

  1. 一种云台的控制方法，所述云台用于搭载拍摄装置，其特征在于，包括：
    获取针对智能跟随功能的开启操作；
    响应于所述开启操作，自动地从所述拍摄装置的多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头；
    根据所述目标摄像头对应的一路图像，控制所述云台转动，以对待跟随目标进行跟随。
  2. 根据权利要求1所述的方法，其特征在于，所述多个摄像头中的第一摄像头对应的一路图像用于当前显示画面。
  3. 根据权利要求2所述的方法，其特征在于，所述目标摄像头为所述第一摄像头，或者，所述目标摄像头为所述多个摄像头中除所述第一摄像头之外的一个摄像头。
  4. 根据权利要求3所述的方法，其特征在于，在所述目标摄像头为所述第一摄像头时，所述目标摄像头对应的一路图像为所述目标摄像头的待显示的一路图像；
    和/或，在所述目标摄像头为所述多个摄像头除所述第一摄像头之外的一个摄像头时，所述目标摄像头对应的一路图像为所述目标摄像头输出的一路图像。
  5. 根据权利要求2所述的方法，其特征在于，所述自动地从所述拍摄装置的多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，包括：
    在确定所述第一摄像头满足第一预设条件或当前显示画面满足第二预设条件时，从所述多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头。
  6. 根据权利要求5所述的方法，其特征在于，所述第一预设条件包括：
    所述第一摄像头为竖拍模式，和/或，所述多个摄像头中存在比第一摄像头的视场角更大的摄像头。
  7. 根据权利要求5所述的方法，其特征在于，所述第二预设条件包括：
    所述当前显示画面的显示时间与所述当前显示画面的拍摄时间的差值大于预设阈值。
  8. 根据权利要求5所述的方法，其特征在于，所述第二预设条件包括：
    所述当前显示画面经过了电子防抖处理。
  9. 根据权利要求5所述的方法，其特征在于，所述在确定所述第一摄像头满足第一预设条件或当前显示画面满足第二预设条件时，从所述多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头之前，还包括：
    响应于所述开启操作，确定所述第一摄像头是否满足所述第一预设条件，和/或，确定当前显示画面是否满足所述第二预设条件。
  10. 根据权利要求5所述的方法，其特征在于，所述在确定所述第一摄像头满足第一预设条件时，从所述多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，包括：
    在确定所述第一摄像头满足第一预设条件时，将所述多个摄像头中视场角最大的摄像头确定为智能跟随功能对应的目标摄像头。
  11. 根据权利要求5所述的方法，其特征在于，所述在确定所述第一摄像头满足第一预设条件时，从所述多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，包括：
    在确定所述第一摄像头满足第一预设条件时，如果当前显示画面不满足第二预设条件，将所述第一摄像头确定为智能跟随功能对应的目标摄像头；
    在确定所述第一摄像头满足第一预设条件时，如果当前显示画面满足第二预设条件，将所述多个摄像头中除第一摄像头外视场角最大的摄像头确定为智能跟随功能对应的目标摄像头。
  12. 根据权利要求5所述的方法，其特征在于，所述在确定当前显示画面满足第二预设条件时，从所述多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，包括：
    在确定当前显示画面满足第二预设条件时，将所述多个摄像头中除第一摄像头外视场角最大的摄像头确定为智能跟随功能对应的目标摄像头。
  13. 根据权利要求5所述的方法，其特征在于，所述在确定当前显示画面满足第二预设条件时，从所述多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，包括：
    在确定当前显示画面满足第二预设条件时，如果所述第一摄像头不满足第一预设条件，将所述第一摄像头确定为智能跟随功能对应的目标摄像头；
    在确定当前显示画面满足第二预设条件时，如果所述第一摄像头满足第一预设条件，将所述多个摄像头中视场角最大的摄像头确定为智能跟随功能对应的目标摄像头。
  14. 根据权利要求1所述的方法，其特征在于，所述从所述多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，包括：
    如果第一摄像头不满足第一预设条件，将所述第一摄像头确定为智能跟随功能对应的目标摄像头；
    如果第一摄像头满足第一预设条件，将所述多个摄像头中视场角最大的摄像头确定为智能跟随功能对应的目标摄像头；
    其中，所述第一摄像头为所述多个摄像头中对应的一路图像用于当前显示画面的摄像头。
