WO2020024104A1 - 返航控制方法、装置及设备 - Google Patents

返航控制方法、装置及设备 Download PDF

Info

Publication number
WO2020024104A1
WO2020024104A1 PCT/CN2018/097776 CN2018097776W WO2020024104A1 WO 2020024104 A1 WO2020024104 A1 WO 2020024104A1 CN 2018097776 W CN2018097776 W CN 2018097776W WO 2020024104 A1 WO2020024104 A1 WO 2020024104A1
Authority
WO
WIPO (PCT)
Prior art keywords
target object
drone
image
identity information
take
Prior art date
Application number
PCT/CN2018/097776
Other languages
English (en)
French (fr)
Inventor
孟国涛
庞磊
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/CN2018/097776 priority Critical patent/WO2020024104A1/zh
Priority to CN201880010540.4A priority patent/CN110291482A/zh
Publication of WO2020024104A1 publication Critical patent/WO2020024104A1/zh

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
    • G05D1/10—Simultaneous control of position or course in three dimensions
    • G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft

Definitions

  • the present invention relates to the field of communication technologies, and in particular, to a method, a device, and a device for returning home control.
  • Embodiments of the present invention provide a home return control method, device, and device, which can automatically identify a target object during the home return process and improve the convenience of operation.
  • a first aspect of the embodiments of the present invention is to provide a home return control method, including:
  • a second aspect of the embodiments of the present invention is to provide a return flight control device, including:
  • 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:
  • a third aspect of the embodiments of the present invention is to provide a drone, including:
  • a power system installed on the fuselage and configured to provide power to the drone
  • An image acquisition device installed on the fuselage and configured to acquire a first image including an object
  • the identity information of the target object is acquired during the take-off of the drone, and the first image containing the object is collected by the image acquisition device during the return of the drone, and according to the first image and the identity information, Recognize the object as a target object and perform corresponding operations based on the target object without the need for the user to re-designate the target object.
  • the target object can be automatically identified during the return flight to improve the convenience of the operation.
  • FIG. 1 is a flowchart of a home return control method according to an embodiment of the present invention
  • FIG. 2 is a schematic diagram of a drone taking off scene provided by an embodiment of the present invention
  • FIG. 3 is a schematic diagram of a drone returning scene provided by an embodiment of the present invention.
  • FIG. 4 is a flowchart of another home return control method according to an embodiment of the present invention.
  • FIG. 5 is a schematic diagram of another drone returning scenario provided by an embodiment of the present invention.
  • FIG. 6 is a structural diagram of a home return control device according to an embodiment of the present invention.
  • a component when a component is called “fixed to” another component, it may be directly on another component or a centered component may exist. When a component is considered to be “connected” to another component, it can be directly connected to another component or a centered component may exist at the same time.
  • An embodiment of the present invention provides a home return control method for acquiring identity information of a target object during a drone take-off process, and during a drone return home process, acquiring a first image including an object through an image acquisition device, According to the first image and the identity information, the identified object is the target object, and the corresponding operation is performed based on the target object.
  • the user can automatically identify the target object without re-designating the target object, which improves the convenience of the operation.
  • the process of taking off of the drone refers to the process of moving the drone to the take-off point, wherein the geographic coordinates of the take-off point are the same as the geographic coordinates selected by the user, and the height of the take-off point is the same as the height selected by the user .
  • the user can select the geographic coordinates of the drone taking off in the digital map displayed on the ground station, and select the altitude in the altitude menu.
  • the ground station sends the geographic coordinates and altitude selected by the user to the drone.
  • the drone can be based on Geographical coordinates and altitude to the takeoff point.
  • the height menu may include at least one height value or a slider for indicating at least one height value.
  • the user can operate the remote control with a joystick to control the drone to fly to the take-off point.
  • the identity information may include human characteristics and / or face characteristics of the target object.
  • Human characteristics may include the clothing style of the target object and its color, or a human body support.
  • the facial features may include structural relationships among the target's eyes, nose, mouth, and chin.
  • the drone may acquire an image containing the target object through an image acquisition device, and process the image to obtain the identity information of the target object.
  • a drone may acquire a human body image including a target object, and process the human body image to obtain the human body characteristics of the target object.
  • a drone may acquire a face image including a target object through an image acquisition device, and process the face image to obtain the facial features of the target object.
  • the drone when the drone is taking off, when the distance between the drone and the target object is relatively long, the drone can obtain an image of a human body containing the target object through an image acquisition device, and process the human body image.
  • the human characteristics of the target object are obtained; when the drone is flying near the target object, the distance between the drone and the target object is relatively short, and a face image including the target object can be collected and processed. To obtain the facial features of the target object. For another example, when the drone is taking off, when the distance between the drone and the target object is relatively long, the drone can obtain an image of a human body containing the target object through an image acquisition device, and process the human body image.
  • the human characteristics of the target object are obtained; further, the drone may adjust the focal length of the image acquisition device, collect a face image containing the target object, and process the face image to obtain the facial characteristics of the target object.
  • the drone may receive the second image containing the target object sent by the ground station, and process the second image to obtain the identity information of the target object. For example, before the drone takes off to the take-off point, the user can find the second image containing the target object in the local station's local memory, and the ground station sends the second image found by the user to the drone, then the drone The second image can be processed to obtain the identity information of the target object. For another example, before the drone takes off to the take-off point, the user can collect a second image containing the target object through the ground station, and the ground station sends the collected second image to the drone, and the drone can The image is processed to obtain the identity information of the target object.
  • the drone in the process of obtaining the identity information of the target object, if the image acquired by the drone through the image acquisition device contains an object, the drone can identify the object as the target object; if the drone passes the image acquisition The image obtained by the device contains multiple objects, and the drone can send the image to the ground station, and the ground station displays the image so that the user can select a target object among the above multiple objects, and the ground station can identify the target object selected by the user The ground station sends the identified image to the drone, and the drone can identify the target object in the identified image and obtain the identity information of the target object.
  • performing the corresponding operation based on the target object may specifically include the following three types:
  • the drone enters the gesture control mode during take-off, performs corresponding operations based on the gesture of the target object, and performs a return operation when the return instruction or a signal loss received by the user is entered; when the drone returns,
  • performing the corresponding operation based on the gesture of the target object may specifically include: performing a spiral flight when the first gesture of the target object is detected; and performing a circular flight with the preset position as the center when the second gesture of the target object is detected ,and many more.
  • the drone enters the self-timer mode during take-off, the drone performs the return operation when a return instruction or a signal loss is input by the user; when the target object is identified during the return flight of the drone, Automatically enters self-timer mode again.
  • target tracking mode based on the target object. For example, when the drone enters the target tracking mode during take-off, the drone performs the return operation when a return instruction or a signal loss is input by the user; when the target is identified during the return flight of the drone, Can automatically enter the target tracking mode again.
  • FIG. 1 is a flowchart of a home return control method according to an embodiment of the present invention. As shown in FIG. 1, the method may include the following steps:
  • Step 101 During the take-off of the drone, obtain the identity information of the target object.
  • an image containing the target object may be collected by an image acquisition device, and the image may be processed to obtain the identity information of the target object.
  • the identity information may include human characteristics and / or face characteristics of the target object.
  • the preset position may be a position on the ground or the target's hand or on the desktop.
  • the drone can collect the image data from the image acquisition device before flying to the take-off point.
  • the image of the target object If the drone does not face the target object during take-off, the drone can rotate the pan / tilt after flying to the take-off point to control the image acquisition device to face the target object.
  • the image of the object If the image acquisition device is facing the target object during the take-off process, the drone can collect the image data from the image acquisition device before flying to the take-off point. The image of the target object. If the drone does not face the target object during take-off, the drone can rotate the pan / tilt after flying to the take-off point to control the image acquisition device to face the target object. The image of the object.
