CN113438399B - Target guidance system, method for unmanned aerial vehicle, and storage medium - Google Patents
Target guidance system, method for unmanned aerial vehicle, and storage medium Download PDFInfo
- Publication number
- CN113438399B CN113438399B CN202110710098.XA CN202110710098A CN113438399B CN 113438399 B CN113438399 B CN 113438399B CN 202110710098 A CN202110710098 A CN 202110710098A CN 113438399 B CN113438399 B CN 113438399B
- Authority
- CN
- China
- Prior art keywords
- wide
- angle
- camera
- optimal
- target
- 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.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/50—Constructional details
- H04N23/55—Optical parts specially adapted for electronic image sensors; Mounting thereof
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/66—Remote control of cameras or camera parts, e.g. by remote control devices
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/67—Focus control based on electronic image sensor signals
- H04N23/671—Focus control based on electronic image sensor signals in combination with active ranging signals, e.g. using light or sound signals emitted toward objects
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/695—Control of camera direction for changing a field of view, e.g. pan, tilt or based on tracking of objects
Abstract
The invention provides a target guidance system and method for an unmanned aerial vehicle, the unmanned aerial vehicle and a storage medium. The system includes a wide angle camera, a tele camera, and a processor. Wide angle camera and tele camera are with strapdown mode fixed mounting on unmanned aerial vehicle, and wide angle camera and tele camera are close to and the optical axis of the two is parallel to each other to the optical axis of wide angle camera and tele camera and the fuselage axis of ordinates of unmanned aerial vehicle keep fixed angle. The processor can search a plurality of suspected targets through appearance matching in a large range by detecting the shot image of the wide-angle camera in real time, complete similarity sorting and optimal target screening, and can identify optimal target details in a small range by detecting the shot image of the long-focus camera in real time to judge whether the targets are given targets. And in the case of confirming the given target, the unmanned aerial vehicle is guided to realize the tracking flight of the given target by continuously calculating the wide-angle guidance instruction. The invention has the advantages of simple structure, stability, reliability and convenient integration.
Description
Technical Field
Embodiments of the present invention generally relate to the field of drone technology, and more particularly, to a target guidance system for a drone, a method, a drone, and a storage medium.
Background
With the gradual maturity of unmanned aerial vehicle technology, unmanned aerial vehicles are increasingly applied to the recognition and shooting of targets.
Patent document CN111683204A discloses a method and an apparatus for shooting by unmanned aerial vehicle, a computer device and a storage medium, wherein a wide-angle camera and a telephoto camera are combined, the wide-angle camera is used to detect a target first, the shooting angle of the telephoto camera is adjusted by controlling and adjusting the rotation angle of a pan-tilt, and then the telephoto camera is used to shoot by zooming, so that the detection and the shooting are performed synchronously.
Patent document two CN111242988A discloses a method for tracking a target by a dual pan-tilt head with a wide-angle camera and a telephoto camera in a linkage manner, which includes: the wide-angle camera identifies a target and marks the target to be tracked; the long-focus camera is aligned with the wide-angle camera, so that the Z axis of a coordinate system where the long-focus camera is located is superposed with the optical center connecting line of the two cameras; rotating the tele camera to search for a target to be tracked; the tele camera tracks the target. Wherein, wide angle camera and tele camera install respectively on two different cloud platforms, and two cloud platforms all can the free rotation.
Patent document three CN110443247A discloses a system and a method for detecting a small ground motion target of an unmanned aerial vehicle in real time, the system includes a zoom camera and a micro-computing platform, the zoom camera is used for zooming according to a zoom instruction and shooting a target video image, and the micro-computing platform sends an attitude adjustment instruction to an unmanned aerial vehicle control system according to the position of the target in the image and sends the zoom instruction to the zoom camera at the same time.
Both of the above patent documents i and ii need to carry a pan-tilt on an unmanned aerial vehicle to control a camera shooting angle, however, the existing pan-tilt tracking system has a large volume and a heavy weight and a complex structure, and for a micro-miniature unmanned aerial vehicle which needs to realize tracking flight, the takeoff weight of the micro-miniature unmanned aerial vehicle is limited, so a light-weight visual guidance system is needed. Therefore, the pan-tilt is not suitable for micro unmanned aerial vehicle carrying application.
Above patent document three adjusts unmanned aerial vehicle's gesture through the micro-computing platform, however, patent document three adopts the camera that can zoom to realize the image shooting of different focus, and at unmanned aerial vehicle fast flight in-process, the distance between camera and the target lasts the rapid change. If a zoom camera is used, zooming and focusing needs to be done to get a sharp image. The zooming and focusing processes not only cause processing delay, but also may miss capturing of the target.
Disclosure of Invention
In order to solve the above problems in the prior art, in a first aspect, an embodiment of the present invention provides an object guidance system for a drone, wherein the system includes a wide-angle camera, a tele-camera and a processor, wherein the wide-angle camera and the tele-camera are fixedly mounted on the drone in a strapdown manner, the wide-angle camera and the tele-camera are in close proximity and have their optical axes parallel to each other, and the optical axes of the wide-angle camera and the tele-camera are held at a fixed angle to a longitudinal axis of a body of the drone. The wide-angle camera is used for acquiring wide-angle images with a large visual field range; the tele camera is used for acquiring a tele image in a small field range. The processor includes a wide guidance module and a tele confirmation module. The wide-angle guidance module is used for: acquiring and detecting the wide-angle image acquired by the wide-angle camera, and determining an optimal suspected target in the wide-angle image according to the overall characteristics of an object in the wide-angle image; calculating an unmanned aerial vehicle wide-angle guidance instruction according to the pixel coordinates of the optimal suspected target, and sending the unmanned aerial vehicle wide-angle guidance instruction to a flight control device of the unmanned aerial vehicle, so that the flight control device adjusts the flight attitude of the unmanned aerial vehicle according to the unmanned aerial vehicle wide-angle guidance instruction, and the optimal suspected target enters an image plane of the telephoto camera; and when the tele confirmation module confirms that the optimal suspected target is a given target, correcting the unmanned aerial vehicle wide-angle guidance instruction in real time according to the pixel coordinates of the optimal suspected target in the wide image, so that the optimal suspected target is kept in a specific area of an image plane of the wide camera and the unmanned aerial vehicle tracks the given target. The tele confirmation module is to: and acquiring the tele image acquired by the tele camera after the optimal suspected target enters an image plane of the tele camera, and judging whether the optimal suspected target is the given target according to the detail features of the optimal suspected target in the tele image.
