WO2018058264A1 - 一种基于影像的控制方法、装置及飞行器 - Google Patents

一种基于影像的控制方法、装置及飞行器 Download PDF

Info

Publication number
WO2018058264A1
WO2018058264A1 PCT/CN2016/100230 CN2016100230W WO2018058264A1 WO 2018058264 A1 WO2018058264 A1 WO 2018058264A1 CN 2016100230 W CN2016100230 W CN 2016100230W WO 2018058264 A1 WO2018058264 A1 WO 2018058264A1
Authority
WO
WIPO (PCT)
Prior art keywords
image information
posture
aircraft
feature
preset
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2016/100230
Other languages
English (en)
French (fr)
Inventor
赵丛
李思晋
封旭阳
张伟兴
邬奇峰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
SZ DJI Technology Co Ltd
Original Assignee
SZ DJI Technology Co Ltd
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 SZ DJI Technology Co Ltd filed Critical SZ DJI Technology Co Ltd
Priority to PCT/CN2016/100230 priority Critical patent/WO2018058264A1/zh
Priority to CN201680065366.4A priority patent/CN108351651B/zh
Publication of WO2018058264A1 publication Critical patent/WO2018058264A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/0094—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots involving pointing a payload, e.g. camera, weapon, sensor, towards a fixed or moving target
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/10—Simultaneous control of position or course in three dimensions
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer

Definitions

  • the present invention relates to the field of electronic technologies, and in particular, to an image-based control method, apparatus, and aircraft.
  • camera devices have been greatly popularized. Taking pictures and videos through camera devices such as cameras and cameras can be said to be one of people's daily lives.
  • the automation of camera devices has also become a research hotspot.
  • the camera device generally has a function of delay automatic shooting. After the camera is set up, the user needs to manually turn on the automatic shooting function, and then the camera starts counting down. While counting down, the user waits for shooting when he goes to the shooting position.
  • the existing camera device automation function processing process is cumbersome, time-consuming and laborious.
  • the embodiment of the invention provides an image-based control method, device and aircraft, which can complete the shooting process simply and quickly.
  • an embodiment of the present invention provides an image-based control method, including:
  • the second image information including the motion posture is collected again;
  • the imaging device is triggered to perform image capturing processing.
  • an embodiment of the present invention further provides an image-based control method, including:
  • the flight control image information is captured by an image pickup device mounted on the aircraft.
  • an embodiment of the present invention further provides an image-based control apparatus, including:
  • An acquisition module configured to collect first image information including a motion posture
  • a determining module configured to analyze a motion gesture in the collected first image information, and determine a first feature posture included in the first image information
  • a prompting module configured to send a first prompt message if the first feature posture satisfies a preset start condition
  • the acquiring module is further configured to: after the first prompt message is sent, collect second image information including a motion posture;
  • the determining module is further configured to analyze the motion gesture in the collected second image information, and determine a second posture gesture of the motion gesture included in the second image information;
  • the processing module is configured to trigger the imaging device to perform image capturing processing when the second feature posture satisfies a preset shooting condition.
  • an embodiment of the present invention further provides another image-based control apparatus, including:
  • Controlling an image capturing module configured to collect flight control image information including a motion posture, the flight control image information being captured by an image capturing device mounted on the aircraft;
  • a posture determining module configured to analyze a motion posture in the flight control image information, and determine a control feature posture included in the flight control image information
  • a flight control module configured to acquire a control instruction associated with the determined control feature gesture, and control aircraft flight according to the acquired control instruction.
  • an embodiment of the present invention further provides an aircraft, including: a memory, a processor, and a communication interface; and a memory, where the program stores program instructions;
  • the processor, the program instruction stored in the memory is used to collect first image information including a motion posture; and the motion posture in the collected first image information is analyzed to determine the first image a first feature gesture included in the message; if the first feature gesture satisfies a preset start condition, issuing a first prompt message; after issuing the first prompt message, acquiring second image information including the motion pose again;
  • the motion posture in the collected second image information is analyzed, Determining a second feature gesture of the motion gesture included in the second image information; if the second feature gesture satisfies a preset photographing condition, generating a trigger command to trigger the image capturing device to perform image capturing processing;
  • the communication interface is configured to send the trigger instruction to a camera.
  • an embodiment of the present invention further provides another aircraft, including: a memory processor and an electronic governor; a memory, wherein the memory stores program instructions;
  • the processor, the program instruction stored in the memory is used to collect flight control image information including a motion posture; and the motion posture in the flight control image information is analyzed to determine that the flight control image information is included a control feature gesture; acquiring a control command associated with the determined control feature gesture, and transmitting an instruction to the electronic governor according to the acquired control command, the flight control image information being mounted on the aircraft Photographed by the camera;
  • the electronic governor is configured to complete flight control of the aircraft according to an instruction sent by the processor.
  • the embodiment of the invention can automatically realize the control of the smart device such as the camera device or the aircraft that mounts the camera device based on the motion posture in the image captured by the camera device, and realize the camera shooting and the aircraft flight conveniently and quickly.
  • Control automation intelligent functions.
  • FIG. 1 is a schematic flow chart of an image-based control method according to an embodiment of the present invention.
  • FIG. 2 is a schematic flow chart of another image-based control method according to an embodiment of the present invention.
  • FIG. 3 is a schematic flow chart of still another image-based control method according to an embodiment of the present invention.
  • FIG. 4 is a schematic structural diagram of an image-based control apparatus according to an embodiment of the present invention.
  • FIG. 5 is a schematic structural diagram of another image-based control apparatus according to an embodiment of the present invention.
  • FIG. 6 is a schematic structural view of an aircraft according to an embodiment of the present invention.
  • Fig. 7 is a schematic structural view of another aircraft according to an embodiment of the present invention.
  • the camera device may be controlled based on the captured image.
  • the user may enter The processing stage of automatic shooting, and after confirmation in the processing stage, automatically captures image information about the user or other objects.
  • the main body that performs the corresponding processing may be an imaging device, or may be an aircraft such as an aircraft or a cloud platform on which the imaging device is mounted.
  • the first image information including the motion posture may be captured by the imaging device, and the motion posture may be a body posture or a hand posture of the user, such as a body jumping posture, a waving posture, or the like.
  • the motion posture can also be a posture of each body part of the user, such as a head posture, a foot posture, etc., and can also be a posture that the user issues through other objects, such as a gesture issued by an object such as a flag or a hat.
  • the user can preset a posture for triggering the camera to turn on the automatic trigger function, and start tracking the target to be photographed, for example, setting a posture of open arms.
  • the camera device captures an object within the lens range thereof by taking a video, and if there are various motion postures such as a user body posture or a hand posture, the first image information including the motion posture is extracted, the first The image information can be a small video captured. If the motion posture is not photographed, the shooting is continued until the acquired image including the motion posture is acquired.
  • the analysis of the motion posture mainly includes: extracting a motion posture in the first image information, and identifying the motion posture based on the depth neural network, and determining a motion feature (first feature posture) corresponding to the motion posture.
  • determining whether the first feature posture satisfies the preset start condition mainly includes: determining whether the identified first feature posture is a motion feature specified in a start condition, and if yes, the first feature posture satisfies a preset start condition .
  • the deep neural network is: by learning different postures, a corresponding model is obtained. Therefore, when the motion pose recognition is performed subsequently, the motion pose is learned based on the model of the deep neural network, and the corresponding motion feature can be analyzed and determined.
  • a first prompt message can be issued, the first prompt message being used to issue a prompt that has started tracking.
  • the prompt can be issued by controlling a functional component that can be instructed to be installed on the camera. If the camera device is mounted on another device, for example, on an aircraft or a pan/tilt head, the first prompt message can be sent through the component capable of giving an indication on the aircraft or the pan/tilt, and the component capable of issuing the indication Including indicator light (LED light), speaker, etc., The first prompt message is used to trigger the way the indicator light is illuminated or the speaker sounds.
  • LED light indicator light
  • the camera device After the first prompt message is sent, the camera device is controlled to collect the second image information.
  • the tracking function of the camera is activated, the user or other objects to be photographed are started to be tracked by the camera.
  • the user can check the shooting direction of the camera lens after receiving the prompt of the first prompt message, or view the scene captured by the camera device through the application APP, and determine whether the camera device has tracked the object that needs to be photographed. If it is determined that the camera device has tracked the object that needs to be photographed, for example, the user is tracked, the user can again issue a body posture or a posture of another portion, such as an in-situ jump, wave, and/or open arms.
  • the second image information may also be a small video including a motion posture in the captured video.
  • the shooting process can be: triggering when the second feature posture is detected to satisfy the preset shooting condition.
  • the camera starts counting down and shoots an image (photo or video) after the countdown ends.
  • the prompting of the prompt message and the shooting of the image can be realized by controlling the aircraft and the camera device.
  • the first prompt message is sent to control the LED indicator on the aircraft to emit light in a first rule, indicating that the tracking target has been started; and the second feature posture is satisfied
  • a shooting prompt message is issued to control the LED indicator on the aircraft to illuminate with the shooting prompt rule, indicating that the camera can take a picture at any time.
  • the first prompt message causes the aircraft's LED indicator to flash at a frequency of 1 Hz, indicating that the tracking function is activated and starts tracking the object to be photographed;
  • the shooting prompt message causes the aircraft's LED indicator to flash at a frequency of 20 Hz, indicating that the camera can be photographed.
  • 1HZ and 20HZ there is a big difference in the blinking of the LED indicator, so the user can distinguish it well.
  • the camera device can be mounted on the aircraft, and based on the image captured by the camera device, the aircraft can also be controlled to take off, land, or perform a specified task.
  • the camera device is controlled to collect control image information including a motion posture, where the motion posture includes the above
  • the user's body posture, hand posture, and posture of other parts of the body are mentioned, such as body jumping, waving, and the like.
  • the motion gesture can be identified based on a preset depth neural network.
  • a control instruction stored in the control feature gesture map is found according to the extracted control feature posture, and then the aircraft flight is controlled according to the acquired control instruction.
  • the storage control feature gestures and control instructions can be mapped in the form of a mapping table, as shown in Table 1 below.
  • Control feature pose Control instruction Jumping character pose take off Waved character pose landing Nodding feature pose Performing a photo task
  • FIG. 1 is a schematic flowchart diagram of an image-based control method according to an embodiment of the present invention.
  • the method of the embodiment of the present invention may be performed by a smart camera device. Of course, it may also be executed by an intelligent device such as an aircraft or a cloud platform that mounts the camera device to perform a corresponding shooting task.
  • the method of the embodiment of the present invention includes the following steps.
  • the first image information including the motion posture may be acquired by calling the camera when detecting that the camera enters the automatic shooting mode or receiving an automatic shooting instruction issued by the user via wireless or wired.
  • the exercise posture includes a user's body posture, or a hand posture, or a posture of another portion. For example, the user's feet are jumping, waving, and the like.
  • a control command may be issued to adjust a shooting position of the aircraft that mounts the camera device, and the control command is used to control the aircraft.
  • a control command that issues a continuous rotation causes the aircraft camera to align with the user, or a control command to fly the aircraft up, or to fly down, or to the left or the like, so that the camera on the camera is aimed at the user.
  • S102 Analyze a motion posture in the collected first image information, and determine a first feature posture included in the first image information. Recognizing a part of the user's hand and the like in the first image information by image recognition technology, and then based on a part of a plurality of video frames, based on a depth neural network pair obtained by learning a large number of motion postures
  • the motion gesture in the first image information performs gesture recognition, for example, gesture recognition, and determines whether it is a first feature gesture such as waving or jumping.
  • S103 Send a first prompt message if the first feature posture satisfies a preset start condition.
  • Determining whether the first feature posture satisfies the preset activation condition may be: determining whether the first feature posture is a feature specified in the activation condition, for example, the first feature posture is a feature of the wave hand specified in the activation condition, and determining The first feature pose satisfies a preset start condition.
  • the first prompt message may be a prompt instruction indicating that the camera, the aircraft, or the pan/tilt is illuminated or audible.
  • the indicator light configured on the triggering device emits a light-emitting prompt according to the preset illumination mechanism.
  • the illumination mechanism can be: controlling the blinking of the indicator light on the aircraft according to the preset frequency.
  • the second image information including the motion posture is collected again. After seeing or hearing the relevant indication based on the first prompt message, the user may again issue a motion gesture that is the same as or different from the motion gesture in the S101.
  • the second feature gesture may be a corresponding feature gesture such as waving, jumping, or the like.
  • the second feature gesture may also be identified by means of gesture recognition or the like by a deep neural network.
  • the second feature posture satisfies a preset shooting condition, triggering the imaging device to perform image capturing processing. After the second feature posture satisfies the preset shooting condition, the image capturing device may be triggered to perform image capturing processing.
  • the second characteristic posture satisfies the shooting condition, wherein the second characteristic posture is the same as the posture specified in the shooting condition, for example, the second characteristic posture is a jump, and the specified posture in the shooting condition is also a jump. Then the second feature pose satisfies a preset shooting condition.
  • the S106 may specifically include: if the second feature posture satisfies a preset shooting condition, further determining a shooting mode associated with the second feature posture; and triggering according to the determined shooting mode
  • the camera captures an image.
  • the determined shooting mode is determined from a preset mapping database in which a mapping relationship between the feature gesture and the shooting mode is stored.
  • the association mapping relationship between the feature pose and the photographing mode can be established in advance, as shown in Table 2.
  • the corresponding shooting mode can be found, and then the camera device, or the pan/tilt head and the aircraft are controlled according to the parameters specified in the shooting mode, and the related photos or videos are captured.
  • the parameters of the shooting mode configuration include: white balance, shutter, aperture and other camera parameters, and may also include shooting position angle parameters, shooting position angle parameters and the like for adjusting parameters of the aircraft or the pan/tilt head device.
  • the embodiment of the present invention can automatically control the smart device such as the camera device or the aircraft that mounts the camera device based on the motion posture in the image captured by the camera device, and intelligently prompt the user to shoot the current device through the interaction of the prompt information.
  • the state of the device makes it easy and quick to realize the automation and intelligent functions of the shooting control of the camera.
  • FIG. 2 it is a schematic flowchart of another image-based control method according to an embodiment of the present invention.
  • the method of the embodiment of the present invention may be performed by a smart camera device.
  • an intelligent device such as an aircraft or a cloud platform that performs the corresponding shooting task by mounting the camera device may also be used.
  • the method of the embodiment of the present invention includes the following steps.
  • the camera is mounted on an aircraft.
  • S201 Collect third image information including a motion posture.
  • the motion posture in the third image information may be the same as or different from the motion posture in the first image information and the motion posture in the second image information.
  • S202 Analyze a motion posture in the third image information, and determine a third feature posture included in the third image information.
  • the third feature posture satisfies a preset tracking confirmation condition
  • the first image information including the motion posture is further collected.
  • the second prompt message may be sent in the embodiment of the present invention, where the second prompt message is used to prompt that the automatic shooting has been initialized, and the tracking may be started.
  • the second prompt message sent is: triggering the indicator light configured on the aircraft to emit a light-emitting prompt according to the preset illumination mechanism.
  • the motion postures of S201 to S203 mainly serve to initialize the tracking function of the shooting.
  • S204 Analyze a motion gesture in the collected first image information, and determine a first feature posture included in the first image information.
  • the S205 Send a first prompt message if the first feature posture satisfies a preset start condition.
  • the S205 may specifically include: if the first feature posture satisfies a preset start condition, The indicator light configured on the triggering aircraft is illuminated according to a preset illumination mechanism.
  • the motion postures of S204 to S205 mainly serve to confirm that the triggering camera starts the tracking function, starts tracking or has tracked the user or other objects to be photographed.
  • S207 Analyze a motion posture in the collected second image information, and determine a second posture posture of the motion posture included in the second image information.
  • the S206 to S208 motion posture is mainly used to determine that the camera device performs photographing.
  • the tracking confirmation condition, the starting condition, and the shooting condition include requirements for the motion posture, and mainly include requirements for the third image information, the first image information, and the motion posture in the second image information, for example, in the tracking confirmation condition.
  • the motion posture requirement of the jump is included, and therefore, the third feature posture that satisfies the tracking confirmation condition that the motion posture included in the third image information is a jump;
  • the start condition includes a motion posture requirement of the wave, and therefore, the start condition is the first image.
  • the first characteristic posture of the information is waved;