  • the drone may receive the second image containing the target object sent by the ground station, and process the second image to obtain the identity information of the target object.
  • Step 102 In the course of returning the drone, a first image including an object is collected by an image acquisition device.
  • the drone when the drone receives a return instruction or a signal loss sent by the ground station, the drone will perform a return operation, that is, fly from the current position to the take-off point or around the take-off point.
  • the first image containing the object may include a human body image and / or a human face image. Taking the scenario of the drone returning as described in FIG. 3 as an example, the drone can fly obliquely below the take-off point, so that the image of the face containing the object can be collected by the image acquisition device.
  • the drone may acquire a first image including an object through an image acquisition device at a preset interval, until it recognizes that the object included in the first image is the target object.
  • the drone can fly from its current position to the take-off point during the return flight. During the flight from the current position to the take-off point, the drone can adjust the focal length of the image acquisition device so that the image acquisition device can capture the human body image containing the object. If the drone captures a human body image containing the object before flying to the takeoff point, the drone can process the human body image to obtain the human body characteristic of the object. If the human body characteristic of the object matches the human body characteristic of the target object, the drone can Identify the object as the target object, and then perform corresponding operations based on the target object.
  • the drone can fly from the current position to the take-off point.
  • the drone can adjust the focal length of the image capture device so that the image capture device can capture a face image containing the object. If the drone collects a face image containing the object before flying to the take-off point, the drone can process the face image to obtain the face feature of the object. If the face feature of the object matches the face feature of the target object , The drone can recognize the object as the target object, and then perform corresponding operations based on the target object.
  • the drone can fly from the current position to below the takeoff point.
  • the drone can collect an image of a human body containing an object through an image acquisition device. If a drone captures a human body image containing an object before flying below the takeoff point, the drone can process the human body image to obtain the human body characteristics of the object. If the human body characteristics of the object and the human body characteristics of the target object match, no one The machine can recognize the object as the target object, and then perform corresponding operations based on the target object.
  • the drone can fly from the current position to below the takeoff point.
  • the drone can acquire an image of a face containing an object through an image acquisition device. If the drone acquires a face image containing the object before flying below the takeoff point, the drone can process the face image to obtain the face feature of the object. If the face feature of the object and the face of the target object Feature matching, the drone can recognize the object as the target object, and then perform corresponding operations based on the target object.
  • the drone can fly from the current position to the area around the takeoff point.
  • the drone can collect an image of the human body containing the object through an image acquisition device. If the drone captures a human body image containing the object before flying around the take-off point, the drone can process the human body image to obtain the human body characteristic of the object. If the human body characteristic of the object matches the human body characteristic of the target object, no The human machine can be lowered to a certain height, so that the face image of the object can be collected by the image acquisition device, and the face image is processed to obtain the face characteristics of the object. If the face characteristics of the object and the face of the target object Feature matching, the drone can recognize the object as the target object, and then perform corresponding operations based on the target object.
  • the drone can fly from the current position to the area around the takeoff point.
  • the drone can collect an image of the human body containing the object through an image acquisition device. If the drone captures a human body image containing the object before flying around the take-off point, the drone can process the human body image to obtain the human body characteristic of the object. If the human body characteristic of the object matches the human body characteristic of the target object, no The human machine can adjust the focal length of the image acquisition device, so that the face image of the object can be collected by the image acquisition device, and the face image is processed to obtain the facial features of the object. If the facial features of the object and the target object's Face features match, the drone can recognize the object as the target object.
  • flight trajectory of the drone flying from the current position to the take-off point and the flight trajectory of the drone flying from the take-off point to the current position may be the same or different, and are not specifically limited by the embodiments of the present invention.
  • Step 103 According to the first image and the identity information, the identification object is a target object.
  • the drone may process the first image to obtain the identity information of the object, and if the identity information of the object and the identity information of the target object match, the identified object is the target object.
  • the identity information of the object includes human characteristics and / or face characteristics of the object.
  • the drone may determine whether the human characteristics of the object and the human characteristics of the target object match through pedestrian re-identification (ReID) technology.
  • the human features match the human features of the target object, and the drone can recognize the object as the target object.
  • the drone may compare the human characteristics of the object with the human characteristics of the target object to obtain a distance value. If the distance value is less than the first preset distance threshold, the drone may determine the human characteristics of the object and the human characteristics of the target object. match.
  • the drone may use face recognition technology to determine whether the facial features of the object and the facial features of the target object match.
  • the facial features of the target object match, and the drone can recognize the object as the target object.
  • the drone may compare the facial features of the object and the facial features of the target object to obtain a distance value. If the distance value is less than a second preset distance threshold, the drone may determine the facial features of the object and the target object. Facial features match.
  • the drone may determine whether the human characteristics of the object and the human characteristics of the target object match, and if the human characteristics of the object and the human characteristics of the target object match Matching, it is judged whether the facial features of the object and the facial features of the target object match, and if the facial features of the object and the facial features of the target object match, the object is identified as the target object.
  • the human characteristics of the object match the human characteristics of the target object.
  • the drone can determine that the similarity between the object and the target object is high. Further, the facial features of the object match the facial features of the target object. The drone can determine the object as the target object, improving the accuracy of target object recognition.
  • Step 104 Perform a corresponding operation based on the target object.
  • the drone After the drone recognizes the object as the target object, it can enter gesture control mode, selfie mode, or target tracking mode based on the target object.
  • the identity information of the target object is acquired during the take-off of the drone, and the first image containing the object is collected by the image acquisition device during the return of the drone, and is identified based on the first image and the identity information.
  • This object is the target object, and the corresponding operation is performed based on the target object. The user does not need to re-designate the target object, and the target object can be automatically identified during the return flight to improve the convenience of the operation.
  • FIG. 4 is a flowchart of another home return control method according to an embodiment of the present invention. As shown in FIG. 4, the method may include the following steps:
  • Step 401 During the take-off of the drone, obtain the identity information of the target object.
  • an image containing the target object may be collected by an image acquisition device, and the image may be processed to obtain the identity information of the target object.
  • the identity information may include human characteristics and / or face characteristics of the target object.
  • the preset position may be a position on the ground or the target's hand or on the desktop.
  • the drone can collect the image data from the image acquisition device before flying to the take-off point.
  • the image of the target object If the drone does not face the target object during take-off, the drone can rotate the pan / tilt after flying to the take-off point to control the image acquisition device to face the target object.
  • the image of the object If the image acquisition device is facing the target object during the take-off process, the drone can collect the image data from the image acquisition device before flying to the take-off point. The image of the target object. If the drone does not face the target object during take-off, the drone can rotate the pan / tilt after flying to the take-off point to control the image acquisition device to face the target object. The image of the object.
  • the drone may receive the second image containing the target object sent by the ground station, and process the second image to obtain the identity information of the target object.
  • Step 402 Obtain a take-off position of the drone during take-off.
  • the drone can obtain the geographic coordinates and altitude of the take-off position. For example, if the drone flies to the takeoff point based on the geographic coordinates and altitude sent by the ground station, the drone can use the geographic coordinates and altitude sent by the ground station as the takeoff position of the drone during takeoff. For another example, if the drone flies to the take-off point based on the user's joystick operation on the remote control, the drone can obtain the take-off position of the drone during take-off by using a preset positioning technology.
