CN106527496B - Aerial target fast tracking method towards unmanned plane image sequence - Google Patents
Aerial target fast tracking method towards unmanned plane image sequence Download PDFInfo
- Publication number
- CN106527496B CN106527496B CN201710023941.0A CN201710023941A CN106527496B CN 106527496 B CN106527496 B CN 106527496B CN 201710023941 A CN201710023941 A CN 201710023941A CN 106527496 B CN106527496 B CN 106527496B
- Authority
- CN
- China
- Prior art keywords
- track
- collection
- image sequence
- unmanned plane
- moment
- 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.)
- Expired - Fee Related
Links
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/12—Target-seeking control
Abstract
The invention discloses a kind of aerial target fast tracking method towards unmanned plane image sequence, the steps include: to assume that present clock is T, then the observation collection of t moment can be expressed as Z (t), i ∈ { 1,2 ... T }, the observation of t moment is several, is expressed as Mk=| Z (t) |, data correlation process can be described as finding the splitting scheme of set Z to track collection Tr;Track collection is obtained from unmanned plane image sequence, chooses some specific time, and track collection is resolved into track;The selection target in frame A clusters profile using the MHT algorithm based on time slip-window, constructs target object;The present invention utilizes a series of images sequence of multiple hypotheis tracking (MHT) algorithm keeps track object (target), the vision content of unmanned plane image is described using SIFT feature, formula and time slip-window scheme are distributed based on multidimensional, MHT frame is constructed, tracking velocity is fast, tracking is accurate.
Description
Technical field
The present invention relates to air vehicle technique field, specifically a kind of sky towards unmanned plane image sequence
Middle target fast tracking method.
Background technique
Unmanned vehicle (UAV) is also referred to as unmanned plane, can realize autonomous flight under no driver conditions.Flight shape
State and course line then pass through Digiplex or computer program control.Unmanned plane supports using air power it in an atmosphere
Flight, have many advantages, such as wide viewing angle, have a high potential, with autonomous flight ability small-sized depopulated helicopter, be suitable for military affairs,
Civilian and scientific research field.The basic structure of unmanned plane is made of 6 modules: 1) unmanned body, and 2) navigation system, 3) flight control
System, 4) wireless information communication system, 5) earth station system and 6) task application system.
As the important component of aircraft, unmanned plane antenna has become the important means of spatial data extraction.Military
With civilian application field, unmanned plane is considered as the effective way for solving many challenges.Especially in recent years, unmanned plane
It cashes and protrudes in the Gulf War and Afghan War, by Successful utilization in military purposes.In past ten years, unmanned plane
It has been used in science and commercial field.Currently, all kinds of unmanned planes can be applied to microwave link, military monitoring, traffic administration,
The fields such as scientific research, atmospheric monitoring, geographical mapping, forest fire protection, emergency management and rescue.
The fast-moving target tracking of unmanned plane image sequence be it is extremely complex, this process need using low cost and
High performance technology, especially UAV system receive the video frame that nadir camera obtains, then by textural characteristics input picture
The process of sequence Target Tracking System.In addition, needing to correct rotation and the scalability information of frame during practical flight to improve
Unmanned plane instrument flight performance, tracking velocity are slow.
Summary of the invention
In view of the defects and deficiencies of the prior art, the present invention intends to provide one kind towards unmanned plane image
The aerial target fast tracking method of sequence.z(t)
To achieve the above object, the technical solution adopted by the present invention is that:
Aerial target fast tracking method towards unmanned plane image sequence, the steps include:
1, assume that present clock is T, then the observation collection of t moment can be expressed as Z (t), t ∈ { 1,2 ..T }, t moment
Observation it is several, be expressed as Mk=| Z (t) |, data correlation process can be described as finding set Z to the segmentation side of track collection Tr
Case;It is further assumed that Z is in [t0, t0+ T] in time range with track collection TrAssociated image sequence, and track collection TrAlways
Remain into t0+ T-1 the moment;
2, track collection is obtained from unmanned plane image sequence, chooses some specific time, and track collection is resolved into rail
Road;
3, the selection target in frame A clusters profile using the MHT algorithm based on time slip-window, constructs target object;
4, the SIFT descriptor of detected target object predicts the next frame where SIFT descriptor using multiple hypotheis tracking
B;
5, the posterior probability that Track association is maximized using the MHT algorithm based on time slip-window, track is classified;
6, the target in detection frame B, if it is detected continuing to track, otherwise tracking terminates.