In some embodiments, determining the optimal suspicious object in the wide-angle image comprises: detecting one or more objects in a current frame of the wide-angle image; calculating an overall similarity value of the one or more objects to the given target; determining an object of the one or more objects with an overall similarity value greater than or equal to an overall similarity threshold as a suspected target; sorting all the suspected targets determined in the wide-angle image in a descending order according to the overall similarity value; selecting one or more suspected targets ranked at the top from the suspected targets as the optimal suspected target; and detecting a subsequent frame of the wide-angle image upon determining that none of the one or more objects in the current frame are suspicious objects.
In some embodiments, calculating the drone wide-angle guidance instruction according to the pixel coordinates of the optimal suspected target includes: calculating the unmanned aerial vehicle wide-angle guidance instruction according to the difference value between the pixel coordinate of the optimal suspected target and the center coordinate of the specific area of the wide-angle image.
In some embodiments, calculating the drone wide-angle guidance instruction according to the pixel coordinates of the optimal suspected target includes: and calculating the unmanned aerial vehicle wide-angle guidance instruction according to the derivative or the numerical difference of the pixel coordinates of the optimal suspected target in the multi-frame images of the wide-angle image with respect to time.
In some embodiments, determining whether the optimal suspected target is the given target according to the detail features of the optimal suspected target in the tele-image comprises: calculating a detail similarity value between the detail features of the optimal suspected target in the tele image and the detail features of the given target; determining that the optimal suspected target is the given target if the detail similarity value is greater than or equal to a detail similarity threshold; determining that the optimal suspected target is not the given target if the detail similarity value is less than the detail similarity threshold.
In some embodiments, the wide angle navigation module is further configured to: when the tele confirmation module confirms that the optimal suspected target is not the given target, continuing to search for a new optimal suspected target from the newly acquired wide-angle image.
In some embodiments, the pixel coordinates of the optimal decoy are the coordinates of the center point of the pixel area occupied by the optimal decoy.
In a second aspect, the embodiments of the present invention propose a target guidance method for a drone, the target guidance method being applied to a target guidance system, the target guidance system including a wide-angle camera, a tele camera and a processor, wherein the wide-angle camera and the tele camera are fixedly mounted on the drone in a strapdown manner, the wide-angle camera and the tele camera are closely adjacent and have optical axes parallel to each other, and the optical axes of the wide-angle camera and the tele camera maintain a fixed angle with a longitudinal axis of a body of the drone; the wide-angle camera is used for acquiring wide-angle images with a large visual field range; the tele camera is used for acquiring a tele image in a small field range. The target guidance method comprises the following steps: acquiring and detecting the wide-angle image acquired by the wide-angle camera, and determining an optimal suspected target in the wide-angle image according to the overall characteristics of an object in the wide-angle image; calculating an unmanned aerial vehicle wide-angle guidance instruction according to the pixel coordinates of the optimal suspected target, and sending the unmanned aerial vehicle wide-angle guidance instruction to a flight control device of the unmanned aerial vehicle, so that the flight control device adjusts the flight attitude of the unmanned aerial vehicle according to the unmanned aerial vehicle wide-angle guidance instruction, and the optimal suspected target enters an image plane of the telephoto camera; acquiring the tele image acquired by the tele camera after the optimal suspected target enters an image plane of the tele camera, and judging whether the optimal suspected target is the given target according to the detail features of the optimal suspected target in the tele image; and when the optimal suspected target is confirmed to be a given target, correcting the unmanned aerial vehicle wide-angle guidance instruction in real time according to the pixel coordinates of the optimal suspected target in the wide-angle image, so that the optimal suspected target is kept in a specific area of an image plane of the wide-angle camera and the unmanned aerial vehicle tracks the given target.
In a third aspect, embodiments of the present invention provide a drone comprising a target guidance system for a drone as described in any of the embodiments above.
In a fourth aspect, embodiments of the present invention propose a computer-readable storage medium having stored thereon instructions that, when executed by a processor, implement the method described in the above embodiments.
The embodiment of the invention provides a target guiding system and method based on a parallel optical axis wide-long-focus camera, a unmanned aerial vehicle comprising the system and a storage medium, which are used for realizing the search, guidance and tracking of a given target by the unmanned aerial vehicle. Through wide angle camera and the long focus camera of parallel optical axis of installing with the strapdown mode on unmanned aerial vehicle, utilize small unmanned aerial vehicle, especially micro unmanned aerial vehicle's high mobility, realize searching and tracking to the target through changing the unmanned aerial vehicle gesture. The processor can search a plurality of suspected targets through appearance matching in a large range by detecting the shot images of the wide-angle camera in real time, and completes similarity sorting and optimal target screening. The processor can identify the optimal target details in a small range and judge whether the target is a given target by detecting the shot image of the telephoto camera in real time. And in the case of confirming the given target, the unmanned aerial vehicle is guided to realize the tracking flight of the given target by continuously calculating the wide-angle guidance instruction.
The embodiment of the invention provides a scheme for searching, guiding and tracking a given target by an unmanned aerial vehicle, in particular a micro unmanned aerial vehicle, can realize continuous tracking flight guidance of the given target by the unmanned aerial vehicle, and can be applied to the technical fields of aerospace, unmanned aerial vehicles and the like.
According to the embodiment of the invention, the wide-angle camera and the telephoto camera are directly and fixedly installed on the unmanned aerial vehicle, and a tripod head is not needed, so that the unmanned aerial vehicle has the following technical effects:
(1) the total weight of the unmanned aerial vehicle is reduced, and the unmanned aerial vehicle is particularly suitable for micro unmanned aerial vehicles. The weight of a traditional unmanned aerial vehicle carrying a tripod head is usually more than 200g, and the system and the method provided by the embodiment of the invention can control the total weight of the miniature unmanned aerial vehicle to be in the order of tens of g, for example, below 50 g; however, the technical solution proposed by the embodiment of the present invention is not only suitable for a micro-unmanned aerial vehicle, but also can be applied to any scene that needs to reduce the weight of the unmanned aerial vehicle;
(2) the unmanned aerial vehicle is not provided with a cloud deck, so that parts are reduced, the system is simplified, and the overall reliability of the unmanned aerial vehicle is improved;
(3) if the cloud deck is carried, when the shooting angle is adjusted, the angle of the cloud deck needs to be measured and controlled, and the situation that the measurement of the rotation angle is inaccurate or the control is inaccurate is likely to occur, so that errors are caused;
(4) on the unmanned aerial vehicle carrying the cloud deck, if a target tracking function is to be realized, firstly, the attitude adjustment of the cloud deck is needed, then the attitude adjustment of the unmanned aerial vehicle is driven by the attitude adjustment of the cloud deck, the response time is long, the number of controlled components is large, and deviation and even error are easily caused due to inaccurate control of a certain component.