  • the preset positioning technology may include Global Positioning System (GPS) technology, base station communication positioning technology, or obtaining the altitude of the drone flying to the take-off point by using a barometric pressure sensor.
  • GPS Global Positioning System
  • Step 403 During the drone return flight, control the drone return flight to the take-off position.
  • the drone when the drone receives a return instruction or a signal loss sent by the ground station, the drone will perform a return operation, that is, fly from the current position to the take-off position. Taking the scenario of the drone returning as described in FIG. 5 as an example, after the drone can fly to the take-off point, it descends to a certain height so that the face image containing the object can be collected by the image acquisition device.
  • Step 404 Collect a first image including an object through an image acquisition device.
  • a first image including an object may be collected by an image acquisition device.
  • the first image containing the object may include a human body image and / or a human face image.
  • a drone can fly from the current position to the take-off point, collect an image of the human body containing the object through an image acquisition device, and process the human image to obtain the human characteristics of the object. If the human characteristics of the object and the target object Matching human characteristics, the drone can recognize the object as the target object.
  • the drone can fly from the current position to the take-off point, adjust the focus of the image acquisition device, so that the image acquisition device can collect the face image containing the object, and process the face image to obtain the object's Face features. If the face features of the object match the face features of the target object, the drone can recognize the object as the target object.
  • the drone can fly from the current position to the take-off point, collect the image of the human body containing the object through the image acquisition device, and process the human image to obtain the human characteristics of the object.
  • the human characteristics of the target object are matched, and the drone can be lowered to a certain height, so that the face image of the object can be collected by the image acquisition device, and the face image is processed to obtain the face characteristics of the object.
  • the facial features match the facial features of the target object, and the drone can recognize the object as the target object.
  • the drone can fly from the current position to the take-off point, collect the image of the human body containing the object through the image acquisition device, and process the human image to obtain the human characteristics of the object.
  • the human characteristics of the target object are matched.
  • the drone can adjust the focal length of the image acquisition device so that the face image of the object can be collected by the image acquisition device, and the face image is processed to obtain the face characteristics of the object.
  • the facial features of the target match the facial features of the target object, and the drone can recognize the object as the target object.
  • flight trajectory of the drone from the current position to the take-off point and the flight trajectory of the drone from the take-off point to the current position may be the same or different, and are not specifically limited by the embodiments of the present invention.
  • Step 405 According to the first image and the identity information, identify the object as a target object.
  • the drone may process the first image to obtain the identity information of the object, and if the identity information of the object and the identity information of the target object match, the identified object is the target object.
  • the identity information of the object includes human characteristics and / or face characteristics of the object.
  • the drone may determine whether the human characteristics of the object and the human characteristics of the target object match through the ReID technology. Matching, the drone can identify the object as the target object. For example, the drone may compare the human characteristics of the object with the human characteristics of the target object to obtain a distance value. If the distance value is less than the first preset distance threshold, the drone may determine the human characteristics of the object and the human characteristics of the target object. match.
  • the drone may use face recognition technology to determine whether the facial features of the object and the facial features of the target object match.
  • the facial features of the target object match, and the drone can recognize the object as the target object.
  • the drone may compare the facial features of the object and the facial features of the target object to obtain a distance value. If the distance value is less than a second preset distance threshold, the drone may determine the facial features of the object and the target object. Facial features match.
  • the drone may determine whether the human characteristics of the object and the human characteristics of the target object match, and if the human characteristics of the object and the human characteristics of the target object match Matching, it is judged whether the facial features of the object and the facial features of the target object match, and if the facial features of the object and the facial features of the target object match, the object is identified as the target object.
  • the human characteristics of the object match the human characteristics of the target object.
  • the drone can determine that the similarity between the object and the target object is high. Further, the facial features of the object match the facial features of the target object. The drone can determine the object as the target object, improving the accuracy of target object recognition.
  • Step 406 Perform a corresponding operation based on the target object.
  • the drone After the drone recognizes the object as the target object, it can enter gesture control mode, selfie mode, or target tracking mode based on the target object.
  • the identity information of the target object and the take-off position of the drone during take-off are acquired.
  • the drone is controlled to return to the take-off position.
  • the image acquisition device collects a first image containing an object, and recognizes the object as a target object based on the first image and identity information, and performs corresponding operations based on the target object. The user does not need to re-specify the target object, and can automatically identify the target object during the return flight To improve the convenience of operation.
  • FIG. 6 is a structural diagram of a home return control device according to an embodiment of the present invention.
  • the home return control device 600 includes a memory 601 and a processor 602.
  • the program code is stored, and the processor 602 calls the program code in the memory.
  • the processor 602 performs the following operations: during the take-off of the drone, obtaining the identity information of the target object; when the drone returns In the process, a first image containing an object is collected by an image acquisition device; according to the first image and identity information, the identified object is a target object; and a corresponding operation is performed based on the target object.
  • the processor 602 when the processor 602 identifies the object as the target object based on the first image and the identity information, the processor 602 performs the following operations:
  • Identity information of the object includes human characteristics and / or face characteristics of the object
  • the identity information of the object matches the identity information of the target object, identifying the object as the target object.
  • the identity information of the target object includes human characteristics and face characteristics of the target object
  • the processor 602 When the identity information of the object and the identity information of the target object match, the processor 602 performs the following operations when the object is identified as the target object:
  • the processor 602 collects the first image containing the object through the image acquisition device during the drone return flight, the following operations are performed:
  • the first image is acquired by the image acquisition device.
  • the processor 602 collects the first image containing the object through the image acquisition device during the drone return flight, the following operations are performed:
  • the first image containing the object is collected by the image acquisition device every preset time interval until it is recognized that the object contained in the first image is the target object.
  • the processor 602 when the processor 602 takes off the identity information of the target object during the take-off of the drone, the processor performs the following operations:
  • the processor 602 when the processor 602 takes off the identity information of the target object during the take-off of the drone, the processor performs the following operations:
  • the processor 602 when performing a corresponding operation based on the target object, performs the following operations:
  • the return-to-home control device provided in this embodiment can execute the return-to-home control method of the drone provided in the foregoing embodiment, and the execution manner and beneficial effects thereof are similar, and will not be repeated here.
  • An embodiment of the present invention also provides a drone.
  • the drone includes:
  • a power system installed on the fuselage and configured to provide power to the drone
  • An image acquisition device installed on the fuselage and configured to acquire a first image including an object
  • the drone further includes a position sensor
  • the position sensor is installed on the fuselage and is used to obtain a take-off position of the drone during take-off.
  • the drone further includes:
  • a communication device is installed on the fuselage and is used for information interaction with a ground station.
  • the disclosed apparatus and method may be implemented in other manners.
  • the device embodiments described above are only schematic.
  • the division of the unit is only a logical function division.
  • multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, which may be electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, may be located in one place, or may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objective of the solution of this embodiment.
  • each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist separately physically, or two or more units may be integrated into one unit.
  • the above integrated unit may be implemented in the form of hardware, or in the form of hardware plus software functional units.
  • the above integrated unit implemented in the form of a software functional unit may be stored in a computer-readable storage medium.
  • the software functional unit is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) or a processor to execute the methods described in the embodiments of the present invention. Some steps.
  • the aforementioned storage media include: U disks, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks or compact discs, and other media that can store program codes .

Landscapes

  • Engineering & Computer Science (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)

Abstract

本发明实施例提供一种返航控制方法、装置及设备，其中该方法包括在无人机起飞的过程中，获取目标对象的身份信息；在无人机返航的过程中，通过图像采集设备采集包含对象的第一图像；根据第一图像和身份信息，识别对象为目标对象；基于目标对象进行相应操作。本发明实施例能够在返航过程中自动识别目标对象，提升操作的便利性。

Description

返航控制方法、装置及设备 技术领域
本发明涉及通信技术领域，尤其涉及返航控制方法、装置及设备。
背景技术
随着科学技术的发展，图像采集设备(如摄像机)在家庭、工业、军事等领域的应用日益广泛，并且随着飞行器技术的发展，无人驾驶飞机简称无人机(Unmanned Aerial Vehicle，UAV)在家庭、工业、军事等领域的应用同样日益广泛。安装图像采集设备的无人机在基于目标对象进行航拍的过程中因为信号丢失等原因可实现返航操作。但是传统的无人机只能返航至无人机的起飞位置，如果用户希望无人机继续基于上述目标对象进行航拍，用户需要重新指定目标对象，以便无人机重新基于用户指定的目标对象进行航拍，操作的便利性不足。
发明内容
本发明实施例提供返航控制方法、装置及设备，可在返航过程中自动识别目标对象，提升操作的便利性。
本发明实施例的第一方面是提供一种返航控制方法，包括：
在无人机起飞的过程中，获取目标对象的身份信息；
在所述无人机返航的过程中，通过图像采集设备采集包含对象的第一图像；
根据所述第一图像和所述身份信息，识别所述对象为所述目标对象；
基于所述目标对象进行相应操作。
本发明实施例的第二方面是提供一种返航控制装置，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：
在无人机起飞的过程中，获取目标对象的身份信息；
在所述无人机返航的过程中，通过图像采集设备采集包含对象的第一图像；
根据所述第一图像和所述身份信息，识别所述对象为所述目标对象；
基于所述目标对象进行相应操作。
本发明实施例的第三方面是提供一种无人机，包括：
机身；
动力***，安装在所述机身，用于为所述无人机提供动力；
图像采集设备，安装在所述机身，用于采集包含对象的第一图像；
以及上述第二方面提供的返航控制装置。
本发明实施例，在无人机起飞的过程中，获取目标对象的身份信息，在无人机返航的过程中，通过图像采集设备采集包含对象的第一图像，根据第一图像和身份信息，识别该对象为目标对象，并基于目标对象进行相应操作，无需用户重新指定目标对象，可在返航过程中自动识别目标对象，提升操作的便利性。
附图说明
图1是本发明实施例提供的一种返航控制方法的流程图；
图2是本发明实施例提供的一种无人机起飞的场景示意图；
图3是本发明实施例提供的一种无人机返航的场景示意图；
图4是本发明实施例提供的另一种返航控制方法的流程图；
图5为本发明实施例提供的另一种无人机返航的场景示意图；
图6是本发明实施例提供的一种返航控制装置的结构图。
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做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。
需要说明的是，当组件被称为“固定于”另一个组件，它可以直接在另一个组件上或者也可以存在居中的组件。当一个组件被认为是“连接”另一个组件， 它可以是直接连接到另一个组件或者可能同时存在居中组件。
除非另有定义，本文所使用的所有的技术和科学术语与属于本发明的技术领域的技术人员通常理解的含义相同。本文中在本发明的说明书中所使用的术语只是为了描述具体的实施例的目的，不是旨在于限制本发明。本文所使用的术语“及/或”包括一个或多个相关的所列项目的任意的和所有的组合。
下面结合附图，对本发明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。
本发明实施例提供一种返航控制方法，用于在无人机起飞的过程中，获取目标对象的身份信息，在无人机返航的过程中，通过图像采集设备采集包含对象的第一图像，根据第一图像和身份信息，识别对象为目标对象，并基于目标对象进行相应操作，无需用户重新指定目标对象，可自动识别目标对象，提高操作的便利性。
本发明实施例中，无人机起飞的过程，指的是无人机移动至起飞点的过程，其中起飞点的地理坐标与用户选取的地理坐标相同，起飞点的高度与用户选取的高度相同。例如，用户可以在地面站显示的数字地图中选取无人机起飞的地理坐标，并在高度菜单中选取高度，地面站将用户选取的地理坐标和高度发送给无人机，无人机可以基于地理坐标和高度飞行至起飞点。其中，高度菜单可以包括至少一个高度数值或者用于指示至少一个高度数值的滑动条。又如，用户可以对遥控器进行摇杆操作，以控制无人机飞行至起飞点。
本发明实施例中，身份信息可以包括目标对象的人体特征和/或人脸特征。人体特征可以包括目标对象的服装样式及其颜色或者人体支架等。人脸特征可以包括目标对象的眼睛、鼻子、嘴巴以及下巴等器官之间的结构关系。
在一些实施例中，在无人机起飞的过程中，无人机可以通过图像采集设备获取包含目标对象的图像，对该图像进行处理，得到目标对象的身份信息。例如，无人机在起飞的过程中，可以获取包含目标对象的人体图像，对该人体图像进行处理，得到所述目标对象的人体特征。又如，无人机在起飞的过程中，可以通过图像采集设备获取包含目标对象的人脸图像，对该人脸图像进行处理，得到所述目标对象的人脸特征。又如，无人机在起飞的过程中，当无人机与目标对象之间的距离较远时，无人机可以通过图像采集设备获取包含目标对象的人体图像，对该人体图像进行处理，得到所述目标对象的人体特征；当无 人机飞行至目标对象附近时，无人机与目标对象之间的距离较近，可以采集包含目标对象的人脸图像，对该人脸图像进行处理，得到所述目标对象的人脸特征。又如，无人机在起飞的过程中，当无人机与目标对象之间的距离较远时，无人机可以通过图像采集设备获取包含目标对象的人体图像，对该人体图像进行处理，得到所述目标对象的人体特征；进一步的，无人机可以调整图像采集设备的焦距，采集包含目标对象的人脸图像，对该人脸图像进行处理，得到所述目标对象的人脸特征。
在一些实施例中，在无人机起飞的过程中，无人机可以接收地面站发送的包含目标对象的第二图像，对第二图像进行处理，得到目标对象的身份信息。例如，在无人机起飞至起飞点之前，用户可以在地面站的本地存储器中查找包含目标对象的第二图像，地面站将用户查找到的第二图像发送至无人机，则无人机可以对第二图像进行处理，得到目标对象的身份信息。又如，在无人机起飞至起飞点之前，用户可以通过地面站采集包含目标对象的第二图像，地面站将采集到的第二图像发送至无人机，则无人机可以对第二图像进行处理，得到目标对象的身份信息。
需要说明的是，在获取目标对象的身份信息的过程中，如果无人机通过图像采集设备获取的图像包含一个对象，无人机可将该对象识别为目标对象；如果无人机通过图像采集设备获取的图像包含多个对象，无人机可将该图像发送至地面站，地面站显示该图像，以便用户在上述多个对象中选取目标对象，地面站可以对用户选取的目标对象进行标识，地面站将标识后的图像发送至无人机，无人机可以在标识后的图像中识别目标对象，并获取目标对象的身份信息。
本发明实施例中，基于目标对象进行相应操作具体可以包括如下三种：
一、基于目标对象进入手势控制模式。例如，无人机在起飞过程中进入手势控制模式，基于目标对象的手势进行相应操作，在接收到用户输入的返航指令或者信号丢失等情况下，无人机进行返航操作；在无人机返航的过程中识别到目标对象时，可自动再次进入手势控制模式。示例性的，基于目标对象的手势进行相应操作具体可以为：检测到目标对象的第一手势时，进行螺旋飞行；检测到目标对象的第二手势时，以预设位置为中心进行环形飞行，等等。
二、基于目标对象进入自拍模式。例如，无人机在起飞过程中进入自拍模式，在接收到用户输入的返航指令或者信号丢失等情况下，无人机进行返航操 作；在无人机返航的过程中识别到目标对象时，可自动再次进入自拍模式。
三、基于目标对象进行目标跟踪模式。例如，无人机在起飞过程中进入目标跟踪模式，在接收到用户输入的返航指令或者信号丢失等情况下，无人机进行返航操作；在无人机返航的过程中识别到目标对象时，可自动再次进入目标跟踪模式。
下面对本发明实施例中的返航控制方法进行举例描述。
本发明实施例提供一种返航控制方法。图1是本发明实施例提供的返航控制方法的流程图，如图1所示，该方法可以包括如下步骤：
步骤101、在无人机起飞的过程中，获取目标对象的身份信息。
在一些实施例中，无人机从预设位置起飞至起飞点的过程中，可以通过图像采集设备采集包含目标对象的图像，并对该图像进行处理，得到目标对象的身份信息。其中，身份信息可以包括目标对象的人体特征和/或人脸特征。预设位置可以为地面或者目标对象的手上或者桌面等位置。
以图2所述的无人机起飞的场景示意图为例，若无人机在起飞过程中，图像采集设备朝向目标对象，则无人机可以在飞行至起飞点之前，通过图像采集设备采集包含目标对象的图像。若无人机在起飞过程中，图像采集设备未朝向目标对象，则无人机可以在飞行至起飞点之后，通过云台转动，控制图像采集设备朝向目标对象，并通过图像采集设备采集包含目标对象的图像。
在一些实施例中，在无人机起飞的过程中，无人机可以接收地面站发送的包含目标对象的第二图像，对第二图像进行处理，得到目标对象的身份信息。
步骤102、在无人机返航的过程中，通过图像采集设备采集包含对象的第一图像。
在本发明实施例中，无人机在接收到地面站发送的返航指令或者信号丢失等情况下会进行返航操作，即从当前位置飞行至起飞点或者起飞点的周围。包含对象的第一图像可以包括人体图像和/或人脸图像。以图3所述的无人机返航的场景示意图为例，无人机可斜飞至起飞点的下方，以便可通过图像采集设备采集到包含对象的人脸图像。
在一些实施例中，在无人机返航至起飞位置之前，无人机可以每间隔预设时长通过图像采集设备采集包含对象的第一图像，直至识别到第一图像所包含 对象为目标对象。
例如，无人机在返航过程中，可以由当前位置飞行至起飞点。无人机在由当前位置飞行至起飞点的过程中可以调整图像采集设备的焦距，以便图像采集设备可采集到包含对象的人体图像。如果无人机飞行至起飞点之前采集到包含对象的人体图像，无人机可以对人体图像进行处理，得到对象的人体特征，如果对象的人体特征和目标对象的人体特征匹配，无人机可以将对象识别为目标对象，进而基于目标对象进行相应操作。
又如，无人机在返航过程中，可以由当前位置飞行至起飞点。无人机在由当前位置飞行至起飞点的过程中可以调整图像采集设备的焦距，以便图像采集设备可采集到包含对象的人脸图像。如果无人机飞行至起飞点之前采集到包含对象的人脸图像，无人机可以对人脸图像进行处理，得到对象的人脸特征，如果对象的人脸特征和目标对象的人脸特征匹配，无人机可以将对象识别为目标对象，进而基于目标对象进行相应操作。
又如，无人机在返航过程中，可以由当前位置飞行至起飞点的下方。无人机在由当前位置飞行至起飞点的下方的过程中可以通过图像采集设备采集包含对象的人体图像。如果无人机飞行至起飞点的下方之前采集到包含对象的人体图像，无人机可以对人体图像进行处理，得到对象的人体特征，如果对象的人体特征和目标对象的人体特征匹配，无人机可以将对象识别为目标对象，进而基于目标对象进行相应操作。
又如，无人机在返航过程中，可以由当前位置飞行至起飞点的下方。无人机在由当前位置飞行至起飞点的下方的过程中可以通过图像采集设备采集包含对象的人脸图像。如果无人机飞行至起飞点的下方之前采集到包含对象的人脸图像，无人机可以对人脸图像进行处理，得到对象的人脸特征，如果对象的人脸特征和目标对象的人脸特征匹配，无人机可以将对象识别为目标对象，进而基于目标对象进行相应操作。
又如，无人机在返航过程中，可以由当前位置飞行至起飞点的周围。无人机在由当前位置飞行至起飞点的周围的过程中可以通过图像采集设备采集包含对象的人体图像。如果无人机飞行至起飞点的周围之前采集到包含对象的人体图像，无人机可以对该人体图像进行处理，得到对象的人体特征，如果对象的人体特征和目标对象的人体特征匹配，无人机可以降落至一定高度，以便可 以通过图像采集设备采集到该对象的人脸图像，对该人脸图像进行处理，得到对象的人脸特征，如果对象的人脸特征和目标对象的人脸特征匹配，无人机可以将对象识别为目标对象，进而基于目标对象进行相应操作。
又如，无人机在返航过程中，可以由当前位置飞行至起飞点的周围。无人机在由当前位置飞行至起飞点的周围的过程中可以通过图像采集设备采集包含对象的人体图像。如果无人机飞行至起飞点的周围之前采集到包含对象的人体图像，无人机可以对该人体图像进行处理，得到对象的人体特征，如果对象的人体特征和目标对象的人体特征匹配，无人机可以调整图像采集设备的焦距，以便可以通过图像采集设备采集到该对象的人脸图像，对该人脸图像进行处理，得到对象的人脸特征，如果对象的人脸特征和目标对象的人脸特征匹配，无人机可以将对象识别为目标对象。
需要说明的是，无人机从当前位置飞行至起飞点的飞行轨迹和无人机从起飞点飞行至当前位置的飞行轨迹可以相同，也可以不相同，具体不受本发明实施例的限制。
步骤103、根据第一图像和身份信息，识别对象为目标对象。
本发明实施例中，无人机可以对第一图像进行处理，得到对象的身份信息，若对象的身份信息和目标对象的身份信息匹配，则识别对象为目标对象。其中，对象的身份信息包括对象的人体特征和/或人脸特征。
在一些实施例中，如果对象的身份信息包括对象的人体特征，无人机可以通过行人重识别(human re-identification，ReID)技术判断对象的人体特征和目标对象的人体特征是否匹配，如果对象的人体特征和目标对象的人体特征匹配，无人机可以识别对象为目标对象。例如，无人机可以将对象的人体特征和目标对象的人体特征进行比较得到距离值，如果该距离值小于第一预设距离阈值，无人机可以确定对象的人体特征和目标对象的人体特征匹配。
在一些实施例中，如果对象的身份信息包括对象的人脸特征，无人机可以通过人脸识别技术判断对象的人脸特征和目标对象的人脸特征是否匹配，如果对象的人脸特征和目标对象的人脸特征匹配，无人机可以识别对象为目标对象。例如，无人机可以将对象的人脸特征和目标对象的人脸特征进行比较得到距离值，如果该距离值小于第二预设距离阈值，无人机可以确定对象的人脸特征和目标对象的人脸特征匹配。
在一些实施例中，如果对象的身份信息包括对象的人体特征和人脸特征，无人机可以判断对象的人体特征和目标对象的人体特征是否匹配，若对象的人体特征和目标对象的人体特征匹配，则判断对象的人脸特征和目标对象的人脸特征是否匹配，若对象的人脸特征和目标对象的人脸特征匹配，则识别对象为目标对象。本发明实施例中，对象的人体特征和目标对象的人体特征匹配，无人机可以确定对象和目标对象的相似度较高，进一步的，对象的人脸特征和目标对象的人脸特征匹配，无人机可以确定该对象为目标对象，提高目标对象识别的精准度。
步骤104、基于目标对象进行相应操作。
无人机识别该对象为目标对象之后，可以基于目标对象进入手势控制模式、自拍模式或者目标跟踪模式。
本实施例，在无人机起飞的过程中，获取目标对象的身份信息，在无人机返航的过程中，通过图像采集设备采集包含对象的第一图像，根据第一图像和身份信息，识别该对象为目标对象，并基于目标对象进行相应操作，无需用户重新指定目标对象，可在返航过程中自动识别目标对象，提升操作的便利性。
图4是本发明实施例提供的另一种返航控制方法的流程图，如图4所示，该方法可以包括如下步骤：
步骤401、在无人机起飞的过程中，获取目标对象的身份信息。
在一些实施例中，无人机从预设位置起飞至起飞点的过程中，可以通过图像采集设备采集包含目标对象的图像，并对该图像进行处理，得到目标对象的身份信息。其中，身份信息可以包括目标对象的人体特征和/或人脸特征。预设位置可以为地面或者目标对象的手上或者桌面等位置。
以图2所述的无人机起飞的场景示意图为例，若无人机在起飞过程中，图像采集设备朝向目标对象，则无人机可以在飞行至起飞点之前，通过图像采集设备采集包含目标对象的图像。若无人机在起飞过程中，图像采集设备未朝向目标对象，则无人机可以在飞行至起飞点之后，通过云台转动，控制图像采集设备朝向目标对象，并通过图像采集设备采集包含目标对象的图像。
在一些实施例中，在无人机起飞的过程中，无人机可以接收地面站发送的包含目标对象的第二图像，对第二图像进行处理，得到目标对象的身份信息。
步骤402、获取无人机在起飞过程中的起飞位置。
无人机在起飞的过程中，可以获取起飞位置的地理坐标和高度。例如，如果无人机基于地面站发送的地理坐标和高度飞行至起飞点，则无人机可以将地面站发送的地理坐标和高度作为无人机在起飞过程中的起飞位置。又如，如果无人机基于用户对遥控器的摇杆操作飞行至起飞点，则无人机可以通过预设定位技术获取无人机在起飞过程中的起飞位置。预设定位技术可以包括全球定位***(Global Positioning System，GPS)技术、基站通信定位技术或者通过气压传感器获取无人机飞行至起飞点的高度等。
步骤403、在无人机返航的过程中，控制无人机返航至起飞位置。
在本发明实施例中，无人机在接收到地面站发送的返航指令或者信号丢失等情况下会进行返航操作，即从当前位置飞行至起飞位置。以图5所述的无人机返航的场景示意图为例，无人机可飞行至起飞点之后，下降一定高度，以便可通过图像采集设备采集到包含对象的人脸图像。
步骤404、通过图像采集设备采集包含对象的第一图像。
本发明实施例中，无人机返航至起飞位置之后，可以通过图像采集设备采集包含对象的第一图像。包含对象的第一图像可以包括人体图像和/或人脸图像。