7, all stable segment iteration are merged, until not more segments can be merged into only.
Further, the step of track is classified is:
1, delete Z in track collection TrAssociated image sequence;
2, building has more hypothesis trees of gate, and the image sequence being stored on T is then assigned to track collection TrOn;
3, multiple hypotheses are gathered into disjoint tree;
4, track collection is obtained using greedy random adaptive local searching algorithm
If 5,Some track concentrated is from TrMiddle extension, then being divided into " continuing to track class ";IfConcentrate some track not with TrIn other Orbital Overlaps, then being divided into " new-track class ";Remaining track is drawn
Assign to t0" terminating class " that+T-1 moment terminates.
The invention has the benefit that
1, a series of images sequence of multiple hypotheis tracking (MHT) algorithm keeps track object (target) is utilized.
2, the vision content of unmanned plane image is described using SIFT feature.
3, based on multidimensional distribution formula and time slip-window scheme, MHT frame is constructed.
4, tracking velocity is fast, tracking is accurate.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, right below in conjunction with specific embodiment
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are only to explain the present invention, and
It is not used in the restriction present invention.
Embodiment:
Aerial target fast tracking method towards unmanned plane image sequence is:
1. hypothesis present clock is T, then the observation collection of t moment can be expressed as Z (t), t ∈ { 1,2 ..T }, t moment
Observation it is several, be expressed as Mk=| Z (t) |.Therefore, data correlation process can be described as finding point of set Z to track collection Tr
Cut scheme.
It is further assumed that Z is in [t0, t0+ T] in time range with track collection TrAssociated image sequence, and track collection
TrT is remained into always0+ T-1 the moment.
2. obtaining track collection from unmanned plane image sequence, some specific time is chosen, track collection is resolved into rail
Road;
3. the selection target in frame A clusters profile using the MHT algorithm based on time slip-window, constructs target object;
4. the SIFT descriptor of detected target object predicts the next frame where SIFT descriptor using multiple hypotheis tracking
B;
5. maximizing the posterior probability of Track association using the MHT algorithm based on time slip-window;Track is classified,
Specific step is as follows:
5.1: delete Z in track collection TrAssociated image sequence;
5.2: constructing more hypothesis trees with gate, the image sequence being stored on T is then assigned to track collection TrOn;
5.3: multiple hypotheses are gathered into disjoint tree;
5.4: obtaining track collection using greedy random adaptive local searching algorithm
5.5: track classification:
If a)Some track concentrated is from TrMiddle extension, then being divided into " continuing to track class ".
If b)Concentrate some track not with TrIn other Orbital Overlaps, then being divided into " new-track
Class ".
C) remaining division of period orbit is to t0" terminating class " that+T-1 moment terminates.
6. the target in detection frame B, if it is detected continuing to track, otherwise tracking terminates.
7. all stable segment iteration are merged, until not more segments can be merged into only.
Present embodiment utilizes a series of images sequence of multiple hypotheis tracking (MHT) algorithm keeps track object (target),
The vision content of unmanned plane image is described using SIFT feature, based on multidimensional distribution formula and time slip-window scheme, structure
MHT frame is built, tracking velocity is fast, tracking is accurate.
It is obvious to a person skilled in the art that invention is not limited to the details of the above exemplary embodiments, Er Qie
In the case where without departing substantially from spirit or essential attributes of the invention, the present invention can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the present invention is by appended power
Benefit requires rather than above description limits, it is intended that all by what is fallen within the meaning and scope of the equivalent elements of the claims
Variation is included within the present invention.