  • the shooting condition includes a nodding motion posture requirement, and therefore, the second characteristic posture that satisfies the shooting condition is the second image information is nodding.
  • the shooting position of the aircraft can be adjusted so as to be able to capture various sports postures.
  • the method further includes: in the automatic shooting mode, issuing a control instruction to adjust a shooting position of the aircraft; wherein the control instruction is used to control the The aircraft rotates or moves until the camera module mounted by the aircraft captures one or more shooting targets capable of emitting a motion posture.
  • the method before the acquiring the first image information of the motion posture, the method further includes: in the automatic shooting mode, issuing a control instruction to adjust a shooting position of the aircraft; wherein the control instruction is used to control the aircraft Rotating or moving until the camera module mounted by the aircraft captures one or more shooting targets capable of emitting a motion posture.
  • the correspondingly used rules for indicating the illumination of the indicator light on the aircraft mainly include:
  • the corresponding second rule includes: the forearm light of the aircraft double flashes, and the forearm indicator of the aircraft first flashes twice at 25 Hz, then turns off the light for 500 ms, This is a cycle.
  • the prompt mode mainly indicates the tracking function of the aircraft initial shooting. Go to Tracking Tracking mode and start tracking the objects you want to shoot (such as the target person).
  • the first rule corresponding to the first prompt message includes: the indicator light of the forearm of the aircraft blinks at a frequency of 1 Hz, and is mainly used to indicate that the target person is already being tracked, The shooting mode is observed. At this time, the aircraft controls the camera to track the target person. After the target person is aligned, the aircraft does not move.
  • the shooting prompt rule corresponding to the upcoming shooting includes: a flashing long light mode, that is, the forearm light of the aircraft flashes at 2.10 s at 10 Hz, and then remains bright. 0.75s, after the light is turned off for 1.5s, the camera is triggered to capture the image of the target person.
  • the embodiment of the present invention can automatically control the smart device such as the camera device or the aircraft that mounts the camera device based on the motion posture in the image captured by the camera device, and intelligently prompt the user to shoot the current device through the interaction of the prompt information.
  • the state of the device makes it easy and quick to realize the automation and intelligent functions of the shooting control of the camera.
  • FIG. 3 it is a schematic flowchart of still another image-based control method according to an embodiment of the present invention.
  • the method of the embodiments of the present invention may be performed by an aircraft, and the control of the aircraft is implemented mainly based on the motion information and the image information of the shooting motion posture.
  • An imaging device is mounted on the aircraft, and the related imaging information including the motion posture is collected by the imaging device.
  • the method of the embodiment of the present invention includes the following steps.
  • the motion posture may be a posture of the user's body, hands, feet, and other parts.
  • the flight control impact information can be a small video including a motion pose.
  • S302 Analyze a motion posture in the flight control image information, and determine a control feature posture included in the flight control image information.
  • a model trained by a deep neural network can be used to learn different motion postures such as gestures. After learning a stable neural network, a specific motion gesture can be identified, for example, based on a deep neural network.
  • Gesture Recognition is a technique that uses image recognition technology to identify gestures.
  • S303 Acquire a control instruction associated with the determined control feature gesture, and control aircraft flight according to the acquired control instruction.
  • the mapping relationship between the control feature posture and the control command can be established, specifically, the mapping relationship of Table 1 above, and then the corresponding control command corresponding to the control posture is obtained, and the aircraft is controlled to take off, land, and perform special tasks.
  • the embodiment of the invention can automatically realize the control of the aircraft based on the motion posture in the image captured by the camera device, and realize the automatic and intelligent function of the aircraft control conveniently and quickly.
  • the device in the embodiment of the present invention may be disposed in an intelligent device such as an imaging device, an aircraft, a cloud platform, or the like.
  • the The device includes the following modules.
  • the acquiring module 401 is configured to collect first image information including a motion posture, and the determining module 402 is configured to analyze the motion posture in the collected first image information, and determine the first part included in the first image information.
  • a feature gesture the prompting module 403 is configured to: if the first feature gesture meets a preset start condition, issue a first prompt message; the collecting module 401 is further configured to collect the first prompt message again a second image information including a motion posture; the determining module 402 is further configured to analyze the motion posture in the collected second image information, and determine a second feature of the motion posture included in the second image information
  • the gesture module 404 is configured to trigger the image capturing device to perform image capturing processing when the second feature posture satisfies a preset shooting condition.
  • the prompting module 403 is specifically configured to: when the first feature posture satisfies a preset start condition, trigger an indicator light configured on the aircraft to emit a light-emitting prompt according to a preset lighting mechanism; wherein The aircraft is used to mount the camera device.
  • the device may further include: a notification module 405, wherein the collection module 401 is further configured to collect third image information including a motion posture; the determining module 402 is further configured to: The motion gesture in the three image information is analyzed to determine a third feature gesture included in the third image information; the notification module 405 is configured to notify when the third feature gesture satisfies a preset tracking confirmation condition
  • the acquisition module 401 collects first image information including a motion posture.
  • the determining module 402 is configured to perform motion analysis on the first image information, the second image information, or the third image information based on a preset depth neural network algorithm.
  • the prompting module 403 is further configured to: when the third feature gesture satisfies a preset tracking confirmation condition, issue a second prompt message, where the second prompt message is used to prompt that automatic shooting has been initialized.
  • the prompting module 403 is specifically configured to: when the third feature posture satisfies a preset tracking confirmation condition, trigger an indicator light configured on the aircraft to issue a light according to a preset lighting mechanism. a light prompt; wherein the aircraft is used to mount the camera.
  • the apparatus may further include: a position adjustment module 406, configured to issue a control instruction to adjust a shooting position of the aircraft in an automatic shooting mode; wherein the control instruction is used to control the rotation of the aircraft or Moving until the camera module mounted by the aircraft captures one or more shooting targets capable of emitting a motion posture.
  • a position adjustment module 406 configured to issue a control instruction to adjust a shooting position of the aircraft in an automatic shooting mode; wherein the control instruction is used to control the rotation of the aircraft or Moving until the camera module mounted by the aircraft captures one or more shooting targets capable of emitting a motion posture.
  • the position adjustment module 406 is configured to, when in the automatic shooting mode, issue a control command to adjust a shooting position of the aircraft; wherein the control instruction is used to control the aircraft to rotate or move until the aircraft
  • the mounted camera module captures one or more shooting targets capable of emitting a motion posture.
  • the processing module 404 is specifically configured to further determine a shooting mode associated with the second feature posture when the second feature posture satisfies a preset shooting condition; according to the determined shooting mode
  • the camera device is triggered to capture an image; the determined shooting mode is determined from a preset mapping database, and the mapping relationship between the feature posture and the shooting mode is stored in the mapping database.
  • each module of the device may be referred to the specific implementation of the related steps in the foregoing embodiments, and details are not described herein.
  • the embodiment of the present invention can automatically control the smart device such as the camera device or the aircraft that mounts the camera device based on the motion posture in the image captured by the camera device, and intelligently prompt the user to shoot the current device through the interaction of the prompt information.
  • the state of the device makes it easy and quick to realize the automation and intelligent functions of the shooting control of the camera.
  • FIG. 5 it is a schematic structural diagram of another image-based control device according to an embodiment of the present invention.
  • the device in the embodiment of the present invention may be disposed in an aircraft, such as an unmanned aerial vehicle.
  • the device may specifically include the following modules.
  • the image capturing module 501 is configured to collect flight control image information including a motion posture, the flight control image information is captured by an image capturing device mounted on the aircraft; and the posture determining module 502 is configured to Controlling the motion posture in the image information to determine the control feature posture included in the flight control image information; the flight control module 503 is configured to acquire a control instruction associated with the determined control feature gesture, and according to the acquired Control commands control the flight of the aircraft.
  • the posture determining module 502 is specifically configured to analyze the motion posture in the flight control image information based on a preset depth neural network algorithm.
  • the device may further include: a confirmation prompting module 504, configured to send a confirmation prompt message after collecting the flight control image information including the motion posture, and issue a confirmation prompt message to perform the collection flight control Image information.
  • a confirmation prompting module 504 configured to send a confirmation prompt message after collecting the flight control image information including the motion posture, and issue a confirmation prompt message to perform the collection flight control Image information.
  • the camera device mounted on the aircraft may be controlled based on the image.
  • the specific control process refer to the description in the foregoing embodiment.
  • the various modules of the device in the embodiments of the present invention reference may be made to the specific implementation of the related steps in the foregoing embodiments, and details are not described herein.
  • the embodiment of the invention can automatically realize the control of the aircraft based on the motion posture in the image captured by the camera device, and realize the automatic and intelligent function of the aircraft control conveniently and quickly.
  • FIG. 6 it is a schematic structural diagram of an aircraft according to an embodiment of the present invention.
  • the aircraft includes a motor, a propeller, a power supply, a rack, and an LED indicator, and further includes: a processor 601, a communication interface 602, and Memory 603.
  • the memory 603 may include a volatile memory such as a random-access memory (RAM); the memory 603 may also include a non-volatile memory such as a flash memory. (flash memory), a solid-state drive (SSD); the memory 603 may also include a combination of the above types of memories.
  • RAM random-access memory
  • non-volatile memory such as a flash memory. (flash memory), a solid-state drive (SSD); the memory 603 may also include a combination of the above types of memories.
  • the processor 601 can be a central processing unit (CPU).
  • the processor 601 may further include a hardware chip.
  • the hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof.
  • the PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a general array logic (GAL), or any combination thereof.
  • the memory 603 is further configured to store program instructions.
  • the processor 601 can invoke the program instructions to implement an image-based control method as shown in the embodiments of Figures 1 and 2 of the present application.
  • the processor 601 is configured to invoke a program instruction stored in the memory 603, to collect first image information including a motion posture, and analyze the motion posture in the collected first image information to determine Determining a first feature gesture included in the first image information; if the first feature gesture satisfies a preset start condition, issuing a first prompt message; after issuing the first prompt message, collecting the first motion gesture Two image information; motion in the collected second image information Performing an analysis to determine a second posture gesture of the motion gesture included in the second image information; if the second feature gesture satisfies a preset shooting condition, generating a triggering instruction to trigger the imaging device to perform image capturing processing;
  • the communication interface 602 is configured to send the trigger instruction to a camera.
  • the processor 601 is configured to: when the first feature gesture meets a preset start condition, issue a first prompt message, specifically, if the first feature pose satisfies a preset The starting condition triggers the indicator light configured on the aircraft to emit a light-emitting prompt according to a preset lighting mechanism; wherein the aircraft is used to mount the camera device.
  • the processor 601 is further configured to collect third image information including a motion posture, analyze the motion posture in the third image information, and determine a third content included in the third image information. a feature gesture; if the third feature gesture satisfies a preset tracking confirmation condition, performing the acquiring the first image information including the motion gesture.
  • the processor 601 is configured to perform analysis based on a preset depth neural network algorithm when analyzing motion poses in the first image information, the second image information, or the third image information. .
  • the processor 601 is further configured to: if the third feature gesture satisfies a preset tracking confirmation condition, issue a second prompt message, where the second prompt message is used to prompt that automatic shooting has been initialized.
  • the processor 601 is specifically configured to: if the third feature gesture satisfies a preset tracking confirmation condition, when the second prompt message is sent, specifically, if the third feature posture satisfies the preset
  • the tracking confirmation condition triggers the indicator light disposed on the aircraft to emit a light-emitting prompt according to a preset lighting mechanism; wherein the aircraft is used to mount the camera device.
  • the processor 601 is further configured to: when the automatic shooting mode is issued, issue a control instruction to adjust a shooting position of the aircraft; wherein the control instruction is used to control the aircraft to rotate or move until the The camera module mounted on the aircraft captures one or more shooting targets capable of emitting a motion posture.
  • the processor 601 is further configured to: when the automatic shooting mode is issued, issue a control instruction to adjust a shooting position of the aircraft; wherein the control instruction is used to control the aircraft to rotate or move until the The camera module mounted on the aircraft captures one or more shooting targets capable of emitting a motion posture.
  • the processor 601 is configured to: when the second feature pose satisfies a preset shooting condition, trigger the camera to capture an image, specifically, if the second feature pose satisfies Setting a shooting condition to further determine a shooting mode associated with the second feature posture; triggering the camera device to capture an image according to the determined shooting mode; the determined shooting mode is from a preset mapping database It is determined that the mapping relationship between the feature gesture and the shooting mode is stored in the mapping database.
  • the embodiment of the present invention can automatically control the smart device such as the camera device or the aircraft that mounts the camera device based on the motion posture in the image captured by the camera device, and intelligently prompt the user to shoot the current device through the interaction of the prompt information.
  • the state of the device makes it easy and quick to realize the automation and intelligent functions of the shooting control of the camera.
  • FIG. 7 it is a schematic structural diagram of another aircraft according to an embodiment of the present invention.
  • the aircraft includes a motor, a propeller, a power supply, a rack, and an LED indicator.
  • the processor further includes: a processor 701 and an electronic strip.
  • the electronic strip is connected to the processor 701 and the motor of the aircraft, and the electronic strip is controlled by the controller 702 to adjust the rotation of the aircraft motor, thereby realizing the control of various flight postures and actions of the aircraft.
  • the memory 703 may include a volatile memory 703 (for example, a RAM).
  • the memory 703 may also include a non-volatile memory 703 (such as a flash memory, SSD).
  • the memory 703 may also include the above categories.
  • the processor 701 can be a CPU.
  • the processor 701103 may further include a hardware chip.
  • the above hardware chip may be an ASIC, a PLD, or a combination thereof.
  • the above PLD may be a complex programmable logic device CPLD, FPGA, GAL or any combination thereof.
  • the memory 703 is further configured to store program instructions.
  • the processor 701 can invoke the program instructions to implement an image-based control method as shown in the embodiment of FIG. 3 of the present application.
  • the processor 701 calls a program instruction stored in the memory 703, and is used to collect flight control image information including a motion posture; and analyzes a motion posture in the flight control image information to determine the flight. Controlling a control feature gesture included in the image information; acquiring a control instruction associated with the determined control feature gesture, and displaying the electronic strip according to the acquired control instruction
  • the controller 702 sends an instruction, the flight control image information is captured by an image capturing device mounted on the aircraft; the electronic strip is 702, configured to complete according to the instruction sent by the processor 701 Flight control of the aircraft.
  • the processor 701 is further configured to: after collecting the image information including the motion posture, issue an acknowledgement prompt message, and issue the confirmation prompt message to execute the collected flight control image information.
  • the processor 701 is specifically configured to perform motion on the flight control image information based on a preset depth neural network algorithm when analyzing the motion posture in the flight control image information. Posture analysis.
  • the specific implementation of the processor in the aircraft may be referred to the specific implementation of the related steps in the foregoing embodiments, and details are not described herein. Further, the processor in the embodiment of the present invention may further complete the control processing of the camera device mounted on the aircraft based on the image.
  • the specific method for controlling the camera device based on the image refers to the description in the foregoing embodiment.
  • the embodiment of the invention can automatically realize the control of the aircraft based on the motion posture in the image captured by the camera device, and realize the automatic and intelligent function of the aircraft control conveniently and quickly.
  • the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Remote Sensing (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • General Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Studio Devices (AREA)
  • Image Analysis (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