(5) In the process of tracking and shooting, because the camera is fixedly installed with the unmanned aerial vehicle, the unmanned aerial vehicle is also aimed at the target at the same time in the process of aiming at the target by the camera, so that the unmanned aerial vehicle can be used for quickly locking and continuously tracking the target, and the distance tracking or direct arrival at the target can be kept.
The embodiment of the invention adopts the combination of the wide-angle camera and the telephoto camera, and has the following technical effects: in the rapid flight process of the unmanned aerial vehicle, the distance between the camera and the target continuously and rapidly changes. If a conventional zoom camera is used, zooming and focusing needs to be done to obtain a sharp image. The zooming and focusing processes not only cause processing delay, but also may miss capturing of the target. The embodiment of the invention adopts a combination mode of two fixed-focus lenses, overcomes the defects brought by a zoom camera, can quickly focus, reduces the response time required by zooming and focusing, can quickly switch shooting with different focal lengths and different visual fields, and can even simultaneously shoot with different focal lengths and different visual fields, thereby being convenient for quickly capturing the target. In addition, the matching of the long focus and the wide angle realizes the searching of a plurality of targets in a short distance and a large range and the confirmation of a specific target in a long distance and a small range of a specific area.
Drawings
The above and other objects, features and advantages of embodiments of the present invention will become readily apparent from the following detailed description read in conjunction with the accompanying drawings. Several embodiments of the invention are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which:
fig. 1 shows a schematic view of a drone carrying a target guidance system according to an embodiment of the invention;
FIG. 2 illustrates a flow diagram of example operations of a target guidance system according to an embodiment of the present invention;
FIG. 3 illustrates a top view operational schematic of a target guidance system according to an embodiment of the present invention;
FIG. 4 illustrates a front view operational schematic of a target guidance system according to an embodiment of the present invention;
FIG. 5 shows a schematic view of wide-angle imaging of an object guidance system according to an embodiment of the invention;
FIG. 6 shows a tele imaging schematic of the target guidance system according to an embodiment of the invention.
In the drawings, the same or corresponding reference numerals indicate the same or corresponding parts.
Detailed Description
The principles and spirit of the present invention will be described with reference to a number of exemplary embodiments. It is understood that these embodiments are given solely for the purpose of enabling those skilled in the art to better understand and to practice the invention, and are not intended to limit the scope of the invention in any way.
In one aspect, embodiments of the present invention provide a target guidance system for a drone. Referring to fig. 1, a schematic diagram of a drone carrying a target guidance system according to an embodiment of the invention is shown. As shown in fig. 1, the object guidance system includes a wide-angle camera 1, a telephoto camera 2, and a processor 3. Wherein, wide angle camera 1 and tele camera 2 are with strapdown mode fixed mounting on unmanned aerial vehicle 4, in the example that fig. 1 shows, wide angle camera 1 and tele camera 2 are with strapdown mode side by side fixed mounting in 4 fuselage fronts of unmanned aerial vehicle. Wide angle camera 1 and tele camera 2 are in close proximity with their optical axes parallel to each other and the optical axes of wide angle camera 1 and tele camera 2 are held at a fixed angle to the longitudinal axis of the body of drone 4, e.g., the optical axes of wide angle camera 1 and tele camera 2 may be parallel to the longitudinal axis of the body of drone 4.
The wide angle camera 1 is installed next to the telephoto camera 2 and the optical axes of the wide angle camera 1 and the telephoto camera 2 are parallel, and this fixed installation may cause the telephoto camera 2 to be directed to a specific region of the view plane (or image plane) of the wide angle camera 1. The specific area may comprise the central area or any other area of the image plane of the wide angle camera 1, depending on the specific application.
Referring to fig. 5 and 6, wherein fig. 5 shows a schematic view of wide-angle imaging of an object guidance system according to an embodiment of the invention; FIG. 6 shows a tele imaging schematic of the target guidance system according to an embodiment of the invention.
The wide-angle camera 1 may be used to acquire wide-angle images of a large field of view. In the fixed mode shown in fig. 1, the wide-angle camera 1 can capture a wide-range image in front of the drone 4. As shown in fig. 5-6, the wide-range wide-angle image is shown as a dashed box in the image area of the wide-angle camera, which includes the image 51 of the subject 5 in the wide-angle camera, and the wide-angle image includes the overall features, or apparent optical features, of the subject. The wide-angle camera is used for shooting a large-range image, so that the large-range target searching and screening guidance in front of the unmanned aerial vehicle 4 can be realized conveniently.
The tele camera 2 may be used to acquire tele images of a small field of view. In the fixed mode shown in fig. 1, the tele camera 2 can be used to capture a small range image in front of the drone 4. As shown in fig. 5-6, the tele image of the small range includes an image 52 of the photographic subject 5 in the tele camera as shown by the solid box in the figure. The tele image includes detailed optical features of the object. The long-focus camera is used for shooting the small-range image, so that the target in the small range in front of the unmanned aerial vehicle 4 can be confirmed conveniently.
In the example shown in the figure, the processor 3 is mounted on the drone, however, the processor 3 may also be mounted elsewhere, for example on a remote control or a third terminal, communicating with the wide angle camera 1, the tele camera 2 and the flight control devices on the drone 4 through wireless transmission means.
The processor 3 may be, for example, a vision computer for calculating and sending wide-angle guidance instructions to the flight control device of the drone 4 in real time, so as to guide the tracked flight of the drone 4 to a given target. The processor 3 may internally run a vision guidance algorithm to calculate guidance instructions by detecting captured images of the wide camera 1 and the tele camera 2 in real time.