In addition, it should be understood that although this specification is described in terms of embodiments, but not each embodiment is only wrapped
Containing an independent technical solution, this description of the specification is merely for the sake of clarity, and those skilled in the art should
It considers the specification as a whole, the technical solutions in the various embodiments may also be suitably combined, forms those skilled in the art
The other embodiments being understood that.
Claims (2)
1. the aerial target fast tracking method towards unmanned plane image sequence, it is characterised in that: the steps include:
(1), assume that present clock is T, then the observation set representations of t moment are Z (t), t ∈ { 1,2 ..T }, the observation of t moment
It is several, it is expressed as Mk=| Z (t) |, data correlation process description is the splitting scheme for finding set Z to track collection Tr;It is further false
If Z is in [t0, t0+ T] in time range with track collection TrAssociated image sequence, and track collection TrT is remained into always0+T-1
Moment;
(2), track collection is obtained from unmanned plane image sequence, chooses some specific time, and track collection is resolved into track;
(3), the selection target in frame A clusters profile using the MHT algorithm based on time slip-window, constructs target object;
(4), the SIFT descriptor of detected target object predicts the next frame B where SIFT descriptor using multiple hypotheis tracking;
(5), the posterior probability that Track association is maximized using the MHT algorithm based on time slip-window, track is classified;
(6), the target in detection frame B, if it is detected continuing to track, otherwise tracking terminates;
(7), all stable segment iteration are merged, until not more segments can be merged into only.
2. the aerial target fast tracking method according to claim 1 towards unmanned plane image sequence, feature
Be: the step of track is classified is:
(1), delete Z in track collection TrAssociated image sequence;
(2), building has more hypothesis trees of gate, and then the image sequence for being saved in the T moment is assigned on track collection Tr;
(3), more hypothesis trees are gathered into disjoint tree;
(4), track collection is obtained using greedy random adaptive local searching algorithm
(5) if,Some track concentrated is from TrMiddle extension, then being divided into " continuing to track class ";IfCollection
In some track not with TrIn other Orbital Overlaps, then being divided into " new-track class ";Remaining division of period orbit is to t0
" terminating class " that+T-1 moment terminates.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710023941.0A CN106527496B (en) | 2017-01-13 | 2017-01-13 | Aerial target fast tracking method towards unmanned plane image sequence |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710023941.0A CN106527496B (en) | 2017-01-13 | 2017-01-13 | Aerial target fast tracking method towards unmanned plane image sequence |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106527496A CN106527496A (en) | 2017-03-22 |
CN106527496B true CN106527496B (en) | 2019-07-02 |
Family
ID=58336721
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710023941.0A Expired - Fee Related CN106527496B (en) | 2017-01-13 | 2017-01-13 | Aerial target fast tracking method towards unmanned plane image sequence |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106527496B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109697393B (en) * | 2017-10-23 | 2021-11-30 | 北京京东尚科信息技术有限公司 | Person tracking method, person tracking device, electronic device, and computer-readable medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101354254A (en) * | 2008-09-08 | 2009-01-28 | 北京航空航天大学 | Method for tracking aircraft course |
CN101944234A (en) * | 2010-07-23 | 2011-01-12 | 中国科学院研究生院 | Multi-object tracking method and device driven by characteristic trace |
EP2778819A1 (en) * | 2013-03-12 | 2014-09-17 | Thomson Licensing | Method for shooting a film performance using an unmanned aerial vehicle |
CN104199460A (en) * | 2014-08-29 | 2014-12-10 | 河南科技大学 | Intelligent four-rotor aircraft reconnoitering system based on image recognition and tracking |