一种基于影像的控制方法、装置及飞行器，其中，方法包括：采集包括运动姿势的第一影像信息；对采集到的第一影像信息中的运动姿势进行分析，确定出该第一影像信息中包含的第一特征姿势；若第一特征姿势满足预置的启动条件，则发出第一提示消息；在发出第一提示消息后，再次采集包括运动姿势的第二影像信息；对采集到的第二影像信息中的运动姿势进行分析，确定出该第二影像信息中包含的运动姿势第二特征姿势；若第二特征姿势满足预置的拍摄条件，则触发摄像装置进行影像拍摄处理。采用该方案，可简便、快捷地实现了拍摄装置拍摄、飞行器飞行控制的自动化、智能化功能。

Description

一种基于影像的控制方法、装置及飞行器 技术领域
本发明涉及电子技术领域，尤其涉及一种基于影像的控制方法、装置及飞行器。
背景技术
随着电子技术的发展和人们生活水平的提高，摄像装置已经得到了极大的普及。通过照相机、摄影机等摄像装置拍摄图片、视频已经可以说是人们的日常生活之一。
摄像装置的自动化也成为研究的热点。目前，摄像装置一般具有延时自动拍摄的功能。在架好摄像装置后，需要用户手动开启自动拍摄功能，然后摄像装置开始倒计时，在倒计时的同时，用户在跑到拍摄的位置等待拍摄。现有的摄像装置自动化功能处理流程较繁琐，费时费力。
发明内容
本发明实施例提供了一种基于影像的控制方法、装置及飞行器，可以简便、快捷地完成拍摄处理。
一方面，本发明实施例提供了一种基于影像的控制方法，包括：
采集包括运动姿势的第一影像信息；
对所述采集到的第一影像信息中的运动姿势进行分析，确定出该第一影像信息中包含的第一特征姿势；
若所述第一特征姿势满足预置的启动条件，则发出第一提示消息；
在发出第一提示消息后，再次采集包括运动姿势的第二影像信息；
对所述采集到的第二影像信息中的运动姿势进行分析，确定出该第二影像信息中包含的运动姿势第二特征姿势；
若所述第二特征姿势满足预置的拍摄条件，则触发所述摄像装置进行影像拍摄处理。
另一方面，本发明实施例还提供了一种基于影像的控制方法，包括：
采集包括运动姿势的飞行控制影像信息；
对所述飞行控制影像信息中的运动姿势进行分析，确定出该飞行控制影像 信息中包含的控制特征姿势；
获取与所述确定出的控制特征姿势关联的控制指令，并根据获取的控制指令控制飞行器飞行；
所述飞行控制影像信息是挂载在所述飞行器上的摄像装置拍摄得到的。
相应地，本发明实施例还提供了一种基于影像的控制装置，包括：
采集模块，用于采集包括运动姿势的第一影像信息；
确定模块，用于对所述采集到的第一影像信息中的运动姿势进行分析，确定出该第一影像信息中包含的第一特征姿势；
提示模块，用于若所述第一特征姿势满足预置的启动条件，则发出第一提示消息；
所述采集模块，还用于在发出第一提示消息后，再次采集包括运动姿势的第二影像信息；
所述确定模块，还用于对所述采集到的第二影像信息中的运动姿势进行分析，确定出该第二影像信息中包含的运动姿势第二特征姿势；
处理模块，用于在所述第二特征姿势满足预置的拍摄条件时，则触发所述摄像装置进行影像拍摄处理。
相应地，本发明实施例还提供了另一种基于影像的控制装置，包括：
控制影像采集模块，用于采集包括运动姿势的飞行控制影像信息，所述飞行控制影像信息是挂载在所述飞行器上的摄像装置拍摄得到的；
姿势确定模块，用于对所述飞行控制影像信息中的运动姿势进行分析，确定出该飞行控制影像信息中包含的控制特征姿势；
飞行控制模块，用于获取与所述确定出的控制特征姿势关联的控制指令，并根据获取的控制指令控制飞行器飞行。
相应地，本发明实施例还提供了一种飞行器，包括：存储器，处理器和通信接口；存储器，所述存储器中存储有程序指令；
所述处理器，调用所述存储器中存储的程序指令，用于采集包括运动姿势的第一影像信息；对所述采集到的第一影像信息中的运动姿势进行分析，确定出该第一影像信息中包含的第一特征姿势；若所述第一特征姿势满足预置的启动条件，则发出第一提示消息；在发出第一提示消息后，再次采集包括运动姿势的第二影像信息；对所述采集到的第二影像信息中的运动姿势进行分析，确 定出该第二影像信息中包含的运动姿势第二特征姿势；若所述第二特征姿势满足预置的拍摄条件，则生成触发指令触发所述摄像装置进行影像拍摄处理；
所述通信接口，用于向拍摄装置发送所述触发指令。
相应地，本发明实施例还提供了另一种飞行器，包括：存储器处理器和电子调速器；存储器，所述存储器中存储有程序指令；
所述处理器，调用所述存储器中存储的程序指令，用于采集包括运动姿势的飞行控制影像信息；对所述飞行控制影像信息中的运动姿势进行分析，确定出该飞行控制影像信息中包含的控制特征姿势；获取与所述确定出的控制特征姿势关联的控制指令，并根据获取的控制指令向所述电子调速器发送指令，所述飞行控制影像信息是挂载在所述飞行器上的摄像装置拍摄得到的；
所述电子调速器，用于根据所述处理器发送的指令完成对所述飞行器的飞行控制。
本发明实施例可以基于摄像装置拍摄得到的影像中的运动姿势，来自动地实现对摄像装置或者是挂载摄像装置的飞行器等智能设备的控制，简便、快捷地实现了拍摄装置拍摄、飞行器飞行控制的自动化、智能化功能。
附图说明
图1是本发明实施例的一种基于影像的控制方法的流程示意图；
图2是本发明实施例的另一种基于影像的控制方法的流程示意图；
图3是本发明实施例的再一种基于影像的控制方法的流程示意图；
图4是本发明实施例的一种基于影像的控制装置的结构示意图；
图5是本发明实施例的另一种基于影像的控制装置的结构示意图；
图6是本发明实施例的一种飞行器的结构示意图；
图7是本发明实施例的另一种飞行器的结构示意图。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，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，都属于本发明保护的范围。
在本发明的一种实施例中，可以基于拍摄到的影像来对摄像装置进行控制，在检测到摄像装置处于自动拍摄模式时，例如检测到用户向摄像装置发出了自动拍摄的指令，可以进入自动拍摄的处理阶段，并在处理阶段确认后，自动拍摄关于用户的或者其他目标物的影像信息。执行相应处理的主体可以是摄像装置，也可以是挂载摄像装置的飞行器、云台等设备。
首先，可以通过摄像装置拍摄包括运动姿势的第一影像信息，所述运动姿势可以为用户的身体姿势或者手部姿势，例如身体跳跃的姿势、挥手的姿势等。当然也可以为用户的各个身体部位的姿势，例如头部姿势、脚部姿势等，还可以为用户通过其他物体发出的姿势，例如通过小旗、或者帽子等物体发出的姿势。用户可以预先设置用于触发摄像装置开启自动触发功能，开始追踪待拍摄目标的姿势，例如设置张开双臂的姿势。摄像装置会以拍摄视频的方式拍摄其镜头范围内的物体，如果拍摄到存在诸如用户身体姿势或手部姿势等各种运动姿势时，则提取包括运动姿势的第一影像信息，所述第一影像信息可以为截取的一段小视频。而如果没有拍摄到运动姿势，则继续拍摄，直至获取到的包括运动姿势的影像。
在获取到所述第一影像信息后，即基于视频帧对其中的运动姿势进行分析，确定该第一影像信息中的运动姿势是否满足追踪的启动条件，如果满足，则发出第一提示消息。若不满足，则重新拍摄视频，或发出运动姿势错误的提醒指示。对运动姿势的分析主要包括：提取出所述第一影像信息中的运动姿势，并基于深度神经网络对该运动姿势进行识别，确定该运动姿势所对应的运动特征(第一特征姿势)。而所述第一特征姿势是否满足预置的启动条件则主要包括：判断识别的第一特征姿势是否为启动条件中指定的运动特征，若是，则所述第一特征姿势满足预置的启动条件。所述深度神经网络为：通过对不同的运动姿势进行学习，得到相应模型。因此，后续进行运动姿势识别时，基于该深度神经网络的模型对运动姿势进行学习，可以分析确定出对应的运动特征。
在满足条件后，可以发出第一提示消息，该第一提示消息用于发出已经开始进行追踪的提示。可以通过控制所述摄像装置上安装的能够发出指示的功能部件来发出提示。如果所述摄像装置挂载在其他设备上，例如挂载在飞行器或者云台上时，则可以通过配置在飞行器或者云台上的能够发出指示的部件发出第一提示消息，能够发出指示的部件包括指示灯(LED灯)、扬声器等，所述 第一提示消息用于触发指示灯发光或者扬声器发声的方式发出提醒。
在发出了第一提示消息后，会控制摄像装置采集第二影像信息。在启动了摄像装置的追踪功能后，开始通过摄像装置追踪用户或者其他待拍摄的物体。用户可以在得到第一提示消息的提示后，查看摄像装置镜头的拍摄方向，或者通过应用APP查看摄像装置拍摄到的景象，确定摄像装置是否追踪到了需要拍摄的对象。如果确定摄像装置追踪到了需要拍摄的对象，例如追踪到了用户自己，则用户可以再次发出身体姿势或者其他部位的姿势，例如原地起跳、挥手、和/或张开双臂等运动姿势。所述第二影像信息同样可以为拍摄到的视频中一段包括运动姿势的小视频。基于上述的对第一影像信息的相关处理所使用的相同方式，判断从第二影像信息中得到的第二特征姿势是否满足拍摄条件，如果满足拍摄条件，即表明摄像装置当前能够拍摄到的用户想要的对象，可以触发摄像装置进入拍摄状态，执行相应的拍摄处理，随时摄取镜头内的对象，例如，拍摄处理可以为：在检测到第二特征姿势满足预置的拍摄条件时，则触发摄像装置开始倒计时，并在倒计时结束后，拍摄影像(照片或者视频)。
当所述摄像装置挂载在飞行器上时，可以通过控制飞行器和摄像装置来实现提示消息的提示和影像的拍摄。具体的，在所述第一特征姿势满足预置的启动条件时，发出第一提示消息控制飞行器上的LED指示灯以第一规则发光，表示已经开始追踪对象；在所述第二特征姿势满足预置的拍摄条件时，发出拍摄提示消息控制飞行器上的LED指示灯以拍摄提示规则发光，表示摄像装置随时可以拍照。例如，第一提示消息使飞行器的LED指示灯以1HZ的频率闪烁，表明追踪功能启动，开始追踪要拍摄的对象；拍摄提示消息使飞行器的LED指示灯以20HZ的频率闪烁，则表明可以拍照了。1HZ和20HZ时，LED指示灯的闪烁情况存在很大的区别，所以用户可以很好地分辨。
在本发明的其他实施例中，摄像装置可以挂载在飞行器上，基于摄像装置拍摄到的影像，还可以控制飞行器起飞、降落或者执行指定的任务。具体的，在飞行器进入影像控制模式时，例如用户通过与飞行器配对的遥控器发出进入影像控制模式的控制指令时，控制所述摄像装置采集包括运动姿势的控制影像信息，所述运动姿势包括上述提到的用户身体姿势、手部姿势以及身体其他部位的姿势等，例如身体跳跃、挥手等。
对控制影像信息中的运动姿势进行分析，可使用手势识别，人体姿势识别 等技术，提取该运动姿势所对应的控制特征姿势。可以基于预置的深度神经网络来对运动姿势进行识别。根据提取的控制特征姿势找到与该控制特征姿势映射存储的控制指令，进而根据获取的控制指令控制飞行器飞行。可以通过一个映射表的形式映射存储控制特征姿势和控制指令，该映射表如下表1所示。
表1
控制特征姿势 控制指令
跳跃的特征姿势 起飞
挥手的特征姿势 降落
点头的特征姿势 执行拍照任务
进一步地，请参见图1，是本发明实施例的一种基于影像的控制方法的流程示意图。本发明实施例的所述方法可以由智能摄像装置来执行，当然，也可以由挂载该摄像装置执行相应拍摄任务的飞行器、云台等智能设备来执行。具体的，本发明实施例的所述方法包括如下步骤。