The processor 3 may include a wide angle navigation module and a tele confirmation module. The vision guidance algorithm may include a pantoscopic guidance algorithm and a tele-confirmation algorithm, operating in the pantoscopic guidance module and the tele-confirmation module, respectively.
The wide-angle guidance module is used for: acquiring and detecting a wide-angle image acquired by a wide-angle camera 1, and determining an optimal suspected target in the wide-angle image according to the overall characteristics of an object in the wide-angle image; calculating an unmanned aerial vehicle wide-angle guidance instruction according to the pixel coordinates of the optimal suspected target, and sending the unmanned aerial vehicle wide-angle guidance instruction to a flight control device of the unmanned aerial vehicle 4, so that the flight control device adjusts the flight attitude of the unmanned aerial vehicle 4 according to the unmanned aerial vehicle wide-angle guidance instruction, and the optimal suspected target enters an image plane of the telephoto camera 1; and, when the tele-confirmation module confirms that the optimal suspicious object is the given target, correcting the drone wide-angle guidance instruction in real time according to the pixel coordinates of the optimal suspicious object in the wide-angle image so that the optimal suspicious object remains in the image plane specific area of the wide-angle camera 1 and the drone 4 tracks the given target.
The tele confirmation module is to: and acquiring a tele image acquired by the tele camera 1 after the optimal suspected target enters an image plane of the tele camera 1, and judging whether the optimal suspected target is a given target according to the detail characteristics of the optimal suspected target in the tele image.
Referring to fig. 2, a flow chart of an example operation of a target guidance system according to an embodiment of the present invention is shown.
As an embodiment of the present invention, determining an optimal suspected target in a wide-angle image includes the following steps: detecting or identifying one or more objects in a current frame of the wide-angle image (shown in fig. 2 as wide-angle camera identification); calculating an overall similarity value of one or more objects to a given target; determining an object with an overall similarity value greater than or equal to an overall similarity threshold value among the one or more objects as a suspected target, the target similarity being shown in fig. 2; sorting all suspected targets determined in the wide-angle image in a descending order according to the overall similarity value; selecting one or more suspected targets ranked in the top as optimal suspected targets (shown as optimal target screening in fig. 2) from the suspected targets; and detecting a subsequent frame of the wide-angle image when it is determined that none of the one or more objects in the current frame is a suspect object.
As shown in fig. 2, as an embodiment of the present invention, the wide-angle guidance module is further configured to: and when the tele confirmation module confirms that the optimal suspected target is not the given target, continuously searching for a new optimal suspected target according to the newly acquired wide-angle image.
As one embodiment of the present invention, the pixel coordinates of the optimal decoy are coordinates of a center point of a pixel area occupied by the optimal decoy. As shown in fig. 5, the pixel coordinates of the imaging of the subject in the wide-angle camera 1 are the center points of the rectangular pixel areas occupied by the vehicle. It is to be understood that the pixel area occupied by the object may be various shapes, such as a circle, a polygon, etc., in addition to a rectangle, and the present invention is not limited in this respect.
As an embodiment of the present invention, determining whether the optimal suspected target is a given target according to the detail features of the optimal suspected target in the tele-image includes: calculating the detail similarity value of the detail features of the optimal suspected target in the tele image and the detail features of the given target; if the detail similarity value is greater than or equal to the detail similarity threshold, determining that the optimal suspected target is the given target; if the detail similarity value is less than the detail similarity threshold, determining that the optimal suspected target is not the given target.
By way of example only, the specific operation of the wide guidance module and the tele confirmation module is described below.
Wide-angle navigation algorithm searches for all objects T in real time in the image taken by the wide-angle camera 1i S1, 2.; the wide-angle guiding algorithm detects the images shot by the wide-angle camera 1 in real time and calculates the overall similarity value f of the appearance samples of the given targetW(Ti S) And is similar to the overall similarity threshold (or called recognition threshold)
In contrast, as shown in the following formula (1):
wherein, Ti SI 1, 2.. denotes all objects within the wide-angle image, fW(Ti S) Representing the overall similarity value of an object within the image to a given target,
indicating the overall similarity threshold.
All suspected targets
Along with its pixel coordinates are recorded in the form of the following equation (2):
wherein the content of the first and second substances,
representing all the suspected objects in the wide-angle image,
representing a suspected object
The pixel coordinates of (a).
After the wide-angle guiding algorithm processes the current frame picture, the similarity data of all suspected targets are sorted, and the optimal target is screened out according to the size of the similarity data
As shown in the following equation (3):
wherein the content of the first and second substances,
the optimal suspected object is represented by the number of pixels,
the pixel coordinates representing the optimal suspected object,
representing the overall similarity value of the suspected object to the given object.
The above formula (3) gives an example of using the suspected target with the largest overall similarity value as the optimal suspected target, however, in practical applications, according to specific scenes and requirements, a plurality of suspected targets with higher overall similarity values may also be selected as the optimal suspected targets.
Wide-angle navigation algorithm based on optimal target
Pixel coordinates of
And calculating a wide-angle guidance instruction and sending the wide-angle guidance instruction to a flight control device of the unmanned aerial vehicle 4.
Referring to fig. 3 and 4, wherein fig. 3 shows a top view operational schematic of a target guidance system according to an embodiment of the present invention; fig. 4 shows a front view operational schematic of a target guidance system according to an embodiment of the invention. In the camera views of fig. 3 and 4, the dotted line indicates the view of the wide-angle camera, and the solid line indicates the view of the telephoto camera.
As an embodiment of the present invention, calculating a wide-angle guidance instruction of an unmanned aerial vehicle according to pixel coordinates of an optimal suspected target includes: and calculating the wide-angle guidance instruction of the unmanned aerial vehicle according to the difference value of the pixel coordinates of the optimal suspected target and the center coordinates of the specific area of the wide-angle image.
Alternatively, the center of a specific area of the wide-angle image may be used as the origin of coordinates. In the case where the specific area is the image center area, the center of the wide-angle image may be taken as the origin of coordinates. In this case, the pixel coordinates may be determined
Proportionally calculating a guidance instruction as shown in the following formula (4):
wherein, uQWIndicating a pitch guidance command, uRWA steering guidance instruction is indicated and,
pixel coordinates, K, representing the optimal suspected objectQRepresenting the pitch guide coefficient, KRIndicating the steering guidance factor.