CN105652891A (en) * | 2016-03-02 | 2016-06-08 | 中山大学 | Unmanned gyroplane moving target autonomous tracking device and control method thereof |
CN106227239A (en) * | 2016-09-22 | 2016-12-14 | 安徽机电职业技术学院 | Four rotor flying robot target lock-ons based on machine vision follow the tracks of system |
-
2017
- 2017-01-13 CN CN201710023941.0A patent/CN106527496B/en not_active Expired - Fee Related
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101354254A (en) * | 2008-09-08 | 2009-01-28 | 北京航空航天大学 | Method for tracking aircraft course |
CN101944234A (en) * | 2010-07-23 | 2011-01-12 | 中国科学院研究生院 | Multi-object tracking method and device driven by characteristic trace |
EP2778819A1 (en) * | 2013-03-12 | 2014-09-17 | Thomson Licensing | Method for shooting a film performance using an unmanned aerial vehicle |
CN104199460A (en) * | 2014-08-29 | 2014-12-10 | 河南科技大学 | Intelligent four-rotor aircraft reconnoitering system based on image recognition and tracking |
CN105652891A (en) * | 2016-03-02 | 2016-06-08 | 中山大学 | Unmanned gyroplane moving target autonomous tracking device and control method thereof |
CN106227239A (en) * | 2016-09-22 | 2016-12-14 | 安徽机电职业技术学院 | Four rotor flying robot target lock-ons based on machine vision follow the tracks of system |
Non-Patent Citations (1)
Title |
---|
基于多假设目标跟踪算法;楼晓祥 等;《指挥信息系统与技术》;20121231;第3卷(第6期);第36-39、50页 |
Also Published As
Publication number | Publication date |
---|---|
CN106527496A (en) | 2017-03-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN208401845U (en) | For promoting the system of nolo flight tool network | |
US10789468B2 (en) | Context-based autonomous perception | |
CN113485441A (en) | Distribution network inspection method combining unmanned aerial vehicle high-precision positioning and visual tracking technology | |
CN113625774B (en) | Local map matching and end-to-end ranging multi-unmanned aerial vehicle co-location system and method | |
US10767996B2 (en) | System and methods for reducing the map search space requirements in a vision-inertial navigation system | |
CN104504748A (en) | Unmanned aerial vehicle oblique photographing infrared three-dimensional imaging system and modeling method thereof | |
CN108646770A (en) | A kind of UAV Flight Control method, apparatus and system | |
Warren et al. | Large scale monocular vision-only mapping from a fixed-wing sUAS | |
CN109668567A (en) | Polarized light orientation method under multi-cloud condition of unmanned aerial vehicle | |
CN113485450A (en) | Unmanned aerial vehicle keeps away barrier system based on computer vision | |
CN111831010A (en) | Unmanned aerial vehicle obstacle avoidance flight method based on digital space slice | |
CN106527496B (en) | Aerial target fast tracking method towards unmanned plane image sequence | |
Miličević et al. | From the early days of unmanned aerial vehicles (UAVS) to their integration into wireless networks | |
CN115811354B (en) | Aerial flight equipment positioning system based on satellite time service | |
Figueira et al. | MISSION-ORIENTED SENSOR ARRAYS AND UAVs–A CASE STUDY ON ENVIRONMENTAL MONITORING | |
Chen et al. | Learning transferable uav for forest visual perception | |
CN112532557B (en) | Attitude correction system and method based on OFDM communication | |
Khanykov et al. | The Application of the High-Speed Pixel Clustering Method in Combining Multi-Angle Images Obtained from Airborne Optical-Location Systems | |
Medrano et al. | UAS: IoT on-line sensors for power line inspection | |
CN108983820B (en) | Ecological protection red line area boundary exploration method and system | |
Dolph et al. | Adversarial Learning Improves Vision-Based Perception from Drones with Imbalanced Datasets | |
Basnayake et al. | The era of the 5G drone is ahead, are we ready? | |
CN116817892B (en) | Cloud integrated unmanned aerial vehicle route positioning method and system based on visual semantic map | |
Zhao et al. | Onboard smart surveillance for micro-uav swarm: an experimental study | |
Fang | Task-Oriented and Energy-Saving Multi-UAV Deployment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | 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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190702 Termination date: 20220113 |
|
CF01 | Termination of patent right due to non-payment of annual fee |