  15. 根据权利要求1所述的方法，其特征在于，所述从所述多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，包括：
    如果当前显示画面不满足第二预设条件，将第一摄像头确定为智能跟随功能对应的目标摄像头，所述第一摄像头为所述多个摄像头中对应的一路图像用于当前显示画面的摄像头；
    如果当前显示画面满足第二预设条件，多个摄像头中不存在比第一摄像头的视场角更大的摄像头时，则确定第一摄像头为目标摄像头；多个摄像头中存在比第一摄像头的视场角更大的摄像头时，则将多个摄像头中视场角最大的摄像头确定为目标摄像头。
  16. 根据权利要求1所述的方法，其特征在于，所述从所述多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，包括：
    如果第一摄像头不满足第一预设条件且当前显示画面不满足第二预设条件，将所述第一摄像头确定为智能跟随功能对应的目标摄像头；
    如果第一摄像头满足第一预设条件或当前显示画面满足第二预设条件，将所述多个摄像头中视场角最大的摄像头确定为智能跟随功能对应的目标摄像头；
    其中，所述第一摄像头为所述多个摄像头中对应的一路图像用于当前显示画面的摄像头。
  17. 根据权利要求1所述的方法，其特征在于，所述多个摄像头包括广角摄像头，普 通摄像头和长焦摄像头，其中，普通摄像头的视场角是介于广角摄像头和长焦摄像头之间。
  18. 根据权利要求1所述的方法，其特征在于，所述云台包括手持云台或机载云台。
  19. 一种云台的控制装置，所述云台用于搭载拍摄控制装置，其特征在于，所述控制装置包括：存储器和处理器；
    所述存储器，用于存储程序代码；
    所述处理器，调用所述程序代码，当程序代码被执行时，用于执行以下操作：
    获取针对智能跟随功能的开启操作；
    响应于所述开启操作，自动地从所述拍摄控制装置的多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头；
    根据所述目标摄像头对应的一路图像，控制所述云台转动，以对待跟随目标进行跟随。
  20. 根据权利要求19所述的控制装置，其特征在于，所述多个摄像头中的第一摄像头对应的一路图像用于当前显示画面。
  21. 根据权利要求20所述的控制装置，其特征在于，所述目标摄像头为所述第一摄像头，或者，所述目标摄像头为所述多个摄像头除所述第一摄像头之外的一个摄像头。
  22. 根据权利要求21所述的控制装置，其特征在于，在所述目标摄像头为所述第一摄像头时，所述目标摄像头对应的一路图像为所述目标摄像头的待显示的一路图像；
    和/或，在所述目标摄像头为所述多个摄像头除所述第一摄像头之外的一个摄像头时，所述目标摄像头对应的一路图像为所述目标摄像头输出的一路图像。
  23. 根据权利要求20所述的控制装置，其特征在于，所述处理器用于从所述拍摄控制装置的多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，包括：在确定所述第一摄像头满足第一预设条件或当前显示画面满足第二预设条件时，从所述多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头。
  24. 根据权利要求23所述的控制装置，其特征在于，所述第一预设条件包括：
    所述第一摄像头为竖拍模式，和/或，所述第一摄像头在所述多个摄像头中并不具有最大的视场角。
  25. 根据权利要求23所述的控制装置，其特征在于，所述第二预设条件包括：所述当前显示画面的显示时间与所述当前显示画面的拍摄时间的差值大于预设阈值。
  26. 根据权利要求23所述的控制装置，其特征在于，所述第二预设条件包括：所述当前显示画面经过了电子防抖处理。
  27. 根据权利要求23所述的控制装置，其特征在于，所述处理器还用于：
    响应于所述开启操作，确定所述第一摄像头是否满足所述第一预设条件，和/或，确定当前显示画面是否满足所述第二预设条件。
  28. 根据权利要求23所述的控制装置，其特征在于，所述处理器用于在确定所述第一摄像头满足第一预设条件时，从所述多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，包括：
    在确定所述第一摄像头满足第一预设条件时，将所述多个摄像头中视场角最大的摄像头确定为智能跟随功能对应的目标摄像头。
  29. 根据权利要求23所述的控制装置，其特征在于，所述处理器用于在确定所述第一摄像头满足第一预设条件时，从所述多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，包括：
    在确定所述第一摄像头满足第一预设条件时，如果当前显示画面不满足第二预设条件，将所述第一摄像头确定为智能跟随功能对应的目标摄像头；
    在确定所述第一摄像头满足第一预设条件时，如果当前显示画面满足第二预设条件，将所述多个摄像头中除第一摄像头外视场角最大的摄像头确定为智能跟随功能对应的目标摄像头。