As another embodiment of the present invention, calculating the wide-angle guidance instruction of the drone according to the pixel coordinates of the optimal suspected target includes: and calculating the unmanned aerial vehicle wide-angle guiding instruction according to the derivative or numerical difference of the pixel coordinates of the optimal suspected target in the multi-frame images of the wide-angle image with respect to time.
Optionally, in terms of pixel coordinates
The derivative of time t proportionally calculates the pilot command as shown in equation (5) below:
wherein, uQWIndicating a pitch guidance command, uRWA steering guidance instruction is indicated and,
pixel coordinates, K, representing the optimal suspected objectQRepresenting the pitch guide coefficient, KRIndicating the steering guidance factor.
The calculation manner of the angular velocity guidance instruction (including the pitch guidance instruction and the steering guidance instruction) in the wide-angle guidance instruction of the drone is exemplified above, however, it is to be noted that the wide-angle guidance instruction of the drone may include an attitude instruction and a velocity instruction in addition to the angular velocity guidance instruction.
As shown in fig. 3 to 4, in the illustrated scene, a right turn guidance instruction and a head lowering guidance instruction are calculated according to the pixel coordinates of the optimal suspected object 5 in the wide-angle image, so that the posture adjustment of the unmanned aerial vehicle 4 drives the wide-angle camera 1 and the telephoto camera 2 to be aligned to the optimal suspected object 5 at the same time.
The wide-angle guidance algorithm continuously detects and processes the captured images of the wide-angle camera 1, and continuously calculates the wide-angle guidance instruction.
The unmanned plane 4 guides the command (uQ) according to the wide angleW,uRW) Make pitching and steering actions change attitude and then steer to the optimal target
Unmanned aerial vehicle 4 is turning to optimum target
Meanwhile, the wide-angle camera 1 and the long-focus camera 2 synchronously change the postures along with the unmanned aerial vehicle 4 until the optimal target
Into the view plane of the tele camera 2.
A tele confirmation algorithm for confirming in real time an optimal target within the image taken by the tele camera 2
Whether it is a given target 5. The long-focus confirmation algorithm detects the images shot by the long-focus camera 2 in real time and calculates the optimal suspected target
With a given target TTDetail similarity values in detail features
And is similar to the threshold of detail similarity (or confirmation threshold)
In contrast, as shown in the following equation (6):
wherein the content of the first and second substances,
the optimal suspected object is represented by the number of pixels,
representing the detail similarity value of the optimal suspected object to the given object,
a detail similarity threshold is indicated.
Thereby identifying the optimal suspected target
Whether or not it is a given target TTWherein:
if it is optimalSuspected target
Not given a target TTThen the tele-confirmation algorithm gives a negative decision that the target in front of the drone 4 is not the given target 5 and the wide-angle-guidance algorithm will continue to search for the target.
If the optimal target
Is given a target TTIf the target in front of the unmanned aerial vehicle 4 is the given target 5, the tele confirmation algorithm gives a positive decision that the target is the given target 5, the wide-angle guidance algorithm locks the existing optimal target, and continuously calculates the wide-angle guidance instruction to correct the flight trajectory of the unmanned aerial vehicle 4, so as to ensure that the given target 5 is always locked in a specific area of a wide-angle camera image plane, and realize the continuous tracking flight of the unmanned aerial vehicle 4 to the given target 5.
In the example shown in fig. 6, the tele-camera is aimed at the detail feature part of the target by scanning the vehicle, i.e. the vehicle has its number on its nose feature area, whereby this vehicle can be locked by the number without being confused with other vehicles of the same vehicle type. It should be noted that this is merely an example, and in fact, it may be possible that the telephoto camera may already recognize the detailed feature of the vehicle when the pose of the drone is adjusted such that the center of the vehicle enters a specific area of the wide-angle image, and then no additional scanning of the detailed feature is required.
The embodiment of the invention provides a parallel optical axis wide-angle and long-focus camera-based strapdown target guidance scheme, which is used for realizing the search, guidance and tracking of an unmanned aerial vehicle 4 on a given target 5. Through wide angle camera 1 and the long focus camera 2 with parallel optical axis of strapdown mode installation on unmanned aerial vehicle 4, synthesized wide angle camera 1 and searched for on a large scale and the long-range advantage of comparing of long focus camera 2, utilized small-size unmanned aerial vehicle 4, especially the high mobility of miniature unmanned aerial vehicle 4, realized the search and the tracking to the target through changing 4 gestures of unmanned aerial vehicle. The processor can search a plurality of suspected targets through appearance matching in a large range by detecting the shot image of the wide-angle camera 1 in real time, and complete similarity ranking and optimal target screening. The processor can determine whether or not the target is a given target 5 by detecting the captured image of the telephoto camera 2 in real time and recognizing the optimum target detail in a small distance range. In the case of confirming a given target, the drone 4 is guided to achieve a follow-up flight to the given target 5 by continuously calculating wide-angle guidance instructions.
The invention relates to a visual guidance system and a visual guidance method which can be applied to a micro unmanned aerial vehicle 4. The advantages of large-scale search of the wide-angle camera 1 and long-distance recognition of the long-focus camera 2 can be integrated by installing the wide-angle camera and the long-focus camera which are parallel to the optical axis on the unmanned aerial vehicle 4 in a strapdown mode; the method mainly comprises the following steps: a vision computer capable of searching for a suspected target through appearance matching in a large field of view by detecting a shot image of the wide-angle camera 1 in real time; by detecting the shot image of the telephoto camera 2 in real time, a given target can be confirmed by detailed recognition within a small field of view; and guiding the unmanned aerial vehicle 4 to realize the tracking flight of the given target by continuously calculating the guiding instruction for the confirmed target. The advantages of the invention include: simple structure, reliable and stable, easy to carry out, the integration of being convenient for are fit for unmanned aerial vehicle 4 especially 4 visual guide flights to given target of miniature unmanned aerial vehicle.