S101：采集包括运动姿势的第一影像信息。可以在检测到摄像装置进入自动拍摄模式时，或者接收到用户通过无线或者有线方式发出的关于自动拍摄指令时，调用摄像装置采集包括运动姿势的第一影像信息。运动姿势包括用户的身体姿势、或者手部姿势、或者其他部位的姿势。例如，用户双脚起跳、挥手等姿势。
可选地，为了保证摄像装置能够拍摄到关于用户的运动姿势的第一影像信息，可以发出控制指令来调整挂载所述摄像装置的飞行器的拍摄位置，所述控制指令用于控制所述飞行器转动或移动，直至所述飞行器挂载的摄像模块拍摄得到一个或者多个能够发出运动姿势的拍摄目标。例如，发出持续旋转的控制指令使飞行器摄像头对准用户，或者使飞行器往上飞、或往下飞、或往左等方向飞行的控制指令，使摄像装置上的摄像头对准用户。
S102：对所述采集到的第一影像信息中的运动姿势进行分析，确定出该第一影像信息中包含的第一特征姿势。通过图像识别技术识别出所述第一影像信息中用户手部等部位，然后根据多个视频帧中手部等部位，基于预先设置的通过对大量的运动姿势进行学习后得到的深度神经网络对所述第一影像信息中的运动姿势进行姿势识别，例如进行手势识别，确定是否为挥手、跳跃等第一特征姿势。
S103：若所述第一特征姿势满足预置的启动条件，则发出第一提示消息。判断第一特征姿势是否满足预置的启动条件可以是：判断第一特征姿势是否为启动条件中指定的特征，例如，第一特征姿势为所述启动条件中指定的挥手的特征，则确定所述第一特征姿势满足预置的启动条件。
第一提示消息可以是指示摄像装置、飞行器、或云台等设备发光、发声等提示指令。具体的，当摄像装置挂载在飞行器上时，若所述第一特征姿势满足预置的启动条件，则触发飞行器上配置的指示灯按照预置的发光机制发出发光提示。发光机制可以为：按照预置的频率控制飞行器上的指示灯闪烁。
S104：在发出第一提示消息后，再次采集包括运动姿势的第二影像信息。用户在看到或者听到相关的基于第一提示消息的指示后，可以再次发出与所述S101中的运动姿势相同或者不相同的运动姿势。
S105：对所述采集到的第二影像信息中的运动姿势进行分析，确定出该第二影像信息中包含的运动姿势第二特征姿势。第二特征姿势可以为对应的挥手、跳跃等特征姿势。所述第二特征姿势也可以是通过深度神经网络进行手势识别等方式识别确定的。
S106：若所述第二特征姿势满足预置的拍摄条件，则触发所述摄像装置进行影像拍摄处理。在第二特征姿势满足预置的拍摄条件后，即可触发所述摄像装置进行影像拍摄处理。所述第二特征姿势满足拍摄条件是指，所述第二特征姿势与所述拍摄条件中指定的姿势相同，例如，第二特征姿势为跳跃，所述拍摄条件中指定的姿势也为跳跃，则所述第二特征姿势满足预置的拍摄条件。
具体的，所述S106具体可以包括：若所述第二特征姿势满足预置的拍摄条件，则进一步确定所述第二特征姿势所关联的拍摄模式；根据所述确定的拍摄模式，触发所述摄像装置拍摄影像。所述确定的拍摄模式是从预置的映射数据库中确定的，所述映射数据库中存储了特征姿势和拍摄模式的映射关系。
可以预先建立特征姿势和拍摄模式的关联映射关系，具体如表2所示。
表2：
特征姿势 拍摄模式
挥手 延时10秒拍摄
点头 连拍10张
跳跃 运动模式拍摄
在确定了第二特征姿势满足预置的拍摄条件后，可以找到对应的拍摄模式，进而根据拍摄模式中规定的参数对摄像装置、或者云台、飞行器进行控制，拍摄获取相关照片或者视频。拍摄模式配置的参数包括：白平衡、快门、光圈等相机参数，还可以包括拍摄位置角度参数，拍摄位置角度参数等用于调节飞行器或云台设备的参数。
本发明实施例可以基于摄像装置拍摄得到的影像中的运动姿势，来自动地实现对摄像装置或者是挂载摄像装置的飞行器等智能设备的控制，通过提示信息的交互，智能地提示用户当前拍摄装置的状态，简便、快捷地实现了拍摄装置拍摄控制的自动化、智能化功能。
再请参见图2，是本发明实施例的另一种基于影像的控制方法的流程示意图。本发明实施例的所述方法可以由智能摄像装置来执行，当然，也可以由挂载该摄像装置执行相应拍摄任务的飞行器、云台等智能设备。具体的，本发明实施例的所述方法包括如下步骤。在本发明实施例中，所述拍摄装置被挂载在飞行器上。
S201：采集包括运动姿势的第三影像信息。所述第三影像信息中的运动姿势，下述的第一影像信息中的运动姿势、第二影像信息中的运动姿势可以相同也可以不相同。
S202：对所述第三影像信息中的运动姿势进行分析，确定出该第三影像信息中包含的第三特征姿势。
S203：若所述第三特征姿势满足预置的追踪确认条件，则进一步采集包括运动姿势的第一影像信息。进一步地，若所述第三特征姿势满足预置的追踪确认条件，本发明实施例中还可以发出第二提示消息，所述第二提示消息用于提示自动拍摄已初始化，可以开始追踪要拍摄的对象。其中具体的，若所述第三特征姿势满足预置的追踪确认条件，所发出的第二提示消息为：触发飞行器上配置的指示灯按照预置的发光机制发出发光提示。
所述S201至S203的运动姿势主要作用在于初始化拍摄的追踪功能。
S204：对所述采集到的第一影像信息中的运动姿势进行分析，确定出该第一影像信息中包含的第一特征姿势。
S205：若所述第一特征姿势满足预置的启动条件，则发出第一提示消息。具体的，所述S205具体可以包括：若所述第一特征姿势满足预置的启动条件， 则触发飞行器上配置的指示灯按照预置的发光机制发出发光提示。
所述S204至S205的运动姿势主要作用在于确认触发摄像装置启动追踪功能，开始追踪或者已经追踪到用户或者其他待拍摄的对象。
S206：在发出第一提示消息后，再次采集包括运动姿势的第二影像信息。
S207：对所述采集到的第二影像信息中的运动姿势进行分析，确定出该第二影像信息中包含的运动姿势第二特征姿势。
S208：若所述第二特征姿势满足预置的拍摄条件，则触发所述摄像装置进行影像拍摄处理。
所述S206至S208运动姿势主要用于确定摄像装置进行拍照。
所述的追踪确认条件、启动条件以及拍摄条件包括了对运动姿势的要求，主要包括对第三影像信息、第一影像信息以及第二影像信息中的运动姿势的要求，例如，追踪确认条件中包括跳跃的运动姿势需求，因此，满足追踪确认条件是第三影像信息中包括的运动姿势的第三特征姿势为跳跃；启动条件中包括挥手的运动姿势需求，因此，满足启动条件是第一影像信息的第一特征姿势为挥手；拍摄条件中包括点头的运动姿势需求，因此，满足拍摄条件是第二影像信息的第二特征姿势为点头。
在本发明实施例中，可选地，可以对飞行器的拍摄位置进行调整，以便于能够拍摄到各种运动姿势。具体可选地，可以在所述采集包括运动姿势的第三影像信息之前，还包括：在自动拍摄模式时，发出控制指令调整所述飞行器的拍摄位置；其中，所述控制指令用于控制所述飞行器转动或移动，直至所述飞行器挂载的摄像模块拍摄得到一个或者多个能够发出运动姿势的拍摄目标。或者可选地，所述采集包括运动姿势的第一影像信息之前，还包括：在自动拍摄模式时，发出控制指令调整所述飞行器的拍摄位置；其中，所述控制指令用于控制所述飞行器转动或移动，直至所述飞行器挂载的摄像模块拍摄得到一个或者多个能够发出运动姿势的拍摄目标。
具体的，在本发明实施例的一种基于飞行器的提示消息的提示方式中，所对应使用的指示飞行器上的指示灯发光的规则主要包括：
所述第三特征姿势满足预置的追踪确认条件时，所对应的第二规则包括：飞行器的前臂灯双闪，且飞行器的前臂指示灯先以25Hz快闪两次,然后灭灯500ms，以此为一个周期。该提示方式主要表明飞行器初始化拍摄的追踪功能， 进入追踪tracking模式，开始追踪需要拍摄的目标物(如目标人物)。
在所述第一特征姿势满足预置的启动条件时，关于第一提示消息所对应的第一规则包括：飞行器前臂的指示灯以1HZ的频率闪烁，主要用于表示已经在追踪目标人物，可以进行拍摄观测watch模式，此时飞行器会控制用于摄像装置追踪对准目标人物，在对准目标人物后，飞行器不会移动。
在所述第二特征姿势满足预置的拍摄条件时，关于即将拍摄所对应的拍摄提示规则包括：快闪长亮模式，即所述飞行器的前臂灯以10Hz快闪2.25s,然后保持长亮0.75s,此后灭灯1.5s后，触发拍照获取目标人物的影像。
本发明实施例可以基于摄像装置拍摄得到的影像中的运动姿势，来自动地实现对摄像装置或者是挂载摄像装置的飞行器等智能设备的控制，通过提示信息的交互，智能地提示用户当前拍摄装置的状态，简便、快捷地实现了拍摄装置拍摄控制的自动化、智能化功能。
再请参见图3，是本发明实施例的再一种基于影像的控制方法的流程示意图。本发明实施例的所述方法可以由飞行器来执行，主要基于运动姿势和拍摄运动姿势的影像信息来实现对该飞行器的控制。该飞行器上挂载了摄像装置，由该摄像装置来采集包括运动姿势的相关影像信息。具体的，本发明实施例的所述方法包括如下步骤。
S301：采集包括运动姿势的飞行控制影像信息。所述的运动姿势可以为用户身体、手部、脚部以及其他部位的姿势。该飞行控制影响信息可以为一段包括运动姿势的小视频。
S302：对所述飞行控制影像信息中的运动姿势进行分析，确定出该飞行控制影像信息中包含的控制特征姿势。通过图像识别技术，得到视频帧中相关部位，然后确定相关部位发出的运动姿势。本发明实施例可以用深度神经网络训练出的模型对不同的运动姿势例如手势进行学习，在学习好稳定的神经网络后，即可对特定的运动姿势进行识别，例如基于深度神经网络完成用户的手势识别。
S303：获取与所述确定出的控制特征姿势关联的控制指令，并根据获取的控制指令控制飞行器飞行。同样可以建立控制特征姿势与控制指令的映射关系，具体如上述表1的映射关系，进而得到相应的控制姿势对应的控制指令，控制飞行器起飞、降落以及执行特殊任务。
本发明实施例可以基于摄像装置拍摄得到的影像中的运动姿势，来自动地实现飞行器的控制，简便、快捷地实现了飞行器控制的自动化、智能化功能。
下面再对本发明实施例的基于影像的控制装置及飞行器进行详细描述。
请参见图4，是本发明实施例的一种基于影像的控制装置的结构示意图，本发明实施例的所述装置可以设置在摄像装置、飞行器、云台等智能设备中，具体的，所述装置包括如下模块。
采集模块401，用于采集包括运动姿势的第一影像信息；确定模块402，用于对所述采集到的第一影像信息中的运动姿势进行分析，确定出该第一影像信息中包含的第一特征姿势；提示模块403，用于若所述第一特征姿势满足预置的启动条件，则发出第一提示消息；所述采集模块401，还用于在发出第一提示消息后，再次采集包括运动姿势的第二影像信息；所述确定模块402，还用于对所述采集到的第二影像信息中的运动姿势进行分析，确定出该第二影像信息中包含的运动姿势第二特征姿势；处理模块404，用于在所述第二特征姿势满足预置的拍摄条件时，则触发所述摄像装置进行影像拍摄处理。
进一步可选地，所述提示模块403，具体用于在所述第一特征姿势满足预置的启动条件时，触发飞行器上配置的指示灯按照预置的发光机制发出发光提示；其中，所述飞行器用于挂载所述摄像装置。
进一步可选地，所述装置还可以包括：通知模块405，其中，所述采集模块401，还用于采集包括运动姿势的第三影像信息；所述确定模块402，还用于对所述第三影像信息中的运动姿势进行分析，确定出该第三影像信息中包含的第三特征姿势；所述通知模块405，用于在所述第三特征姿势满足预置的追踪确认条件时，通知所述采集模块401采集包括运动姿势的第一影像信息。