  30. 根据权利要求23所述的控制装置，其特征在于，所述处理器用于在确定当前显示画面满足第二预设条件时，从所述多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，包括：
    在确定当前显示画面满足第二预设条件时，将所述多个摄像头中除第一摄像头外视场角最大的摄像头确定为智能跟随功能对应的目标摄像头。
  31. 根据权利要求23所述的控制装置，其特征在于，所述处理器用于在确定当前显示画面满足第二预设条件时，从所述多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，包括：
    在确定当前显示画面满足第二预设条件时，如果所述第一摄像头不满足第一预设条件，将所述第一摄像头确定为智能跟随功能对应的目标摄像头；
    在确定当前显示画面满足第二预设条件时，如果所述第一摄像头满足第一预设条件，将所述多个摄像头中视场角最大的摄像头确定为智能跟随功能对应的目标摄像头。
  32. 根据权利要求19所述的控制装置，其特征在于，所述处理器用于从所述多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，包括：
    如果第一摄像头不满足第一预设条件，将所述第一摄像头确定为智能跟随功能对应的目标摄像头；
    如果第一摄像头满足第一预设条件，将所述多个摄像头中视场角最大的摄像头确定为智能跟随功能对应的目标摄像头；
    其中，所述第一摄像头为所述多个摄像头中对应的一路图像用于当前显示画面的摄像头。
  33. 根据权利要求19所述的控制装置，其特征在于，所述处理器用于从所述多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，包括：
    如果当前显示画面不满足第二预设条件，将第一摄像头确定为智能跟随功能对应的目标摄像头，所述第一摄像头为所述多个摄像头中对应的一路图像用于当前显示画面的摄像头；
    如果当前显示画面满足第二预设条件，多个摄像头中不存在比第一摄像头的视场角更大的摄像头时，则确定第一摄像头为目标摄像头；多个摄像头中存在比第一摄像头的视场角更大的摄像头时，将所述多个摄像头中视场角最大的摄像头确定为智能跟随功能对应的目标摄像头。
  34. 根据权利要求19所述的控制装置，其特征在于，所述处理器用于从所述多个摄像头中确定一个摄像头作为智能跟随功能对应的目标摄像头，包括：
    如果第一摄像头不满足第一预设条件且当前显示画面不满足第二预设条件，将所述第一摄像头确定为智能跟随功能对应的目标摄像头；
    如果第一摄像头满足第一预设条件或当前显示画面满足第二预设条件，将所述多个摄像头中视场角最大的摄像头确定为智能跟随功能对应的目标摄像头；
    其中，所述第一摄像头为所述多个摄像头中对应的一路图像用于当前显示画面的摄像头。
  35. 根据权利要求19所述的控制装置，其特征在于，所述多个摄像头包括广角摄像头，普通摄像头和长焦摄像头，其中，普通摄像头的视场角是介于广角摄像头和长焦摄像头之间。
  36. 根据权利要求19所述的控制装置，其特征在于，所述云台包括手持云台或机载云台。
  37. 一种云台系统，其特征在于，包括：
    云台，所述云台用于搭载拍摄装置；以及，
    权利要求19-36中任一项所述的云台的控制装置，所述控制装置控制所述云台转动，以对待跟随目标进行跟随。
  38. 根据权利要求37所述的云台系统，其特征在于，所述控制装置设置于所述云台上，或，所述控制装置设置于搭载所述云台的可移动平台上。
  39. 一种可移动平台，其特征在于，包括：
    可移动平台本体；
    云台，搭载于所述可移动平台本体上；
    权利要求19-36中任一项所述的控制装置，设置于所述可移动平台上；
    其中，所述控制装置控制所述云台转动，以对待跟随目标进行跟随。
  40. 一种计算机可读存储介质，其特征在于，其上存储有计算机程序，当所述计算机程序被执行时，实现如权利要求1至18中任一项所述的方法。
PCT/CN2022/093081 2022-05-16 2022-05-16 云台的控制方法、装置、可移动平台和存储介质 WO2023220866A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2022/093081 WO2023220866A1 (zh) 2022-05-16 2022-05-16 云台的控制方法、装置、可移动平台和存储介质
CN202280050281.4A CN117730527A (zh) 2022-05-16 2022-05-16 云台的控制方法、装置、可移动平台和存储介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2022/093081 WO2023220866A1 (zh) 2022-05-16 2022-05-16 云台的控制方法、装置、可移动平台和存储介质

Publications (1)

Publication Number Publication Date
WO2023220866A1 true WO2023220866A1 (zh) 2023-11-23

Family

ID=88834356

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/093081 WO2023220866A1 (zh) 2022-05-16 2022-05-16 云台的控制方法、装置、可移动平台和存储介质

Country Status (2)

Country Link
CN (1) CN117730527A (zh)
WO (1) WO2023220866A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH09247515A (ja) * 1996-03-08 1997-09-19 Hitachi Constr Mach Co Ltd 遠隔操縦機械のカメラ視野角制御装置
CN103763519A (zh) * 2013-12-31 2014-04-30 南宁市公安局 一种基于多摄像头的自动监控方法
CN104125433A (zh) * 2014-07-30 2014-10-29 西安冉科信息技术有限公司 基于多球机联动结构的视频运动目标监控方法
CN111491093A (zh) * 2019-01-26 2020-08-04 初速度（苏州）科技有限公司 一种摄像头视场角的调整方法和装置
CN111866437A (zh) * 2020-06-30 2020-10-30 厦门亿联网络技术股份有限公司 一种用于视频会议双摄像头的自动切换方法及装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH09247515A (ja) * 1996-03-08 1997-09-19 Hitachi Constr Mach Co Ltd 遠隔操縦機械のカメラ視野角制御装置
CN103763519A (zh) * 2013-12-31 2014-04-30 南宁市公安局 一种基于多摄像头的自动监控方法
CN104125433A (zh) * 2014-07-30 2014-10-29 西安冉科信息技术有限公司 基于多球机联动结构的视频运动目标监控方法
CN111491093A (zh) * 2019-01-26 2020-08-04 初速度（苏州）科技有限公司 一种摄像头视场角的调整方法和装置
CN111866437A (zh) * 2020-06-30 2020-10-30 厦门亿联网络技术股份有限公司 一种用于视频会议双摄像头的自动切换方法及装置

Also Published As

Publication number Publication date
CN117730527A (zh) 2024-03-19

Similar Documents

Publication Publication Date Title
JP6557768B2 (ja) 追尾制御装置、追尾制御方法、追尾制御プログラム、及び、自動追尾撮影システム
US11696028B2 (en) Blur correction device, interchangeable lens and image-capturing device
WO2018205902A1 (zh) 防抖控制方法和装置
US8823814B2 (en) Imaging apparatus
US11265477B2 (en) Image capturing apparatus and non-transitory recording medium
WO2021047077A1 (zh) 基于多摄像模块的图像处理方法、装置、设备及介质
WO2019227384A1 (zh) 一种云台控制方法及云台
WO2020215215A1 (zh) 一种云台控制方法、设备、云台、系统及存储介质
WO2020140295A1 (zh) 手持云台控制方法和手持云台
WO2021212445A1 (zh) 拍摄方法、可移动平台、控制设备和存储介质
US20210018138A1 (en) Gimbal mode switching method, device, mobile platform and storage medium
CN115802158B (zh) 切换摄像头的方法与电子设备
WO2017177369A1 (zh) 一种稳定器的跟踪拍摄控制方法及系统
CN113805607A (zh) 一种无人机拍摄方法、装置、无人机及存储介质
CN114125268A (zh) 对焦方法及装置
TW202011218A (zh) 用於攝影的設備控制系統及方法
WO2022000138A1 (zh) 拍摄控制方法和装置、云台、拍摄系统
WO2022141271A1 (zh) 云台系统的控制方法、控制设备、云台系统和存储介质
WO2023220866A1 (zh) 云台的控制方法、装置、可移动平台和存储介质
WO2020026901A1 (ja) 情報処理装置及びその制御方法
US11245830B2 (en) Image capture apparatus and control method for same, and storage medium
WO2022082439A1 (zh) 轨迹延时拍摄方法、装置、云台相机、无人机及手持云台
US11265478B2 (en) Tracking apparatus and control method thereof, image capturing apparatus, and storage medium
WO2022000211A1 (zh) 拍摄系统的控制方法、设备、及可移动平台、存储介质
WO2021237565A1 (zh) 音频处理方法、电子设备及计算机可读存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22941917

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 202280050281.4

Country of ref document: CN