The embodiment of the invention also provides a target guiding method for the unmanned aerial vehicle, which is applied to a target guiding system, wherein the target guiding system comprises a wide-angle camera, a long-focus camera and a processor, the wide-angle camera and the long-focus camera are fixedly installed on the unmanned aerial vehicle in a strapdown mode, the wide-angle camera and the long-focus camera are closely adjacent and have optical axes parallel to each other, and the optical axes of the wide-angle camera and the long-focus camera keep a fixed angle with the longitudinal axis of the unmanned aerial vehicle body; the wide-angle camera is used for acquiring wide-angle images with a large visual field range; the tele camera is used for acquiring a tele image of a small field of view.
The target guiding method comprises the following steps:
acquiring and detecting a wide-angle image acquired by a wide-angle camera, and determining an optimal suspected target in the wide-angle image according to the overall characteristics of an object in the wide-angle image;
calculating an unmanned aerial vehicle wide-angle guidance instruction according to the pixel coordinates of the optimal suspected target, and sending the unmanned aerial vehicle wide-angle guidance instruction to a flight control device of the unmanned aerial vehicle, so that the flight control device adjusts the flight attitude of the unmanned aerial vehicle according to the unmanned aerial vehicle wide-angle guidance instruction, and the optimal suspected target enters an image plane of a telephoto camera;
acquiring a tele image acquired by a tele camera after the optimal suspected target enters an image plane of the tele camera, and judging whether the optimal suspected target is a given target according to the detail characteristics of the optimal suspected target in the tele image; and the number of the first and second groups,
and when the optimal suspected target is confirmed to be the given target, correcting the wide-angle guiding instruction of the unmanned aerial vehicle in real time according to the pixel coordinates of the optimal suspected target in the wide-angle image, so that the optimal suspected target is kept in the specific area of the image plane of the wide-angle camera and the unmanned aerial vehicle tracks the given target.
It should be noted that the target guidance method provided by the present invention corresponds to the functions of each module in the target guidance system, and for the specific embodiment and examples of the method, reference is made to the description of the target guidance system above, which is not repeated herein.
Embodiments of the present invention also provide a drone comprising a target guidance system for a drone as described in any of the embodiments.
Embodiments of the present invention also provide a computer-readable storage medium having stored thereon instructions which, when executed by a processor, implement the method described in the above embodiments.
The embodiment of the invention provides a parallel optical axis wide-angle and long-focus camera-based strapdown target guidance system, a parallel optical axis wide-angle and long-focus camera-based strapdown target guidance method, an unmanned aerial vehicle comprising the system and a storage medium, which are used for realizing the search, guidance and tracking of a given target by the unmanned aerial vehicle. Through wide angle camera and the long focus camera of parallel optical axis of installing with the strapdown mode on unmanned aerial vehicle, utilize small unmanned aerial vehicle, especially micro unmanned aerial vehicle's high mobility, realize searching and tracking to the target through changing the unmanned aerial vehicle gesture. The processor can search a plurality of suspected targets through appearance matching in a large range by detecting the shot images of the wide-angle camera in real time, and completes similarity sorting and optimal target screening. The processor can identify the optimal target details in a long distance and in a small range to judge whether the target is a given target by detecting the shot image of the telephoto camera in real time. And in the case of confirming the given target, the unmanned aerial vehicle is guided to realize the tracking flight of the given target by continuously calculating the wide-angle guidance instruction.
The embodiment of the invention provides a scheme for searching, guiding and tracking a given target by an unmanned aerial vehicle, in particular a micro unmanned aerial vehicle, can realize continuous tracking flight guidance of the given target by the unmanned aerial vehicle, and can be applied to the technical fields of aerospace, unmanned aerial vehicles and the like.
According to the embodiment of the invention, the wide-angle camera and the telephoto camera are directly and fixedly installed on the unmanned aerial vehicle, and a tripod head is not needed, so that the unmanned aerial vehicle has the following technical effects:
(1) the total weight of the unmanned aerial vehicle is reduced, and the unmanned aerial vehicle is particularly suitable for micro unmanned aerial vehicles. The weight of a traditional unmanned aerial vehicle carrying a tripod head is usually more than 200g, and the system and the method provided by the embodiment of the invention can control the total weight of the miniature unmanned aerial vehicle to be in the order of tens of g, for example, below 50 g; however, the technical solution proposed by the embodiment of the present invention is not only suitable for a micro-unmanned aerial vehicle, but also can be applied to any scene that needs to reduce the weight of the unmanned aerial vehicle;
(2) the unmanned aerial vehicle is not provided with a cloud deck, so that parts are reduced, the system is simplified, and the overall reliability of the unmanned aerial vehicle is improved;
(3) if the cloud deck is carried, when the shooting angle is adjusted, the angle of the cloud deck needs to be measured and controlled, and the situation that the measurement of the rotation angle is inaccurate or the control is inaccurate is likely to occur, so that errors are caused;
(4) on the unmanned aerial vehicle carrying the cloud deck, if a target tracking function is to be realized, the attitude adjustment of the cloud deck is needed firstly, and then the attitude adjustment of the unmanned aerial vehicle is driven by the attitude adjustment of the cloud deck, so that the response time is long, the number of controlled components is large, and deviation and even error are easily caused due to inaccurate control of a certain component.
(5) In the process of tracking and shooting, because the camera is fixedly installed with the unmanned aerial vehicle, the unmanned aerial vehicle is also aimed at the target at the same time in the process of aiming at the target by the camera, so that the unmanned aerial vehicle can track the target conveniently and can keep a distance tracking or directly reach the target.
The embodiment of the invention adopts the combination of the wide-angle camera and the telephoto camera, and has the following technical effects: in the rapid flight process of the unmanned aerial vehicle, the distance between the camera and the target continuously and rapidly changes. If a conventional zoom camera is used, zooming and focusing needs to be done to obtain a sharp image. The zooming and focusing processes not only cause processing delay, but also may miss capturing of the target. The embodiment of the invention adopts a combination mode of two fixed-focus lenses, overcomes the defects brought by a zoom camera, can quickly focus, reduces the response time required by zooming and focusing, can quickly switch shooting with different focal lengths and different visual fields, and can even simultaneously shoot with different focal lengths and different visual fields, thereby being convenient for quickly capturing the target. In addition, the matching of the long focus and the wide angle realizes the searching of the target in a short distance and a large range and the confirmation of the target in a long distance and a small range in a specific area.