例如，无人机在返航过程中，可以由当前位置飞行至起飞点，通过图像采集设备采集包含对象的人体图像，对人体图像进行处理，得到对象的人体特征，如果对象的人体特征和目标对象的人体特征匹配，无人机可以将对象识别为目标对象。
又如，无人机在返航过程中，可以由当前位置飞行至起飞点，调整图像采集设备的焦距，以便通过图像采集设备采集包含对象的人脸图像，对人脸图像进行处理，得到对象的人脸特征，如果对象的人脸特征和目标对象的人脸特征匹配，无人机可以将对象识别为目标对象。
又如，无人机在返航过程中，可以由当前位置飞行至起飞点，通过图像采集设备采集包含对象的人体图像，对该人体图像进行处理，得到对象的人体特征，如果对象的人体特征和目标对象的人体特征匹配，无人机可以降落至一定高度，以便可以通过图像采集设备采集到该对象的人脸图像，对该人脸图像进行处理，得到对象的人脸特征，如果对象的人脸特征和目标对象的人脸特征匹 配，无人机可以将对象识别为目标对象。
又如，无人机在返航过程中，可以由当前位置飞行至起飞点，通过图像采集设备采集包含对象的人体图像，对该人体图像进行处理，得到对象的人体特征，如果对象的人体特征和目标对象的人体特征匹配，无人机可以调整图像采集设备的焦距，以便可以通过图像采集设备采集到该对象的人脸图像，对该人脸图像进行处理，得到对象的人脸特征，如果对象的人脸特征和目标对象的人脸特征匹配，无人机可以将对象识别为目标对象。
需要说明的是，无人机从当前位置飞行至起飞点的飞行轨迹和无人机从起飞点飞行至当前位置的飞行轨迹可以相同，也可以不相同，具体不受本发明实施例的限制。
步骤405、根据第一图像和身份信息，识别对象为目标对象。
本发明实施例中，无人机可以对第一图像进行处理，得到对象的身份信息，若对象的身份信息和目标对象的身份信息匹配，则识别对象为目标对象。其中，对象的身份信息包括对象的人体特征和/或人脸特征。
在一些实施例中，如果对象的身份信息包括对象的人体特征，无人机可以通过行ReID技术判断对象的人体特征和目标对象的人体特征是否匹配，如果对象的人体特征和目标对象的人体特征匹配，无人机可以识别对象为目标对象。例如，无人机可以将对象的人体特征和目标对象的人体特征进行比较得到距离值，如果该距离值小于第一预设距离阈值，无人机可以确定对象的人体特征和目标对象的人体特征匹配。
在一些实施例中，如果对象的身份信息包括对象的人脸特征，无人机可以通过人脸识别技术判断对象的人脸特征和目标对象的人脸特征是否匹配，如果对象的人脸特征和目标对象的人脸特征匹配，无人机可以识别对象为目标对象。例如，无人机可以将对象的人脸特征和目标对象的人脸特征进行比较得到距离值，如果该距离值小于第二预设距离阈值，无人机可以确定对象的人脸特征和目标对象的人脸特征匹配。
在一些实施例中，如果对象的身份信息包括对象的人体特征和人脸特征，无人机可以判断对象的人体特征和目标对象的人体特征是否匹配，若对象的人体特征和目标对象的人体特征匹配，则判断对象的人脸特征和目标对象的人脸特征是否匹配，若对象的人脸特征和目标对象的人脸特征匹配，则识别对象为 目标对象。本发明实施例中，对象的人体特征和目标对象的人体特征匹配，无人机可以确定对象和目标对象的相似度较高，进一步的，对象的人脸特征和目标对象的人脸特征匹配，无人机可以确定该对象为目标对象，提高目标对象识别的精准度。
步骤406、基于目标对象进行相应操作。
无人机识别该对象为目标对象之后，可以基于目标对象进入手势控制模式、自拍模式或者目标跟踪模式。
本实施例，在无人机起飞的过程中，获取目标对象的身份信息以及无人机在起飞过程中的起飞位置，在无人机返航的过程中，控制无人机返航至起飞位置，通过图像采集设备采集包含对象的第一图像，根据第一图像和身份信息，识别该对象为目标对象，并基于目标对象进行相应操作，无需用户重新指定目标对象，可在返航过程中自动识别目标对象，提升操作的便利性。
本发明实施例提供一种返航控制装置，图6是本发明实施例提供的返航控制装置的结构图，如图6所示，返航控制装置600包括存储器601和处理器602，其中，存储器602中存储有程序代码，处理器602调用存储器中的程序代码，当程序代码被执行时，处理器602执行如下操作：在无人机起飞的过程中，获取目标对象的身份信息；在无人机返航的过程中，通过图像采集设备采集包含对象的第一图像；根据第一图像和身份信息，识别对象为目标对象；基于目标对象进行相应操作。
可选的，所述处理器602在根据所述第一图像和所述身份信息，识别所述对象为所述目标对象时，执行如下操作：
对所述第一图像进行处理，得到所述对象的身份信息，所述对象的身份信息包括所述对象的人体特征和/或人脸特征；
若所述对象的身份信息和所述目标对象的身份信息匹配，则识别所述对象为所述目标对象。
可选的，所述目标对象的身份信息包括所述目标对象的人体特征和人脸特征；
所述处理器602在所述对象的身份信息和所述目标对象的身份信息匹配，则识别所述对象为所述目标对象时，执行如下操作：
若所述对象的人体特征和所述目标对象的人体特征匹配，则判断所述对象的人脸特征和所述目标对象的人脸特征是否匹配；
若所述对象的人脸特征和所述目标对象的人脸特征匹配，则识别所述对象为所述目标对象。
可选的，所述处理器602在所述无人机返航的过程中，通过图像采集设备采集包含对象的第一图像时，执行如下操作：
获取所述无人机在起飞过程中的起飞位置；
在所述无人机返航的过程中，控制所述无人机返航至所述起飞位置；
通过所述图像采集设备采集所述第一图像。
可选的，所述处理器602在所述无人机返航的过程中，通过图像采集设备采集包含对象的第一图像时，执行如下操作：
在所述无人机返航至起飞位置之前，每间隔预设时长通过所述图像采集设备采集所述包含对象的第一图像，直至识别到所述第一图像所包含对象为所述目标对象。
可选的，所述处理器602在无人机起飞的过程中，获取目标对象的身份信息时，执行如下操作：
在所述无人机起飞的过程中，通过图像采集设备采集包含所述目标对象的第二图像；
对所述第二图像进行处理，得到所述目标对象的身份信息。
可选的，所述处理器602在无人机起飞的过程中，获取目标对象的身份信息时，执行如下操作：
在所述无人机起飞的过程中，接收地面站发送的包含所述目标对象的第二图像；
对所述第二图像进行处理，得到所述目标对象的身份信息。
可选的，所述处理器602在基于所述目标对象进行相应操作时，执行如下操作：
基于所述目标对象进入手势控制模式、自拍模式或者目标跟踪模式。
本实施例提供的返航控制装置能够执行前述实施例提供的无人机的返航控制方法，其执行方式和有益效果类似，在这里不再赘述。
本发明实施例还提供一种无人机，该无人机包括：
机身；
动力***，安装在所述机身，用于为所述无人机提供动力；
图像采集设备，安装在所述机身，用于采集包含对象的第一图像；
以及上述实施例提供的返航控制装置。
可选的，该无人机还包括位置传感器；
所述位置传感器，安装在所述机身，用于获取所述无人机在起飞过程中的起飞位置。
可选的，所述无人机还包括：
通信设备，安装在所述机身，用于与地面站进行信息交互。
本实施例提供的移动平台其执行方式和有益效果与前述实施例提供的移动装置类似，在这里不再赘述。
在本发明所提供的几个实施例中，应该理解到，所揭露的装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个***，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本发明各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用硬件加软件功能单元的形式实现。
上述以软件功能单元的形式实现的集成的单元，可以存储在一个计算机可读取存储介质中。上述软件功能单元存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)或处 理器(processor)执行本发明各个实施例所述方法的部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
本领域技术人员可以清楚地了解到，为描述的方便和简洁，仅以上述各功能模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能模块完成，即将装置的内部结构划分成不同的功能模块，以完成以上描述的全部或者部分功能。上述描述的装置的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
最后应说明的是：以上各实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述各实施例对本发明进行了详细的说明，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发明各实施例技术方案的范围。

Claims (19)

  1. 一种返航控制方法，其特征在于，所述方法包括：
    在无人机起飞的过程中，获取目标对象的身份信息；
    在所述无人机返航的过程中，通过图像采集设备采集包含对象的第一图像；
    根据所述第一图像和所述身份信息，识别所述对象为所述目标对象；
    基于所述目标对象进行相应操作。
  2. 根据权利要求1所述的方法，其特征在于，所述根据所述第一图像和所述身份信息，识别所述对象为所述目标对象，包括：
    对所述第一图像进行处理，得到所述对象的身份信息，所述对象的身份信息包括所述对象的人体特征和/或人脸特征；
    若所述对象的身份信息和所述目标对象的身份信息匹配，则识别所述对象为所述目标对象。
  3. 根据权利要求2所述的方法，其特征在于，所述目标对象的身份信息包括所述目标对象的人体特征和人脸特征；
    所述若所述对象的身份信息和所述目标对象的身份信息匹配，则识别所述对象为所述目标对象，包括：
    若所述对象的人体特征和所述目标对象的人体特征匹配，则判断所述对象的人脸特征和所述目标对象的人脸特征是否匹配；
    若所述对象的人脸特征和所述目标对象的人脸特征匹配，则识别所述对象为所述目标对象。
  4. 根据权利要求1-3任一项所述的方法，其特征在于，所述在所述无人机返航的过程中，通过图像采集设备采集包含对象的第一图像，包括：
    获取所述无人机在起飞过程中的起飞位置；
    在所述无人机返航的过程中，控制所述无人机返航至所述起飞位置；
    通过所述图像采集设备采集所述第一图像。
  5. 根据权利要求1-3任一项所述的方法，其特征在于，所述在所述无人机返航的过程中，通过图像采集设备采集包含对象的第一图像，包括：
    在所述无人机返航至起飞位置之前，每间隔预设时长通过所述图像采集设备采集所述包含对象的第一图像，直至识别到所述第一图像所包含对象为所述目标对象。
  6. 根据权利要求1所述的方法，其特征在于，所述在无人机起飞的过程中，获取目标对象的身份信息，包括：
    在所述无人机起飞的过程中，通过图像采集设备采集包含所述目标对象的第二图像；
    对所述第二图像进行处理，得到所述目标对象的身份信息。
  7. 根据权利要求1所述的方法，其特征在于，所述在无人机起飞的过程中，获取目标对象的身份信息，包括：
    在所述无人机起飞的过程中，接收地面站发送的包含所述目标对象的第二图像；
    对所述第二图像进行处理，得到所述目标对象的身份信息。
  8. 根据权利要求1所述的方法，其特征在于，所述基于所述目标对象进行相应操作，包括：
    基于所述目标对象进入手势控制模式、自拍模式或者目标跟踪模式。
  9. 一种返航控制装置，其特征在于，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：
    在无人机起飞的过程中，获取目标对象的身份信息；
    在所述无人机返航的过程中，通过图像采集设备采集包含对象的第一图像；
    根据所述第一图像和所述身份信息，识别所述对象为所述目标对象；
    基于所述目标对象进行相应操作。
  10. 根权利要求9所述的装置，其特征在于，所述处理器在根据所述第一图像和所述身份信息，识别所述对象为所述目标对象时，执行如下操作：
    对所述第一图像进行处理，得到所述对象的身份信息，所述对象的身份信息包括所述对象的人体特征和/或人脸特征；
    若所述对象的身份信息和所述目标对象的身份信息匹配，则识别所述对象为所述目标对象。
  11. 根据权利要求10所述的装置，其特征在于，所述目标对象的身份信息包括所述目标对象的人体特征和人脸特征；
    所述处理器在所述对象的身份信息和所述目标对象的身份信息匹配，则识别所述对象为所述目标对象时，执行如下操作：
    若所述对象的人体特征和所述目标对象的人体特征匹配，则判断所述对象的人脸特征和所述目标对象的人脸特征是否匹配；
    若所述对象的人脸特征和所述目标对象的人脸特征匹配，则识别所述对象为所述目标对象。
  12. 根据权利要求9-11任一项所述的装置，其特征在于，所述处理器在所述无人机返航的过程中，通过图像采集设备采集包含对象的第一图像时，执行如下操作：
    获取所述无人机在起飞过程中的起飞位置；
    在所述无人机返航的过程中，控制所述无人机返航至所述起飞位置；
    通过所述图像采集设备采集所述第一图像。
  13. 根据权利要求9-11任一项所述的装置，其特征在于，所述处理器在所述无人机返航的过程中，通过图像采集设备采集包含对象的第一图像时，执行如下操作：
    在所述无人机返航至起飞位置之前，每间隔预设时长通过所述图像采集设备采集所述包含对象的第一图像，直至识别到所述第一图像所包含对象为所述 目标对象。
  14. 根据权利要求9所述的装置，其特征在于，所述处理器在无人机起飞的过程中，获取目标对象的身份信息时，执行如下操作：
    在所述无人机起飞的过程中，通过图像采集设备采集包含所述目标对象的第二图像；
    对所述第二图像进行处理，得到所述目标对象的身份信息。
  15. 根据权利要求9所述的装置，其特征在于，所述处理器在无人机起飞的过程中，获取目标对象的身份信息时，执行如下操作：
    在所述无人机起飞的过程中，接收地面站发送的包含所述目标对象的第二图像；
    对所述第二图像进行处理，得到所述目标对象的身份信息。
  16. 根据权利要求9所述的装置，其特征在于，所述处理器在基于所述目标对象进行相应操作时，执行如下操作：
    基于所述目标对象进入手势控制模式、自拍模式或者目标跟踪模式。
  17. 一种无人机，其特征在于，包括：
    机身；
    动力***，安装在所述机身，用于为所述无人机提供动力；
    图像采集设备，安装在所述机身，用于采集包含对象的第一图像；
    以及如权利要求9-16中任一项所述的返航控制装置。
  18. 根据权利要求17所述的无人机，其特征在于，所述无人机还包括位置传感器；
    所述位置传感器，安装在所述机身，用于获取所述无人机在起飞过程中的起飞位置。
  19. 根据权利要求17所述的无人机，其特征在于，所述无人机还包括：
    通信设备，安装在所述机身，用于与地面站进行信息交互。
PCT/CN2018/097776 2018-07-31 2018-07-31 返航控制方法、装置及设备 WO2020024104A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2018/097776 WO2020024104A1 (zh) 2018-07-31 2018-07-31 返航控制方法、装置及设备
CN201880010540.4A CN110291482A (zh) 2018-07-31 2018-07-31 返航控制方法、装置及设备

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2018/097776 WO2020024104A1 (zh) 2018-07-31 2018-07-31 返航控制方法、装置及设备

Publications (1)

Publication Number Publication Date
WO2020024104A1 true WO2020024104A1 (zh) 2020-02-06

Family

ID=68001271

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/097776 WO2020024104A1 (zh) 2018-07-31 2018-07-31 返航控制方法、装置及设备

Country Status (2)

Country Link
CN (1) CN110291482A (zh)
WO (1) WO2020024104A1 (zh)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110781821B (zh) * 2019-10-25 2022-11-01 上海商汤智能科技有限公司 基于无人机的目标检测方法及装置、电子设备和存储介质
CN111580546B (zh) * 2020-04-13 2023-06-06 深圳蚁石科技有限公司 无人机自动返航方法和装置
CN111629149A (zh) * 2020-06-09 2020-09-04 金陵科技学院 一种基于小型无人机的自拍装置及其控制方法
CN111813145B (zh) * 2020-06-30 2024-06-11 深圳市万翼数字技术有限公司 无人机巡航的控制方法及相关***
CN111679695B (zh) * 2020-08-11 2020-11-10 中航金城无人***有限公司 一种基于深度学习技术的无人机巡航及追踪***和方法
WO2023159611A1 (zh) * 2022-02-28 2023-08-31 深圳市大疆创新科技有限公司 图像拍摄方法、装置与可移动平台

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105974940A (zh) * 2016-04-29 2016-09-28 优利科技有限公司 适用于飞行器的目标跟踪方法
CN106292721A (zh) * 2016-09-29 2017-01-04 腾讯科技（深圳）有限公司 一种控制飞行器跟踪目标对象的方法、设备及***
CN106598071A (zh) * 2016-12-20 2017-04-26 北京小米移动软件有限公司 跟随式的飞行控制方法及装置、无人机
US20180149479A1 (en) * 2015-06-29 2018-05-31 Yuneec Technology Co., Limited Geo-location or navigation camera, and aircraft and navigation method therefor

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9442485B1 (en) * 2014-08-13 2016-09-13 Trace Live Network Inc. Pixel based image tracking system for unmanned aerial vehicle (UAV) action camera system
CN106020227B (zh) * 2016-08-12 2019-02-26 北京奇虎科技有限公司 无人机的控制方法、装置
KR20180051996A (ko) * 2016-11-09 2018-05-17 삼성전자주식회사 무인 비행 장치 및 이를 이용한 피사체 촬영 방법
CN110119154A (zh) * 2016-11-30 2019-08-13 深圳市大疆创新科技有限公司 飞行器的控制方法、装置和设备以及飞行器
CN106909172A (zh) * 2017-03-06 2017-06-30 重庆零度智控智能科技有限公司 环绕跟踪方法、装置和无人机
CN107749942A (zh) * 2017-09-13 2018-03-02 深圳天珑无线科技有限公司 悬浮拍摄方法、移动终端及计算机可读存储介质
CN107948583A (zh) * 2017-10-31 2018-04-20 易瓦特科技股份公司 基于无人机对目标对象进行标识的方法、***及装置
CN107863153A (zh) * 2017-11-24 2018-03-30 中南大学 一种基于智能大数据的人体健康特征建模测量方法与平台

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180149479A1 (en) * 2015-06-29 2018-05-31 Yuneec Technology Co., Limited Geo-location or navigation camera, and aircraft and navigation method therefor
CN105974940A (zh) * 2016-04-29 2016-09-28 优利科技有限公司 适用于飞行器的目标跟踪方法
CN106292721A (zh) * 2016-09-29 2017-01-04 腾讯科技（深圳）有限公司 一种控制飞行器跟踪目标对象的方法、设备及***
CN106598071A (zh) * 2016-12-20 2017-04-26 北京小米移动软件有限公司 跟随式的飞行控制方法及装置、无人机

Also Published As

Publication number Publication date
CN110291482A (zh) 2019-09-27

Similar Documents

Publication Publication Date Title
WO2020024104A1 (zh) 返航控制方法、装置及设备
US11914370B2 (en) System and method for providing easy-to-use release and auto-positioning for drone applications
US11566915B2 (en) Method, device and system for processing a flight task
US10674062B2 (en) Control method for photographing using unmanned aerial vehicle, photographing method using unmanned aerial vehicle, mobile terminal, and unmanned aerial vehicle
US20190369613A1 (en) Electronic device and method for controlling multiple drones
CN110494360B (zh) 用于提供自主摄影及摄像的***和方法
CN105425815B (zh) 一种利用无人飞行器的牧场智能管理***及方法
WO2021223124A1 (zh) 位置信息获取方法、设备及存储介质
WO2018001245A1 (en) Robot control using gestures
US20220270277A1 (en) Computing a point cloud from stitched images
CN110688914A (zh) 一种手势识别的方法、智能设备、存储介质和电子设备
EP4174716A1 (en) Pedestrian tracking method and device, and computer readable storage medium
US11509809B2 (en) Following control method, control terminal, and unmanned aerial vehicle
WO2019061111A1 (zh) 一种路径调整方法及无人机
CN110546682A (zh) 信息处理装置、航拍路径生成方法、航拍路径生成***、程序以及记录介质
JP2021526743A (ja) 撮像制御方法、装置及び制御装置、撮像装置
US9141191B2 (en) Capturing photos without a camera
US11945583B2 (en) Method for generating search information of unmanned aerial vehicle and unmanned aerial vehicle
US20220157032A1 (en) Multi-modality localization of users
Abate et al. Remote 3D face reconstruction by means of autonomous unmanned aerial vehicles
CN112631333B (zh) 无人机的目标跟踪方法、装置及图像处理芯片
WO2022082440A1 (zh) 确定目标跟随策略的方法、装置、***、设备及存储介质
EP4246463A1 (en) Multiscale object detection device and method
CN113449566A (zh) 人在回路的“低慢小”目标智能图像跟踪方法及***
CN115649501A (zh) 一种无人机夜行照明***及方法

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: 18928530

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: 18928530

Country of ref document: EP

Kind code of ref document: A1