其中具体的，所述确定模块402在用于对对所述第一影像信息、第二影像信息或者第三影像信息中的运动姿势进行分析时是基于预置的深度神经网络算法进行的。
进一步可选地，所述提示模块403，还用于在所述第三特征姿势满足预置的追踪确认条件时，发出第二提示消息，所述第二提示消息用于提示自动拍摄已初始化。
进一步可选地，所述提示模块403，具体用于在所述第三特征姿势满足预置的追踪确认条件时，触发飞行器上配置的指示灯按照预置的发光机制发出发 光提示；其中，所述飞行器用于挂载所述摄像装置。
进一步可选地，所述装置还可以包括：位置调整模块406，用于在自动拍摄模式时，发出控制指令调整所述飞行器的拍摄位置；其中，所述控制指令用于控制所述飞行器转动或移动，直至所述飞行器挂载的摄像模块拍摄得到一个或者多个能够发出运动姿势的拍摄目标。
和/或，所述位置调整模块406，用于在自动拍摄模式时，发出控制指令调整所述飞行器的拍摄位置；其中，所述控制指令用于控制所述飞行器转动或移动，直至所述飞行器挂载的摄像模块拍摄得到一个或者多个能够发出运动姿势的拍摄目标。
进一步可选地，所述处理模块404，具体用于在所述第二特征姿势满足预置的拍摄条件时，进一步确定所述第二特征姿势所关联的拍摄模式；根据所述确定的拍摄模式，触发所述摄像装置拍摄影像；所述确定的拍摄模式是从预置的映射数据库中确定的，所述映射数据库中存储了特征姿势和拍摄模式的映射关系。
在本发明实施例中，所述装置的各个模块的具体实现可参考上述各实施例中相关步骤的具体实现，在此不赘述。
本发明实施例可以基于摄像装置拍摄得到的影像中的运动姿势，来自动地实现对摄像装置或者是挂载摄像装置的飞行器等智能设备的控制，通过提示信息的交互，智能地提示用户当前拍摄装置的状态，简便、快捷地实现了拍摄装置拍摄控制的自动化、智能化功能。
再请参见图5，是本发明实施例的另一种基于影像的控制装置的结构示意图，本发明实施例的所述装置可以设置在飞行器中，例如无人飞行器中。所述装置具体可以包括如下模块。
控制影像采集模块501，用于采集包括运动姿势的飞行控制影像信息，所述飞行控制影像信息是挂载在所述飞行器上的摄像装置拍摄得到的；姿势确定模块502，用于对所述飞行控制影像信息中的运动姿势进行分析，确定出该飞行控制影像信息中包含的控制特征姿势；飞行控制模块503，用于获取与所述确定出的控制特征姿势关联的控制指令，并根据获取的控制指令控制飞行器飞行。其中可选地，所述姿势确定模块502，具体用于基于预置的深度神经网络算法对所述飞行控制影像信息中的运动姿势进行分析。
进一步可选地，所述装置还可以包括：确认提示模块504，用于在采集到包括运动姿势的飞行控制影像信息后，发出确认提示消息，并在发出确认提示消息，执行所述采集飞行控制影像信息。
在本发明实施例中，除了对飞行器基于影像进行控制外，还可以基于影像对该飞行器上挂载的摄像装置进行控制，具体控制过程可参考上述实施例中的描述。并且，进一步地，本发明实施例的所述装置的各个模块的具体实现可参考上述各实施例中相关步骤的具体实现，在此不赘述。
本发明实施例可以基于摄像装置拍摄得到的影像中的运动姿势，来自动地实现飞行器的控制，简便、快捷地实现了飞行器控制的自动化、智能化功能。
再请参见图6，是本发明实施例的一种飞行器的结构示意图，所述飞行器包括电机、螺旋桨、供电电源、机架以及LED指示灯等结构，还包括：处理器601、通信接口602以及存储器603。
所述存储器603可以包括易失性存储器(volatile memory)，例如随机存取存储器(random-access memory，RAM)；存储器603也可以包括非易失性存储器(non-volatile memory)，例如快闪存储器(flash memory)，固态硬盘(solid-state drive，SSD)；存储器603还可以包括上述种类的存储器的组合。
所述处理器601可以是中央处理器(central processing unit，CPU)。所述处理器601还可以进一步包括硬件芯片。上述硬件芯片可以是专用集成电路(application-specific integrated circuit，ASIC)，可编程逻辑器件(programmable logic device，PLD)或其组合。上述PLD可以是复杂可编程逻辑器件(complex programmable logic device，CPLD)，现场可编程逻辑门阵列(field-programmable gate array，FPGA)，通用阵列逻辑(generic array logic，GAL)或其任意组合。
可选地，所述存储器603还用于存储程序指令。所述处理器601可以调用所述程序指令，实现如本申请图1和2实施例中所示的基于影像的控制方法。
具体的，所述处理器601，调用所述存储器603中存储的程序指令，用于采集包括运动姿势的第一影像信息；对所述采集到的第一影像信息中的运动姿势进行分析，确定出该第一影像信息中包含的第一特征姿势；若所述第一特征姿势满足预置的启动条件，则发出第一提示消息；在发出第一提示消息后，再次采集包括运动姿势的第二影像信息；对所述采集到的第二影像信息中的运动 姿势进行分析，确定出该第二影像信息中包含的运动姿势第二特征姿势；若所述第二特征姿势满足预置的拍摄条件，则生成触发指令触发所述摄像装置进行影像拍摄处理；
所述通信接口602，用于向拍摄装置发送所述触发指令。
进一步可选地，所述处理器601，在用于若所述第一特征姿势满足预置的启动条件，则发出第一提示消息时，具体用于若所述第一特征姿势满足预置的启动条件，则触发飞行器上配置的指示灯按照预置的发光机制发出发光提示；其中，所述飞行器用于挂载所述摄像装置。
进一步可选地，所述处理器601，还用于采集包括运动姿势的第三影像信息；对所述第三影像信息中的运动姿势进行分析，确定出该第三影像信息中包含的第三特征姿势；若所述第三特征姿势满足预置的追踪确认条件，则执行所述采集包括运动姿势的第一影像信息。
进一步可选地，所述处理器601，在用于对所述第一影像信息、第二影像信息或者第三影像信息中的运动姿势进行分析时是基于预置的深度神经网络算法进行的分析。
进一步可选地，所述处理器601，还用于若所述第三特征姿势满足预置的追踪确认条件，则发出第二提示消息，所述第二提示消息用于提示自动拍摄已初始化。
进一步可选地，所述处理器601，在用于若所述第三特征姿势满足预置的追踪确认条件，则发出第二提示消息时，具体用于若所述第三特征姿势满足预置的追踪确认条件，则触发飞行器上配置的指示灯按照预置的发光机制发出发光提示；其中，所述飞行器用于挂载所述摄像装置。
进一步可选地，所述处理器601，还用于在自动拍摄模式时，发出控制指令调整所述飞行器的拍摄位置；其中，所述控制指令用于控制所述飞行器转动或移动，直至所述飞行器挂载的摄像模块拍摄得到一个或者多个能够发出运动姿势的拍摄目标。
进一步可选地，所述处理器601，还用于在自动拍摄模式时，发出控制指令调整所述飞行器的拍摄位置；其中，所述控制指令用于控制所述飞行器转动或移动，直至所述飞行器挂载的摄像模块拍摄得到一个或者多个能够发出运动姿势的拍摄目标。
进一步可选地，所述处理器601，在用于若所述第二特征姿势满足预置的拍摄条件，则触发所述摄像装置拍摄影像时，具体用于若所述第二特征姿势满足预置的拍摄条件，则进一步确定所述第二特征姿势所关联的拍摄模式；根据所述确定的拍摄模式，触发所述摄像装置拍摄影像；所述确定的拍摄模式是从预置的映射数据库中确定的，所述映射数据库中存储了特征姿势和拍摄模式的映射关系。
在本发明实施例中，所述飞行器中处理器的具体实现可参考上述各实施例中相关步骤的具体实现，在此不赘述。
本发明实施例可以基于摄像装置拍摄得到的影像中的运动姿势，来自动地实现对摄像装置或者是挂载摄像装置的飞行器等智能设备的控制，通过提示信息的交互，智能地提示用户当前拍摄装置的状态，简便、快捷地实现了拍摄装置拍摄控制的自动化、智能化功能。
再请参见图7，是本发明实施例的另一种飞行器的结构示意图，所述飞行器包括电机、螺旋桨、供电电源、机架以及LED指示灯等结构，还包括：处理器701、电子条是器702以及存储器703。
所述电子条是器702与所述处理器701和飞行器的电机相连，通过控制电子条是器702以达到调整飞机电机转动，进而实现对飞行器的各种飞行姿势和动作的控制。
所述存储器703可以包括易失性存储器703(volatile memory)，例如RAM；存储器703也可以包括非易失性存储器703(non-volatile memory)，例如flash memory，SSD；存储器703还可以包括上述种类的存储器703的组合。
所述处理器701可以是CPU。所述处理器701103还可以进一步包括硬件芯片。上述硬件芯片可以是ASIC，PLD或其组合。上述PLD可以是复杂可编程逻辑器件CPLD，FPGA，GAL或其任意组合。
可选地，所述存储器703还用于存储程序指令。所述处理器701可以调用所述程序指令，实现如本申请图3实施例中所示的基于影像的控制方法。
具体的，所述处理器701，调用所述存储器703中存储的程序指令，用于采集包括运动姿势的飞行控制影像信息；对所述飞行控制影像信息中的运动姿势进行分析，确定出该飞行控制影像信息中包含的控制特征姿势；获取与所述确定出的控制特征姿势关联的控制指令，并根据获取的控制指令向所述电子条 是器702发送指令，所述飞行控制影像信息是挂载在所述飞行器上的摄像装置拍摄得到的；所述电子条是器702，用于根据所述处理器701发送的指令完成对所述飞行器的飞行控制。
进一步可选地，所述处理器701，还用于在采集到包括运动姿势的影像信息后，发出确认提示消息，并在发出确认提示消息，执行所述采集飞行控制影像信息。
进一步可选地，所述处理器701，在用于对所述飞行控制影像信息中的运动姿势进行分析时，具体用于基于预置的深度神经网络算法对所述飞行控制影像信息中的运动姿势进行分析。
在本发明实施例中，所述飞行器中的处理器的具体实现可参考上述各实施例中相关步骤的具体实现，在此不赘述。并且进一步地，本发明实施例中的所述处理器还可以基于影像完成对飞行器上挂载的摄像装置的控制处理，具体的基于影像对摄像装置进行控制的方式参考上述实施例中的描述。
本发明实施例可以基于摄像装置拍摄得到的影像中的运动姿势，来自动地实现飞行器的控制，简便、快捷地实现了飞行器控制的自动化、智能化功能。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，所述的程序可存储于一计算机可读取存储介质中，该程序在执行时，可包括如上述各方法的实施例的流程。其中，所述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory，ROM)或随机存储记忆体(Random Access Memory，RAM)等。
以上所揭露的仅为本发明较佳实施例而已，当然不能以此来限定本发明之权利范围，因此依本发明权利要求所作的等同变化，仍属本发明所涵盖的范围。

Claims (36)

  1. 一种基于影像的控制方法，其特征在于，包括：
    采集包括运动姿势的第一影像信息；
    对所述采集到的第一影像信息中的运动姿势进行分析，确定出该第一影像信息中包含的第一特征姿势；
    若所述第一特征姿势满足预置的启动条件，则发出第一提示消息；
    在发出第一提示消息后，再次采集包括运动姿势的第二影像信息；
    对所述采集到的第二影像信息中的运动姿势进行分析，确定出该第二影像信息中包含的第二特征姿势；
    若所述第二特征姿势满足预置的拍摄条件，则触发所述摄像装置进行影像拍摄处理。
  2. 如权利要求1所述的方法，其特征在于，所述若所述第一特征姿势满足预置的启动条件，则发出第一提示消息，包括：
    若所述第一特征姿势满足预置的启动条件，则触发飞行器上配置的指示灯按照预置的发光机制发出发光提示；其中，所述飞行器用于挂载所述摄像装置。
  3. 如权利要求1或2所述的方法，其特征在于，所述采集包括运动姿势的第一影像信息之前，还包括：
    采集包括运动姿势的第三影像信息；
    对所述第三影像信息中的运动姿势进行分析，确定出该第三影像信息中包含的第三特征姿势；
    若所述第三特征姿势满足预置的追踪确认条件，则执行所述采集包括运动姿势的第一影像信息。
  4. 如权利要求3所述的方法，其特征在于，
    对所述第一影像信息、第二影像信息或者第三影像信息中的运动姿势进行分析是基于预置的深度神经网络算法进行的。
  5. 如权利要求3或4所述的方法，其特征在于，还包括：
    若所述第三特征姿势满足预置的追踪确认条件，则发出第二提示消息，所述第二提示消息用于提示自动拍摄已初始化。
  6. 如权利要求5所述的方法，其特征在于，所述若所述第三特征姿势满足预置的追踪确认条件，则发出第二提示消息，包括：
    若所述第三特征姿势满足预置的追踪确认条件，则触发飞行器上配置的指示灯按照预置的发光机制发出发光提示；其中，所述飞行器用于挂载所述摄像装置。
  7. 如权利要求1-6任一项所述的方法，其特征在于，所述采集包括运动姿势的第一影像信息之前，还包括：
    在自动拍摄模式时，发出控制指令调整所述飞行器的拍摄位置；
    其中，所述控制指令用于控制所述飞行器转动或移动，直至所述飞行器挂载的摄像模块拍摄得到一个或者多个能够发出运动姿势的拍摄目标。
  8. 如权利要求1-7任一项所述的方法，其特征在于，所述采集包括运动姿势的第三影像信息之前，还包括：
    在自动拍摄模式时，发出控制指令调整所述飞行器的拍摄位置；
    其中，所述控制指令用于控制所述飞行器转动或移动，直至所述飞行器挂载的摄像模块拍摄得到一个或者多个能够发出运动姿势的拍摄目标。
  9. 如权利要求1-8任一项所述的方法，其特征在于，所述若所述第二特征姿势满足预置的拍摄条件，则触发所述摄像装置拍摄影像，包括：
    若所述第二特征姿势满足预置的拍摄条件，则进一步确定所述第二特征姿势所关联的拍摄模式；
    根据所述确定的拍摄模式，触发所述摄像装置拍摄影像；
    所述确定的拍摄模式是从预置的映射数据库中确定的，所述映射数据库中存储了特征姿势和拍摄模式的映射关系。
  10. 一种基于影像的控制方法，其特征在于，包括：
    采集包括运动姿势的飞行控制影像信息；
    对所述飞行控制影像信息中的运动姿势进行分析，确定出该飞行控制影像信息中包含的控制特征姿势；
    获取与所述确定出的控制特征姿势关联的控制指令，并根据获取的控制指令控制飞行器飞行；
    所述飞行控制影像信息是挂载在所述飞行器上的摄像装置拍摄得到的。
  11. 如权利要求10所述的方法，其特征在于，所述采集飞行控制影像信息之前，还包括：
    在采集到包括运动姿势的影像信息后，发出确认提示消息，并在发出确认提示消息，执行所述采集飞行控制影像信息。
  12. 如权利要求10或11所述的方法，其特征在于，所述对所述飞行控制影像信息中的运动姿势进行分析包括：
    基于预置的深度神经网络算法对所述飞行控制影像信息中的运动姿势进行分析。
  13. 一种基于影像的控制装置，其特征在于，包括：
    采集模块，用于采集包括运动姿势的第一影像信息；
    确定模块，用于对所述采集到的第一影像信息中的运动姿势进行分析，确定出该第一影像信息中包含的第一特征姿势；
    提示模块，用于若所述第一特征姿势满足预置的启动条件，则发出第一提示消息；
    所述采集模块，还用于在发出第一提示消息后，再次采集包括运动姿势的第二影像信息；
    所述确定模块，还用于对所述采集到的第二影像信息中的运动姿势进行分析，确定出该第二影像信息中包含的运动姿势第二特征姿势；
    处理模块，用于在所述第二特征姿势满足预置的拍摄条件时，则触发所述 摄像装置进行影像拍摄处理。
  14. 如权利要求13所述的装置，其特征在于，
    所述提示模块，具体用于在所述第一特征姿势满足预置的启动条件时，触发飞行器上配置的指示灯按照预置的发光机制发出发光提示；其中，所述飞行器用于挂载所述摄像装置。
  15. 如权利要求13或14所述的装置，其特征在于，所述装置还包括：通知模块，其中，
    所述采集模块，还用于采集包括运动姿势的第三影像信息；
    所述确定模块，还用于对所述第三影像信息中的运动姿势进行分析，确定出该第三影像信息中包含的第三特征姿势；
    所述通知模块，用于在所述第三特征姿势满足预置的追踪确认条件时，通知所述采集模块采集包括运动姿势的第一影像信息。
  16. 如权利要求15所述的装置，其特征在于，所述确定模块在用于对对所述第一影像信息、第二影像信息或者第三影像信息中的运动姿势进行分析时是基于预置的深度神经网络算法进行的。
  17. 如权利要求15或16所述的装置，其特征在于，
    所述提示模块，还用于在所述第三特征姿势满足预置的追踪确认条件时，发出第二提示消息，所述第二提示消息用于提示自动拍摄已初始化。
  18. 如权利要求17所述的装置，其特征在于，
    所述提示模块，具体用于在所述第三特征姿势满足预置的追踪确认条件时，触发飞行器上配置的指示灯按照预置的发光机制发出发光提示；其中，所述飞行器用于挂载所述摄像装置。
  19. 如权利要求13-18任一项所述的装置，其特征在于，还包括：
    位置调整模块，用于在自动拍摄模式时，发出控制指令调整所述飞行器的 拍摄位置；其中，所述控制指令用于控制所述飞行器转动或移动，直至所述飞行器挂载的摄像模块拍摄得到一个或者多个能够发出运动姿势的拍摄目标。
  20. 如权利要求13-19任一项所述的装置，其特征在于，还包括：
    位置调整模块，用于在自动拍摄模式时，发出控制指令调整所述飞行器的拍摄位置；其中，所述控制指令用于控制所述飞行器转动或移动，直至所述飞行器挂载的摄像模块拍摄得到一个或者多个能够发出运动姿势的拍摄目标。
  21. 如权利要求13-20任一项所述的装置，其特征在于，
    所述处理模块，具体用于在所述第二特征姿势满足预置的拍摄条件时，进一步确定所述第二特征姿势所关联的拍摄模式；根据所述确定的拍摄模式，触发所述摄像装置拍摄影像；所述确定的拍摄模式是从预置的映射数据库中确定的，所述映射数据库中存储了特征姿势和拍摄模式的映射关系。
  22. 一种基于影像的控制装置，其特征在于，包括：
    控制影像采集模块，用于采集包括运动姿势的飞行控制影像信息，所述飞行控制影像信息是挂载在所述飞行器上的摄像装置拍摄得到的；
    姿势确定模块，用于对所述飞行控制影像信息中的运动姿势进行分析，确定出该飞行控制影像信息中包含的控制特征姿势；
    飞行控制模块，用于获取与所述确定出的控制特征姿势关联的控制指令，并根据获取的控制指令控制飞行器飞行。
  23. 如权利要求22所述的装置，其特征在于，还包括：
    确认提示模块，用于在采集到包括运动姿势的影像信息后，发出确认提示消息，并在发出确认提示消息，执行所述采集飞行控制影像信息。
  24. 如权利要求22或23所述的装置，其特征在于，
    所述姿势确定模块，具体用于基于预置的深度神经网络算法对所述飞行控制影像信息中的运动姿势进行分析。
  25. 一种飞行器，其特征在于，包括：存储器、处理器和通信接口；
    存储器，所述存储器中存储有程序指令；
    所述处理器，调用所述存储器中存储的程序指令，用于采集包括运动姿势的第一影像信息；对所述采集到的第一影像信息中的运动姿势进行分析，确定出该第一影像信息中包含的第一特征姿势；若所述第一特征姿势满足预置的启动条件，则发出第一提示消息；在发出第一提示消息后，再次采集包括运动姿势的第二影像信息；对所述采集到的第二影像信息中的运动姿势进行分析，确定出该第二影像信息中包含的运动姿势第二特征姿势；若所述第二特征姿势满足预置的拍摄条件，则生成触发指令触发所述摄像装置进行影像拍摄处理；
    所述通信接口，用于向拍摄装置发送所述触发指令。
  26. 如权利要求25所述的飞行器，其特征在于，
    所述处理器，在用于若所述第一特征姿势满足预置的启动条件，则发出第一提示消息时，具体用于若所述第一特征姿势满足预置的启动条件，则触发飞行器上配置的指示灯按照预置的发光机制发出发光提示；其中，所述飞行器用于挂载所述摄像装置。
  27. 如权利要求25或26所述的飞行器，其特征在于，
    所述处理器，还用于采集包括运动姿势的第三影像信息；对所述第三影像信息中的运动姿势进行分析，确定出该第三影像信息中包含的第三特征姿势；若所述第三特征姿势满足预置的追踪确认条件，则执行所述采集包括运动姿势的第一影像信息。
  28. 如权利要求27所述的飞行器，其特征在于，
    所述处理器，在用于对所述第一影像信息、第二影像信息或者第三影像信息中的运动姿势进行分析时是基于预置的深度神经网络算法进行的分析。
  29. 如权利要求27或28所述的飞行器，其特征在于，
    所述处理器，还用于若所述第三特征姿势满足预置的追踪确认条件，则发出第二提示消息，所述第二提示消息用于提示自动拍摄已初始化。
  30. 如权利要求29所述的飞行器，其特征在于，
    所述处理器，在用于若所述第三特征姿势满足预置的追踪确认条件，则发出第二提示消息时，具体用于若所述第三特征姿势满足预置的追踪确认条件，则触发飞行器上配置的指示灯按照预置的发光机制发出发光提示；其中，所述飞行器用于挂载所述摄像装置。
  31. 如权利要求25-30任一项所述的飞行器，其特征在于，
    所述处理器，还用于在自动拍摄模式时，发出控制指令调整所述飞行器的拍摄位置；其中，所述控制指令用于控制所述飞行器转动或移动，直至所述飞行器挂载的摄像模块拍摄得到一个或者多个能够发出运动姿势的拍摄目标。
  32. 如权利要求25-31任一项所述的飞行器，其特征在于，
    所述处理器，还用于在自动拍摄模式时，发出控制指令调整所述飞行器的拍摄位置；其中，所述控制指令用于控制所述飞行器转动或移动，直至所述飞行器挂载的摄像模块拍摄得到一个或者多个能够发出运动姿势的拍摄目标。
  33. 如权利要求25-32任一项所述的飞行器，其特征在于，
    所述处理器，在用于若所述第二特征姿势满足预置的拍摄条件，则触发所述摄像装置拍摄影像时，具体用于若所述第二特征姿势满足预置的拍摄条件，则进一步确定所述第二特征姿势所关联的拍摄模式；根据所述确定的拍摄模式，触发所述摄像装置拍摄影像；所述确定的拍摄模式是从预置的映射数据库中确定的，所述映射数据库中存储了特征姿势和拍摄模式的映射关系。
  34. 一种飞行器，其特征在于，包括：存储器，处理器和电子调速器；
    存储器，所述存储器中存储有程序指令；
    所述处理器，调用所述存储器中存储的程序指令，所述处理器，用于采集包括运动姿势的飞行控制影像信息；对所述飞行控制影像信息中的运动姿势进行分析，确定出该飞行控制影像信息中包含的控制特征姿势；获取与所述确定出的控制特征姿势关联的控制指令，并根据获取的控制指令向所述电子调速器 发送指令，所述飞行控制影像信息是挂载在所述飞行器上的摄像装置拍摄得到的；
    所述电子调速器，用于根据所述处理器发送的指令完成对所述飞行器的飞行控制。
  35. 如权利要求34所述的飞行器，其特征在于，
    所述处理器，还用于在采集到包括运动姿势的影像信息后，发出确认提示消息，并在发出确认提示消息，执行所述采集飞行控制影像信息。
  36. 如权利要求34或35所述的飞行器，其特征在于，
    所述处理器，在用于对所述飞行控制影像信息中的运动姿势进行分析时，具体用于基于预置的深度神经网络算法对所述飞行控制影像信息中的运动姿势进行分析。
PCT/CN2016/100230 2016-09-27 2016-09-27 一种基于影像的控制方法、装置及飞行器 Ceased WO2018058264A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2016/100230 WO2018058264A1 (zh) 2016-09-27 2016-09-27 一种基于影像的控制方法、装置及飞行器
CN201680065366.4A CN108351651B (zh) 2016-09-27 2016-09-27 一种基于影像的控制方法、装置及飞行器

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2016/100230 WO2018058264A1 (zh) 2016-09-27 2016-09-27 一种基于影像的控制方法、装置及飞行器

Publications (1)

Publication Number Publication Date
WO2018058264A1 true WO2018058264A1 (zh) 2018-04-05

Family

ID=61762382

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/100230 Ceased WO2018058264A1 (zh) 2016-09-27 2016-09-27 一种基于影像的控制方法、装置及飞行器

Country Status (2)

Country Link
CN (1) CN108351651B (zh)
WO (1) WO2018058264A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021026780A1 (zh) * 2019-08-13 2021-02-18 深圳市大疆创新科技有限公司 拍摄控制方法、终端、云台、系统及存储介质

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130253733A1 (en) * 2012-03-26 2013-09-26 Hon Hai Precision Industry Co., Ltd. Computing device and method for controlling unmanned aerial vehicle in flight space
CN104767925A (zh) * 2014-01-03 2015-07-08 北京云狐时代科技有限公司 一种实现自动延时拍照、自动摄像功能的技术
CN105223957A (zh) * 2015-09-24 2016-01-06 北京零零无限科技有限公司 一种手势操控无人机的方法和装置
CN105578058A (zh) * 2016-02-03 2016-05-11 北京光年无限科技有限公司 一种面向智能机器人的拍摄控制方法、装置及机器人
CN105677300A (zh) * 2016-02-04 2016-06-15 普宙飞行器科技（深圳）有限公司 基于手势识别操控无人机的方法、无人机及系统
CN105847684A (zh) * 2016-03-31 2016-08-10 深圳奥比中光科技有限公司 无人机
CN105892474A (zh) * 2016-03-31 2016-08-24 深圳奥比中光科技有限公司 无人机以及无人机控制方法

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4896115B2 (ja) * 2008-11-21 2012-03-14 三菱電機株式会社 空中移動体からの自動追尾撮影装置
CN103365294A (zh) * 2012-03-29 2013-10-23 鸿富锦精密工业（深圳）有限公司 无人飞行载具控制系统及方法
JPWO2014077046A1 (ja) * 2012-11-13 2017-01-05 ソニー株式会社 画像表示装置及び画像表示方法、移動体装置、画像表示システム、並びにコンピューター・プログラム
JP6211428B2 (ja) * 2014-01-30 2017-10-11 Kddi株式会社 ガイダンス表示装置、方法及びプログラム
KR101771492B1 (ko) * 2014-12-29 2017-08-28 서울시립대학교 산학협력단 복수의 센서를 탑재한 무인 비행체를 이용하는 매핑 방법 및 시스템
CN105828062A (zh) * 2016-03-23 2016-08-03 常州视线电子科技有限公司 无人机3d虚拟现实拍摄系统
CN105912980B (zh) * 2016-03-31 2019-08-30 深圳奥比中光科技有限公司 无人机以及无人机系统
CN105912989B (zh) * 2016-04-01 2020-01-03 成都云图秀色科技有限公司 基于图像识别的飞行指令生成系统及方法
CN105867362A (zh) * 2016-04-20 2016-08-17 北京博瑞爱飞科技发展有限公司 终端设备和无人驾驶飞行器的控制系统

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130253733A1 (en) * 2012-03-26 2013-09-26 Hon Hai Precision Industry Co., Ltd. Computing device and method for controlling unmanned aerial vehicle in flight space
CN104767925A (zh) * 2014-01-03 2015-07-08 北京云狐时代科技有限公司 一种实现自动延时拍照、自动摄像功能的技术
CN105223957A (zh) * 2015-09-24 2016-01-06 北京零零无限科技有限公司 一种手势操控无人机的方法和装置
CN105578058A (zh) * 2016-02-03 2016-05-11 北京光年无限科技有限公司 一种面向智能机器人的拍摄控制方法、装置及机器人
CN105677300A (zh) * 2016-02-04 2016-06-15 普宙飞行器科技（深圳）有限公司 基于手势识别操控无人机的方法、无人机及系统
CN105847684A (zh) * 2016-03-31 2016-08-10 深圳奥比中光科技有限公司 无人机
CN105892474A (zh) * 2016-03-31 2016-08-24 深圳奥比中光科技有限公司 无人机以及无人机控制方法

Also Published As

Publication number Publication date
CN108351651B (zh) 2024-06-21
CN108351651A (zh) 2018-07-31

Similar Documents

Publication Publication Date Title
CN106791420B (zh) 一种拍摄控制方法及装置
JP7423685B2 (ja) 画像処理装置およびその制御方法
US12149819B2 (en) Autonomous media capturing
JP6388706B2 (ja) 無人航空機の撮影制御方法及び撮影制御装置、電子デバイス
US20160292886A1 (en) Apparatus and method for photographing people using a movable remote device
CN109196439B (zh) 无人飞行器的控制方法、设备及无人飞行器
US20160124435A1 (en) 3d scanning and imaging method utilizing a self-actuating compact unmanned aerial device
KR102475999B1 (ko) 화상 처리장치 및 그 제어방법
CN105518560B (zh) 一种基于位置的控制方法、装置、可移动机器以及机器人
WO2019179357A1 (zh) 拍摄方法、装置、智能设备及存储介质
CN108702458B (zh) 拍摄方法和装置
TW201249713A (en) Unmanned aerial vehicle control system and method
CN110574040A (zh) 自动抓拍方法及装置、无人机及存储介质
JP2019212967A (ja) 撮像装置及びその制御方法
WO2019179364A1 (zh) 拍摄方法、装置和智能设备
WO2015154359A1 (zh) 拍摄的实现方法及装置
CN111182204B (zh) 一种基于可穿戴设备的拍摄方法及可穿戴设备
CN205051791U (zh) 一种航拍设备的拍摄范围标定装置及其无人机
CN107925724A (zh) 在具有相机的设备中支持摄影的技术及其设备
TWI696122B (zh) 無人機互動拍攝系統方法
WO2018058264A1 (zh) 一种基于影像的控制方法、装置及飞行器
TW201725897A (zh) 拍照系統及方法
CN111031202A (zh) 基于四旋翼的智能拍照无人机、智能拍照系统及其方法
CN103379269A (zh) 拍摄系统、拍摄方法及使用该拍摄系统的电子设备
WO2019024010A1 (zh) 一种图像处理方法、系统以及智能导盲装置

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 201680065366.4

Country of ref document: CN

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

Ref document number: 16917024

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16917024

Country of ref document: EP

Kind code of ref document: A1