The foregoing description of the embodiments of the invention has been presented for purposes of illustration and description, and is not intended to be exhaustive or to limit the invention to the precise forms disclosed. It will be understood by those skilled in the art that various changes may be made and equivalents may be substituted for elements thereof without departing from the scope of the invention. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the invention without departing from the essential scope thereof. Therefore, it is intended that the invention not be limited to the particular embodiment disclosed as the best mode contemplated for carrying out this invention, but that the invention will include all embodiments falling within the scope of the appended claims.
Claims (9)
1. An object guidance system for a drone, the system comprising a wide angle camera, a tele camera, and a processor, wherein the wide angle camera and the tele camera are fixedly mounted on the drone in a strapdown fashion, the wide angle camera and the tele camera are in close proximity and have their optical axes parallel to each other, and the optical axes of the wide angle camera and the tele camera maintain a fixed angle with a longitudinal axis of the body of the drone,
the wide-angle camera is used for acquiring wide-angle images with a large visual field range;
the long-focus camera is used for acquiring a long-focus image in a small visual field range;
the processor includes a wide angle guidance module and a tele confirmation module, wherein,
the wide-angle guidance module is used for: acquiring and detecting the wide-angle image acquired by the wide-angle camera, and determining an optimal suspected target in the wide-angle image according to the overall characteristics of an object in the wide-angle image; calculating an unmanned aerial vehicle wide-angle guidance instruction according to the pixel coordinates of the optimal suspected target, and sending the unmanned aerial vehicle wide-angle guidance instruction to a flight control device of the unmanned aerial vehicle, so that the flight control device adjusts the flight attitude of the unmanned aerial vehicle according to the unmanned aerial vehicle wide-angle guidance instruction, and the optimal suspected target enters an image plane of the telephoto camera; and, when the tele-confirmation module confirms that the optimal decoy is a given target, correcting the drone wide-angle guidance instruction in real-time according to the pixel coordinates of the optimal decoy in the wide-angle image, so that the optimal decoy remains at an image plane specific area of the wide-angle camera and the drone tracks the given target; when the tele confirmation module confirms that the optimal suspected target is not the given target, continuing to search for a new optimal suspected target according to the newly acquired wide-angle image;
the tele confirmation module is to: and acquiring the tele image acquired by the tele camera after the optimal suspected target enters an image plane of the tele camera, and judging whether the optimal suspected target is the given target according to the detail features of the optimal suspected target in the tele image.
2. The system of claim 1, wherein determining the optimal suspicious object in the wide-angle image comprises:
detecting one or more objects in a current frame of the wide-angle image;
calculating an overall similarity value of the one or more objects to the given target;
determining an object of the one or more objects with an overall similarity value greater than or equal to an overall similarity threshold as a suspected target;
sorting all the suspected targets determined in the wide-angle image in a descending order according to the overall similarity value;
selecting one or more suspected targets ranked at the top from the suspected targets as the optimal suspected target; and
detecting a subsequent frame of the wide-angle image upon determining that none of the one or more objects in the current frame are suspect objects.
3. The system of claim 1, wherein calculating the drone wide-angle guidance instruction based on the pixel coordinates of the optimal suspected target comprises:
calculating the unmanned aerial vehicle wide-angle guidance instruction according to the difference value between the pixel coordinate of the optimal suspected target and the center coordinate of the specific area of the wide-angle image.
4. The system of claim 1, wherein calculating the drone wide-angle guidance instruction based on the pixel coordinates of the optimal suspected target comprises:
and calculating the unmanned aerial vehicle wide-angle guidance instruction according to the derivative or the numerical difference of the pixel coordinates of the optimal suspected target in the multi-frame images of the wide-angle image with respect to time.
5. The system of claim 1, wherein determining whether the optimal suspected target is the given target according to the detail features of the optimal suspected target in the tele-image comprises:
calculating a detail similarity value between the detail features of the optimal suspected target in the tele image and the detail features of the given target;
determining that the optimal suspected target is the given target if the detail similarity value is greater than or equal to a detail similarity threshold;
determining that the optimal suspected target is not the given target if the detail similarity value is less than the detail similarity threshold.
6. The system of claim 1, wherein the pixel coordinates of the optimal decoy object are coordinates of a center point of a pixel area occupied by the optimal decoy object.
7. An object guidance method for a drone, wherein the object guidance method is applied to an object guidance system, the object guidance system comprises a wide-angle camera, a tele-focus camera and a processor, wherein the wide-angle camera and the tele-focus camera are fixedly mounted on the drone in a strapdown manner, the wide-angle camera and the tele-focus camera are closely adjacent and have optical axes parallel to each other, and the optical axes of the wide-angle camera and the tele-focus camera are kept at a fixed angle with a longitudinal axis of a body of the drone; the wide-angle camera is used for acquiring wide-angle images with a large visual field range; the long-focus camera is used for acquiring a long-focus image in a small visual field range;
the target guidance method comprises the following steps:
acquiring and detecting the wide-angle image acquired by the wide-angle camera, and determining an optimal suspected target in the wide-angle image according to the overall characteristics of an object in the wide-angle image;
calculating an unmanned aerial vehicle wide-angle guidance instruction according to the pixel coordinates of the optimal suspected target, and sending the unmanned aerial vehicle wide-angle guidance instruction to a flight control device of the unmanned aerial vehicle, so that the flight control device adjusts the flight attitude of the unmanned aerial vehicle according to the unmanned aerial vehicle wide-angle guidance instruction, and the optimal suspected target enters an image plane of the telephoto camera;
acquiring the tele image acquired by the tele camera after the optimal suspected target enters an image plane of the tele camera, and judging whether the optimal suspected target is a given target according to the detail features of the optimal suspected target in the tele image; and the number of the first and second groups,
when the optimal suspected target is confirmed to be a given target, correcting the unmanned aerial vehicle wide-angle guidance instruction in real time according to the pixel coordinates of the optimal suspected target in the wide-angle image, so that the optimal suspected target is kept in a specific area of an image plane of the wide-angle camera and the unmanned aerial vehicle tracks the given target;
when the optimal suspected target is confirmed not to be the given target, continuing to search for a new optimal suspected target according to the newly acquired wide-angle image.
8. A drone, characterized in that it comprises a target guidance system for a drone according to any one of claims 1 to 6.
9. A computer-readable storage medium having instructions stored thereon, which when executed by a processor, implement the method of claim 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110710098.XA CN113438399B (en) | 2021-06-25 | 2021-06-25 | Target guidance system, method for unmanned aerial vehicle, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110710098.XA CN113438399B (en) | 2021-06-25 | 2021-06-25 | Target guidance system, method for unmanned aerial vehicle, and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113438399A CN113438399A (en) | 2021-09-24 |
| CN113438399B true CN113438399B (en) | 2022-04-08 |
Family
ID=77754459
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110710098.XA Active CN113438399B (en) | 2021-06-25 | 2021-06-25 | Target guidance system, method for unmanned aerial vehicle, and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113438399B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116109956A (en) * | 2023-04-12 | 2023-05-12 | 安徽省空安信息技术有限公司 | Unmanned aerial vehicle self-adaptive zooming high-precision target detection intelligent inspection method |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09318382A (en) * | 1996-05-27 | 1997-12-12 | Nec Corp | Method and device for detecting posture angle |
| CN104820434A (en) * | 2015-03-24 | 2015-08-05 | 南京航空航天大学 | Velocity measuring method of ground motion object by use of unmanned plane |
| CN106707296A (en) * | 2017-01-09 | 2017-05-24 | 华中科技大学 | Dual-aperture photoelectric imaging system-based unmanned aerial vehicle detection and recognition method |
| CN108566513A (en) * | 2018-03-28 | 2018-09-21 | 深圳臻迪信息技术有限公司 | A kind of image pickup method of unmanned plane to moving target |
| CN110083174A (en) * | 2019-04-12 | 2019-08-02 | 上海歌尔泰克机器人有限公司 | Unmanned aerial vehicle (UAV) control method, apparatus and system |
| CN110443247A (en) * | 2019-08-22 | 2019-11-12 | 中国科学院国家空间科学中心 | A kind of unmanned aerial vehicle moving small target real-time detecting system and method |
| CN111242988A (en) * | 2020-01-14 | 2020-06-05 | 青岛联合创智科技有限公司 | Method for tracking target by using double pan-tilt coupled by wide-angle camera and long-focus camera |
| CN111683204A (en) * | 2020-06-18 | 2020-09-18 | 南方电网数字电网研究院有限公司 | Unmanned aerial vehicle shooting method and device, computer equipment and storage medium |
-
2021
- 2021-06-25 CN CN202110710098.XA patent/CN113438399B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09318382A (en) * | 1996-05-27 | 1997-12-12 | Nec Corp | Method and device for detecting posture angle |
| CN104820434A (en) * | 2015-03-24 | 2015-08-05 | 南京航空航天大学 | Velocity measuring method of ground motion object by use of unmanned plane |
| CN106707296A (en) * | 2017-01-09 | 2017-05-24 | 华中科技大学 | Dual-aperture photoelectric imaging system-based unmanned aerial vehicle detection and recognition method |
| CN108566513A (en) * | 2018-03-28 | 2018-09-21 | 深圳臻迪信息技术有限公司 | A kind of image pickup method of unmanned plane to moving target |
| CN110083174A (en) * | 2019-04-12 | 2019-08-02 | 上海歌尔泰克机器人有限公司 | Unmanned aerial vehicle (UAV) control method, apparatus and system |
| CN110443247A (en) * | 2019-08-22 | 2019-11-12 | 中国科学院国家空间科学中心 | A kind of unmanned aerial vehicle moving small target real-time detecting system and method |
| CN111242988A (en) * | 2020-01-14 | 2020-06-05 | 青岛联合创智科技有限公司 | Method for tracking target by using double pan-tilt coupled by wide-angle camera and long-focus camera |
| CN111683204A (en) * | 2020-06-18 | 2020-09-18 | 南方电网数字电网研究院有限公司 | Unmanned aerial vehicle shooting method and device, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113438399A (en) | 2021-09-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20230078078A1 (en) | Camera ball turret having high bandwidth data transmission to external image processor | |
| KR101645722B1 (en) | Unmanned aerial vehicle having Automatic Tracking and Method of the same | |
| CN113038016B (en) | Unmanned aerial vehicle image acquisition method and unmanned aerial vehicle | |
| CN105744163B (en) | A kind of video camera and image capture method based on depth information tracking focusing | |
| CN101790048B (en) | Intelligent camera system and method | |
| CN104125372B (en) | Target photoelectric search and detection method | |
| KR101769601B1 (en) | Unmanned aerial vehicle having Automatic Tracking | |
| KR100842101B1 (en) | Automatic recovery method of uav using vision information | |
| CN105120146A (en) | Shooting device and shooting method using unmanned aerial vehicle to perform automatic locking of moving object | |
| CN105979147A (en) | Intelligent shooting method of unmanned aerial vehicle | |
| CN110764537B (en) | Automatic tripod head locking system and method based on motion estimation and visual tracking | |
| WO2004106856A9 (en) | Device and method of supporting stereo camera, device and method of detecting calibration, and stereo camera system | |
| CN106973221B (en) | Unmanned aerial vehicle camera shooting method and system based on aesthetic evaluation | |
| CN112699839B (en) | Automatic video target locking and tracking method under dynamic background | |
| CN113111715B (en) | Unmanned aerial vehicle target tracking and information acquisition system and method | |
| CN111242988A (en) | Method for tracking target by using double pan-tilt coupled by wide-angle camera and long-focus camera | |
| CN113438399B (en) | Target guidance system, method for unmanned aerial vehicle, and storage medium | |
| CN114281100B (en) | Unmanned aerial vehicle inspection system and method without hovering | |
| CN109584264B (en) | Unmanned aerial vehicle vision guiding aerial refueling method based on deep learning | |
| CN112422905B (en) | Image acquisition method, device, equipment and medium for power equipment | |
| KR102473142B1 (en) | Fast zoom and focusing apparatus predicting for high-speed tracking and predicting moving object for camera for providing a high-quality image seamlessly and method for high-speed tracking and predicting moving object for camera using the same | |
| TWI641265B (en) | Mobile target position tracking system | |
| CN104113733A (en) | Television search detection method for small slow low-altitude target | |
| JP6719738B2 (en) | Autonomous mobile shooting control system and autonomous mobile body | |
| WO2023159611A1 (en) | Image photographing method and device, and movable platform |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |