CN102043964A - Tracking algorithm and tracking system for taking-off and landing of aircraft based on tripod head and camera head - Google Patents
Tracking algorithm and tracking system for taking-off and landing of aircraft based on tripod head and camera head Download PDFInfo
- Publication number
- CN102043964A CN102043964A CN2010106146511A CN201010614651A CN102043964A CN 102043964 A CN102043964 A CN 102043964A CN 2010106146511 A CN2010106146511 A CN 2010106146511A CN 201010614651 A CN201010614651 A CN 201010614651A CN 102043964 A CN102043964 A CN 102043964A
- Authority
- CN
- China
- Prior art keywords
- submodule
- aircraft
- speed
- module
- cloud terrace
- 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.)
- Pending
Links
Images
Abstract
The invention belongs to the technical field of video monitoring, in particular to a tracking algorithm and tracking system for taking-off and landing of an aircraft based on a tripod head and a camera head. The monitoring camera head and the tripod head are mounted on a control tower, and analog signals sent by the monitoring camera head are converted into digital video signals to serve as input of the tracking system; and the tracking system comprises a target detection module, a tracking module and a tripod head control module. By utilizing the tracking algorithm and tracking system provided by the invention, the taking-off trajectory and landing trajectory of the aircraft can be tracked automatically. According to the invention, the accuracy and real-time property of monitoring of the aircraft with the fixed taking-off and landing mode in an airport can be improved, the consumption of human resources can be reduced, a variety of difficulties in object extraction and tracking in a large-scale high-speed moving background video can be overcome, the automatic intelligent tracking, recording and analysis can be realized, and the monitoring quality can be further improved.
Description
Technical field
The invention belongs to technical field of video monitoring, be specifically related to a kind of Video processing and application at the aircraft flight field that the landing pattern of fixedly taking off is arranged, and, relate in particular to aircraft landing track algorithm and tracker based on the The Cloud Terrace camera based on the cradle head control and the technology that follows the trail of the objective that the Video processing historical data is excavated.
Background technology
Along with world economy develops rapidly, aeronautical technology is advanced by leaps and bounds, between the various places people-to-people contacts more and more closer, Aeronautical Service and the transportation in people's daily life proportion increasing.For a long time, the intelligent information management of air transport and aviation safety are important topics of relevant all circles exploratory development always.
The monitoring of aircraft landing still rests on the level of manual record observation so far, and there are two deficiencies in this traditional mode: the one, and the real-time of observation work, validity and integrality have directly influenced the quality of airport observation data; The 2nd, the data criterion deviation of artificial observation gained is big, later stage arrangement Treatment Analysis process means complexity, and the information recycle value is not high.Along with number of vehicles constantly increases, and great aviation accident increases in the landing stage, and aircraft management and monitoring work is apparent more important, the technological means of need a kind ofly can to become more meticulous to the aircraft landing, robotization being followed the tracks of.
Algorithm of the present invention can solve the problems referred to above well.Analyze the airport situation by the real-time monitor video picture that the monitoring camera that is installed on the approach tower is taken, major advantage is that its signal comprises complete picture, content is abundanter, real-time information is provided, also understood by computing machine and related work personnel, for fast early warning, start emergency measures and intelligent recording monitor provides possible with post-processed, therefore improve the operability of reliability, accuracy and the subsequent treatment analysis of monitoring greatly, had greatly using value and prospect.
At in the landing process, the key that aircraft is followed the tracks of is to detect on the runway and aircraft in the air, the extraction that difficult point is foreground object in the background frame of light variation outdoor monitoring environment under and high-speed motion with separate, be difficult to set up sample.The present invention has proposed well to separate annual reporting law in this respect.
Summary of the invention
The objective of the invention is to overcome above-mentioned difficult point, propose a kind of intelligent takeoff and landing track algorithm and tracker based on the The Cloud Terrace camera.
Technical scheme of the present invention is as follows:
A kind of aircraft landing track algorithm and tracker based on the The Cloud Terrace camera, by being installed in monitoring camera and the The Cloud Terrace on the control tower, mould and the analog signal sent by monitoring camera are after being converted into digital video signal, as the input of tracker.This tracker comprises: module of target detection (1), tracking module (2) and cradle head control module (3); Wherein:
Described module of target detection (1) comprises that background modeling submodule (1.1), Analysis on Prospect submodule (1.2), target are extracted submodule (1.3) and color characteristic extracts submodule (1.4), is sent to tracking module (2) by the detected aircraft coordinate of module of target detection (1);
Described tracking module (2) comprises angle point feature extraction submodule (2.1), by optical flow tracking submodule (2.2), Kalman filtering submodule (2.3), Mean-shift submodule (2.4), decision-making submodule (2.5); Angle point feature extraction submodule (2.1) receives by the detected aircraft coordinate of module of target detection (1); By optical flow tracking submodule (2.2) output, obtain a predicted value by Kalman filtering submodule (2.3), Mean-shift submodule (2.4) draws another predicted value, is made adjustment by decision-making submodule (2.5); Tracking module (2) predicts that with aircraft coordinate is sent to cradle head control module (3);
Described cradle head control module (3) comprises that velocity estimation submodule (3.1) and instruction send submodule (3.2); Carry out the calculating and the analysis of aircraft speed by velocity estimation submodule (3.1) according to aircraft prediction coordinate, obtain corresponding The Cloud Terrace speed, send submodule (3.2) by instruction and implement corresponding The Cloud Terrace operation.
Among the present invention:
Described background modeling submodule (1.1), this module is learnt background by the two field picture of average taking-up, sets up mixed Gauss model and renewal.
Described Analysis on Prospect submodule (1.2), this module is carried out binary conversion treatment according to the background information that the background modeling submodule provides to the two field picture that takes out, promptly the difference pixel within the specific limits with the gray-scale value of background model is made as black, thereby has extracted needed foreground information with the bigger white that is made as of the background subtraction opposite sex.
Described target is extracted submodule (1.3), this module is partly checked the prospect of picture frame according to the binary image that the Analysis on Prospect submodule generates, when prospect triggers frame according to predesigned order by some appointed areas, sign has obtained target, produce trigger event, obtain the position range of aircraft in the prospect, the color characteristic that passes in the tracking module extracts submodule.
Described color characteristic extracts submodule (1.4), and this module is extracted the colouring information of input rectangle in original frame, by setting up color histogram, extracts main color, as the aircraft color information.
Described angle point feature extraction submodule (2.1) in the rectangular extent that triggers the submodule input, is sought unique point, and this unique point is an angle point, storage unique point coordinate.
Described optical flow tracking submodule (2.2) based on the light stream pyramid, carries out Feature Points Matching around the unique point of obtaining in former frame, if the unique point number that the match is successful is less than specifies number, then detected characteristics point again.Otherwise, with the median of the unique point obtained as output.
Described Kalman filtering submodule (2.3) is used for by aircraft movements track cylindricality barrier.Aircraft is by the cylindricality barrier time, and all unique points all can be blocked, and cause and lose.Kalman filter is that aircraft is set up motion model, if wave filter predicted value and actual measurement unique point coordinate figure greater than specifying deviation, detected characteristics point again in the specified scope around the revised coordinate of Kalman filter then.Otherwise, will revise coordinate and pass to decision-making submodule (2.5).
Described Mean-shift submodule (2.4), by extracting the color histogram that obtains in the submodule (1.4) at color characteristic, each frame is carried out background plane, promptly mark respective regions according to ratio in the histogram in the drawings, the color degree of approximation is represented with gray-scale value, and the low gray-scale value of elimination.Obtain new predicted position scope by the Mean-shift algorithm again.Described Mean-shift algorithm, by the maximum probability offset direction of iterative computation target point set, convergence obtains corresponding position range, is sent to the decision-making submodule.
Described decision-making submodule (2.5) is accepted the target location of Mean-shift submodule and Kalman filtering submodule.If both centre distances greater than certain value, are then looked for the angle point feature again around the place-centric of Mean-shift.Here Mean-shift is used for definite aircraft position roughly, because colouring information can not be subjected to big disturbing effect, finds target rapidly after disturbing.And light stream and the Kalman result who obtains that combines can obtain relative accurately target location, but is vulnerable to disturb and loses, so do coarse adjustment with Mean-shif, light stream and Kalman filter are done accurate adjustment.
Aircraft coordinate by tracking module (2) output is used to carry out velocity estimation, the The Cloud Terrace instruction sends.
Described velocity estimation submodule (3.1) adopts the Markov chain method that speed calculation is predicted, its main thought is that first speed with the aircraft central point is converted into actual The Cloud Terrace speed:
,
,
The air speed of the location of pixels gained of two frame aircraft central points and actual The Cloud Terrace velocity of rotation before and after representing respectively.
Be the experience factor that can obtain by experiment in advance.
By excavation to historical data, obtain the sequence of operation of complete tracking aircraft, be unit with The Cloud Terrace speed, carry out curve fitting, obtain the landing curve of the aircraft of some different type of machines.
Then the experience speed in actual speed and the Markov model is compared, utilizes time and positional information and m kind state to be complementary:
Obtain this and be curve j constantly, next is the probability of curve i constantly.Find maximum probability, and carry out curve and shift.After determining curve,,,, obtain further speed accurately in conjunction with present speed by the curve speed coupling again by in the certain limit around the current time.
Described instruction sending module (3.2), this module is called corresponding hardware interface according to the above-mentioned optimum speed that matches, and then instruction is mail to camera and The Cloud Terrace.
Description of drawings
Fig. 1 is a modular structure synoptic diagram of the present invention.
Embodiment
Provide better embodiment of the present invention according to Fig. 1 below, and described in detail, enable to understand the present invention better rather than be used for limiting the scope of the invention.
Core algorithm and peripheral composition part-structure are as shown in Figure 1
At first digital video signal is carried out background modeling, thus make it to have background information with as extract aircraft according to model.According to this model, propose mobile aircraft and carry out analysis monitoring thereby can carry out foreground extraction to monitor video.In the middle of this process, aircraft passes through a plurality of triggering frames successively, produces trigger event, and target is extracted, and calls tracking module.Tracking module obtains one group of predicted value by to the tracking of unique point with to the filtering of coordinate; Carry out the Mean-shift algorithm computation by background plane figure, obtain another class value color characteristic; Judgement adjustment by decision-making module draws the target location.Obtain accurately that the aircraft coordinate inputs to the cradle head control module, carry out speed computing and Model Matching and give an order.
The lexical or textual analysis of various piece and function are as follows among Fig. 1:
Digital video signal: it is the video flowing by digital signal encoding.At first judge whether to have obtained target.If obtain then directly vision signal to be sent to tracking module, if not, be sent to module of target detection.
Module of target detection 1: based on the background learning method, tracker is triggered in the position of sense aircraft.
This module of background modeling submodule (1.1) is learnt background by the two field picture of average taking-up, sets up mixed Gauss model and renewal.
Analysis on Prospect submodule (1.2), this module is carried out binary conversion treatment according to the background information that the background modeling module provides to the two field picture that takes out, promptly the difference pixel within the specific limits with the gray-scale value of background model is made as black, thereby has extracted needed foreground information with the bigger white that is made as of the background subtraction opposite sex.
Target is extracted submodule (1.3), this module is partly checked the prospect of picture frame according to the binary image that Analysis on Prospect submodule (1.2) generates, when prospect triggers frame according to predesigned order by some appointed areas, sign has obtained target, produce trigger event, obtain the position range of aircraft in the prospect, this executive mode is effectively avoided the false triggering of noise.
Color characteristic extracts submodule (1.4), and this module obtains the color histogram of appointed area at the color characteristic of aircraft initial position extraction aircraft, is once remaining unchanged in the complete tracing process.
Tracking module 2: the tracking based on two kinds of methods is regulated, and predicts for the position of aircraft next frame.
Angle point feature extraction submodule (2.1), this module is extracted the angle point feature of aircraft at the aircraft initial position, and after when Kalman's predicted value and light stream predicted value are inconsistent, regain angle point.
Optical flow tracking submodule (2.2) based on the light stream pyramid, carries out Feature Points Matching around the unique point of obtaining in former frame, if the unique point number that the match is successful is less than specifies number, then detected characteristics point again.Otherwise, with the median of the unique point obtained as output.
Kalman filtering submodule (2.3) is used for by aircraft movements track cylindricality barrier.Aircraft is by the cylindricality barrier time, and all unique points all can be blocked once.For Kalman filter has been specified the aircraft movements model, if wave filter predicted value and actual measurement unique point coordinate figure are greater than specifying deviation, detected characteristics point again in the specified scope around the revised coordinate of Kalman filter then is by angle point feature submodule (2.1).Otherwise, will revise coordinate and import decision-making submodule (2.5) into.
Mean-shift submodule (2.4), by extracting the color histogram that obtains in the submodule (1.4) at color characteristic, each frame is carried out background plane, promptly mark respective regions according to ratio in the histogram in the drawings, the color degree of approximation is represented with gray-scale value, and the low gray-scale value of elimination.Obtain new predicted position scope by the Mean-shift algorithm again.Described Mean-shift algorithm, by the maximum probability offset direction of iterative computation target point set, convergence obtains corresponding position range, is sent to decision-making submodule (2.5).
Decision-making submodule (2.5) is accepted the target location of Mean-shift submodule (2.4) and Kalman filtering submodule (2.3).If both centre distances greater than certain value, are then looked for the angle point feature by angle point feature extraction submodule (2.1) again around the place-centric of Mean-shift.Here Mean-shift is used for definite aircraft position roughly, because colouring information can not be subjected to big disturbing effect, finds target rapidly after disturbing.And light stream and the Kalman result who obtains that combines can obtain relative accurately target location, but is vulnerable to disturb and loses, so do coarse adjustment with Mean-shift, light stream and Kalman filter are done accurate adjustment.
Cradle head control module 3: in conjunction with predetermined speed, send suitable instruction and give camera and The Cloud Terrace by data mining algorithm.
Velocity estimation submodule (3.1) adopts the Markov chain method that speed calculation is predicted, its main thought is that first speed with the aircraft central point is converted into actual The Cloud Terrace speed:
,
,
The air speed of the location of pixels gained of two frame aircraft central points and actual The Cloud Terrace velocity of rotation before and after representing respectively.
Be the experience factor that can obtain by experiment in advance.By excavation to historical data, obtain the sequence of operation of complete tracking aircraft, be unit with The Cloud Terrace speed, carry out curve fitting, obtain the landing curve of the aircraft of some different type of machines.
Then the experience speed in actual speed and the Markov model is compared, utilizes time and positional information and m kind state to be complementary:
Obtain this and be curve j constantly, next is the probability of curve i constantly.Find maximum probability, and carry out curve and shift.After determining curve,,,, obtain further speed accurately in conjunction with present speed by the curve speed coupling again by in the certain limit around the current time.
Instruction sending module (3.2), this module is called corresponding hardware interface according to the above-mentioned optimum speed that matches, and then instruction is mail to camera and The Cloud Terrace.
Claims (8)
1. aircraft landing tracker based on the The Cloud Terrace camera, it is characterized in that by being installed in monitoring camera and the The Cloud Terrace on the control tower, the simulating signal of being sent by controlled camera head is after being converted into digital video signal, as the input of described tracker; This tracker comprises: module of target detection (1), tracking module (2) and cradle head control module (3); Wherein:
Described module of target detection (1) comprises that background modeling submodule (1.1), Analysis on Prospect submodule (1.2), target are extracted submodule (1.3) and color characteristic extracts submodule (1.4), is sent to tracking module (2) by the detected aircraft coordinate of module of target detection (1);
Described tracking module (2) comprises angle point feature extraction submodule (2.1), by optical flow tracking submodule (2.2), Kalman filtering submodule (2.3), Mean-shift submodule (2.4), decision-making submodule (2.5); Angle point feature extraction submodule (2.1) receives by the detected aircraft coordinate of module of target detection (1); By optical flow tracking submodule (2.2) output, obtain a predicted value by Kalman filtering submodule (2.3), Mean-shift submodule (2.4) draws another predicted value, is made adjustment by decision-making submodule (2.5); Tracking module (2) predicts that with aircraft coordinate is sent to cradle head control module (3);
Described cradle head control module (3) comprises that velocity estimation submodule (3.1) and instruction send submodule (3.2); Carry out the calculating and the analysis of aircraft speed by velocity estimation submodule (3.1) according to aircraft prediction coordinate, obtain corresponding The Cloud Terrace speed, send submodule (3.2) by instruction and implement corresponding The Cloud Terrace operation.
2. the aircraft landing tracker based on the The Cloud Terrace camera according to claim 1 is characterized in that in module of target detection (1):
Described background modeling submodule (1.1) is learnt background by the two field picture of average taking-up, sets up mixed Gauss model and renewal;
Described Analysis on Prospect submodule (1.2), according to the background information that the background modeling submodule provides the two field picture that takes out is carried out binary conversion treatment, promptly the difference pixel within the specific limits with the gray-scale value of background model is made as black, thereby has extracted needed foreground information with the bigger white that is made as of the background subtraction opposite sex;
Described target is extracted submodule (1.3), according to the binary image that the Analysis on Prospect submodule generates the prospect of picture frame is partly checked, when prospect triggers frame according to predesigned order by some appointed areas, sign has obtained target, produce trigger event, obtain the position range of aircraft in the prospect, the color characteristic that passes in the tracking module extracts submodule;
Described color characteristic extracts submodule (1.4), extracts the colouring information of input rectangle in original frame, by setting up color histogram, extracts main color, as the aircraft color information.
3. the aircraft landing tracker based on the The Cloud Terrace camera according to claim 2 is characterized in that in tracking module (2):
Described angle point feature extraction submodule (2.1) in the rectangular extent that triggers the submodule input, is sought unique point, and this unique point is an angle point, storage unique point coordinate;
Described optical flow tracking submodule (2.2) based on the light stream pyramid, carries out Feature Points Matching around the unique point of obtaining in former frame, if the unique point number that the match is successful is less than specifies number, then detected characteristics point again; Otherwise, with the median of the unique point obtained as output;
Described Kalman filtering submodule (2.3), be used for by aircraft movements track cylindricality barrier, Kalman filter is that aircraft is set up motion model, if wave filter predicted value and actual measurement unique point coordinate figure be greater than specifying deviation, detected characteristics point again in the specified scope around the revised coordinate of Kalman filter then; Otherwise, will revise coordinate and pass to decision-making submodule (2.5);
Described Mean-shift submodule (2.4), by extracting the color histogram that obtains in the submodule (1.4) at color characteristic, each frame is carried out background plane, promptly mark respective regions according to ratio in the histogram in the drawings, the color degree of approximation is represented with gray-scale value, and the low gray-scale value of elimination, obtain new predicted position scope by the Mean-shift algorithm again; Described Mean-shift algorithm, by the maximum probability offset direction of iterative computation target point set, convergence obtains corresponding position range, is sent to the decision-making submodule;
Described decision-making submodule (2.5) is accepted the target location of Mean-shift submodule and Kalman filtering submodule; If both centre distances greater than certain value, are then looked for the angle point feature again around the place-centric of Mean-shift.
4. the aircraft landing tracker based on the The Cloud Terrace camera according to claim 3 is characterized in that in cradle head control module (3):
Described velocity estimation submodule (3.1) adopts the Markov chain method that speed calculation is predicted, its method is that first speed with the aircraft central point is converted into actual The Cloud Terrace speed:
,
,
The air speed of the location of pixels gained of two frame aircraft central points and actual The Cloud Terrace velocity of rotation before and after representing respectively,
Be the experience factor that can obtain by experiment in advance;
By excavation to historical data, obtain the sequence of operation of complete tracking aircraft, be unit with The Cloud Terrace speed, carry out curve fitting, obtain the landing curve of the aircraft of some different type of machines;
Then the experience speed in actual speed and the Markov model is compared, utilizes time and positional information and m kind state to be complementary:
Obtain this and be curve j constantly, next is the probability of curve i constantly; Find maximum probability, and carry out curve and shift; After determining curve,,,, obtain further speed accurately in conjunction with present speed by the curve speed coupling again by in the certain limit around the current time;
Described instruction sending module (3.2) according to the above-mentioned optimum speed that matches, is called corresponding hardware interface, then instruction is mail to camera and The Cloud Terrace.
5. aircraft landing track algorithm based on the The Cloud Terrace camera, it is characterized in that by being installed in monitoring camera and the The Cloud Terrace on the control tower, the simulating signal of being sent by controlled camera head is after being converted into digital video signal, as the input of described tracker; This tracker comprises: module of target detection (1), tracking module (2) and cradle head control module (3); Wherein:
Described module of target detection (1) comprises that background modeling submodule (1.1), Analysis on Prospect submodule (1.2), target are extracted submodule (1.3) and color characteristic extracts submodule (1.4), is sent to tracking module (2) by the detected aircraft coordinate of module of target detection (1);
Described tracking module (2) comprises angle point feature extraction submodule (2.1), by optical flow tracking submodule (2.2), Kalman filtering submodule (2.3), Mean-shift submodule (2.4), decision-making submodule (2.5); Angle point feature extraction submodule (2.1) receives by the detected aircraft coordinate of module of target detection (1); By optical flow tracking submodule (2.2) output, obtain a predicted value by Kalman filtering submodule (2.3), Mean-shift submodule (2.4) draws another predicted value, is made adjustment by decision-making submodule (2.5); Tracking module (2) predicts that with aircraft coordinate is sent to cradle head control module (3);
Described cradle head control module (3) comprises that velocity estimation submodule (3.1) and instruction send submodule (3.2); Carry out the calculating and the analysis of aircraft speed by velocity estimation submodule (3.1) according to aircraft prediction coordinate, obtain corresponding The Cloud Terrace speed, send submodule (3.2) by instruction and implement corresponding The Cloud Terrace operation.
6. the aircraft landing track algorithm based on the The Cloud Terrace camera according to claim 5 is characterized in that in module of target detection (1):
Described background modeling submodule (1.1) is learnt background by the two field picture of average taking-up, sets up mixed Gauss model and renewal;
Described Analysis on Prospect submodule (1.2), according to the background information that the background modeling submodule provides the two field picture that takes out is carried out binary conversion treatment, promptly the difference pixel within the specific limits with the gray-scale value of background model is made as black, thereby has extracted needed foreground information with the bigger white that is made as of the background subtraction opposite sex;
Described target is extracted submodule (1.3), according to the binary image that the Analysis on Prospect submodule generates the prospect of picture frame is partly checked, when prospect triggers frame according to predesigned order by some appointed areas, sign has obtained target, produce trigger event, obtain the position range of aircraft in the prospect, the color characteristic that passes in the tracking module extracts submodule;
Described color characteristic extracts submodule (1.4), extracts the colouring information of input rectangle in original frame, by setting up color histogram, extracts main color, as the aircraft color information.
7. the aircraft landing track algorithm based on the The Cloud Terrace camera according to claim 6 is characterized in that in tracking module (2):
Described angle point feature extraction submodule (2.1) in the rectangular extent that triggers the submodule input, is sought unique point, and this unique point is an angle point, storage unique point coordinate;
Described optical flow tracking submodule (2.2) based on the light stream pyramid, carries out Feature Points Matching around the unique point of obtaining in former frame, if the unique point number that the match is successful is less than specifies number, then detected characteristics point again; Otherwise, with the median of the unique point obtained as output;
Described Kalman filtering submodule (2.3), be used for by aircraft movements track cylindricality barrier, Kalman filter is that aircraft is set up motion model, if wave filter predicted value and actual measurement unique point coordinate figure be greater than specifying deviation, detected characteristics point again in the specified scope around the revised coordinate of Kalman filter then; Otherwise, will revise coordinate and pass to decision-making submodule (2.5);
Described Mean-shift submodule (2.4), by extracting the color histogram that obtains in the submodule (1.4) at color characteristic, each frame is carried out background plane, promptly mark respective regions according to ratio in the histogram in the drawings, the color degree of approximation is represented with gray-scale value, and the low gray-scale value of elimination, obtain new predicted position scope by the Mean-shift algorithm again; Described Mean-shift algorithm, by the maximum probability offset direction of iterative computation target point set, convergence obtains corresponding position range, is sent to the decision-making submodule;
Described decision-making submodule (2.5) is accepted the target location of Mean-shift submodule and Kalman filtering submodule; If both centre distances greater than certain value, are then looked for the angle point feature again around the place-centric of Mean-shift.
8. the aircraft landing track algorithm based on the The Cloud Terrace camera according to claim 7 is characterized in that in cradle head control module (3):
Described velocity estimation submodule (3.1) adopts the Markov chain method that speed calculation is predicted, its method is that first speed with the aircraft central point is converted into actual The Cloud Terrace speed:
,
,
The air speed of the location of pixels gained of two frame aircraft central points and actual The Cloud Terrace velocity of rotation before and after representing respectively,
Be the experience factor that can obtain by experiment in advance;
By excavation to historical data, obtain the sequence of operation of complete tracking aircraft, be unit with The Cloud Terrace speed, carry out curve fitting, obtain the landing curve of the aircraft of some different type of machines;
Then the experience speed in actual speed and the Markov model is compared, utilizes time and positional information and m kind state to be complementary:
Obtain this and be curve j constantly, next is the probability of curve i constantly; Find maximum probability, and carry out curve and shift; After determining curve,,,, obtain further speed accurately in conjunction with present speed by the curve speed coupling again by in the certain limit around the current time;
Described instruction sending module (3.2) according to the above-mentioned optimum speed that matches, is called corresponding hardware interface, then instruction is mail to camera and The Cloud Terrace.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010106146511A CN102043964A (en) | 2010-12-30 | 2010-12-30 | Tracking algorithm and tracking system for taking-off and landing of aircraft based on tripod head and camera head |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010106146511A CN102043964A (en) | 2010-12-30 | 2010-12-30 | Tracking algorithm and tracking system for taking-off and landing of aircraft based on tripod head and camera head |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102043964A true CN102043964A (en) | 2011-05-04 |
Family
ID=43910090
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010106146511A Pending CN102043964A (en) | 2010-12-30 | 2010-12-30 | Tracking algorithm and tracking system for taking-off and landing of aircraft based on tripod head and camera head |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102043964A (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102270348A (en) * | 2011-08-23 | 2011-12-07 | 中国科学院自动化研究所 | Method for tracking deformable hand gesture based on video streaming |
| CN102707724A (en) * | 2012-06-05 | 2012-10-03 | 清华大学 | Visual localization and obstacle avoidance method and system for unmanned plane |
| CN103680291A (en) * | 2012-09-09 | 2014-03-26 | 复旦大学 | Method for realizing simultaneous locating and mapping based on ceiling vision |
| CN104157105A (en) * | 2014-08-24 | 2014-11-19 | 无锡北斗星通信息科技有限公司 | Airplane state detecting and alarming system on runway |
| CN106155081A (en) * | 2016-06-17 | 2016-11-23 | 北京理工大学 | A kind of rotor wing unmanned aerial vehicle target monitoring on a large scale and accurate positioning method |
| CN107423667A (en) * | 2017-04-12 | 2017-12-01 | 杭州奥腾电子股份有限公司 | A kind of method of prediction barrier on car body |
| WO2018053877A1 (en) * | 2016-09-26 | 2018-03-29 | 深圳市大疆创新科技有限公司 | Control method, control device, and delivery system |
| CN108230371A (en) * | 2017-12-29 | 2018-06-29 | 厦门市美亚柏科信息股份有限公司 | Tracking target velocity Forecasting Methodology and storage medium based on holder |
| CN108230370A (en) * | 2017-12-29 | 2018-06-29 | 厦门市美亚柏科信息股份有限公司 | Tracking target velocity Forecasting Methodology and storage medium based on holder |
| CN109902543A (en) * | 2017-12-11 | 2019-06-18 | 北京京东尚科信息技术有限公司 | Target trajectory estimation method, device and Target Tracking System |
| CN110415277A (en) * | 2019-07-24 | 2019-11-05 | 中国科学院自动化研究所 | Based on light stream and the multi-target tracking method of Kalman filtering, system, device |
| CN110956647A (en) * | 2019-11-02 | 2020-04-03 | 上海交通大学 | System and method for dynamically tracking object behaviors in video based on behavior dynamic line model |
| CN111079525A (en) * | 2019-11-05 | 2020-04-28 | 阿里巴巴集团控股有限公司 | Image processing method, apparatus, system and storage medium |
| CN111382784A (en) * | 2020-03-04 | 2020-07-07 | 厦门脉视数字技术有限公司 | Moving target tracking method |
| CN112650298A (en) * | 2020-12-30 | 2021-04-13 | 广东工业大学 | Unmanned aerial vehicle tracking landing method and system |
| CN114727013A (en) * | 2013-04-19 | 2022-07-08 | 索尼公司 | Flight camera and method performed by flight camera |
| CN116991182A (en) * | 2023-09-26 | 2023-11-03 | 北京云圣智能科技有限责任公司 | Unmanned aerial vehicle holder control method, device, system, computer device and medium |
-
2010
- 2010-12-30 CN CN2010106146511A patent/CN102043964A/en active Pending
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102270348A (en) * | 2011-08-23 | 2011-12-07 | 中国科学院自动化研究所 | Method for tracking deformable hand gesture based on video streaming |
| CN102707724A (en) * | 2012-06-05 | 2012-10-03 | 清华大学 | Visual localization and obstacle avoidance method and system for unmanned plane |
| CN102707724B (en) * | 2012-06-05 | 2015-01-14 | 清华大学 | Visual localization and obstacle avoidance method and system for unmanned plane |
| CN103680291A (en) * | 2012-09-09 | 2014-03-26 | 复旦大学 | Method for realizing simultaneous locating and mapping based on ceiling vision |
| CN103680291B (en) * | 2012-09-09 | 2016-12-21 | 复旦大学 | The method synchronizing location and mapping based on ceiling vision |
| CN114727013A (en) * | 2013-04-19 | 2022-07-08 | 索尼公司 | Flight camera and method performed by flight camera |
| US11953904B2 (en) | 2013-04-19 | 2024-04-09 | Sony Group Corporation | Flying camera and a system |
| CN104157105A (en) * | 2014-08-24 | 2014-11-19 | 无锡北斗星通信息科技有限公司 | Airplane state detecting and alarming system on runway |
| CN106155081A (en) * | 2016-06-17 | 2016-11-23 | 北京理工大学 | A kind of rotor wing unmanned aerial vehicle target monitoring on a large scale and accurate positioning method |
| CN106155081B (en) * | 2016-06-17 | 2018-02-16 | 北京理工大学 | A kind of a wide range of target monitoring of rotor wing unmanned aerial vehicle and accurate positioning method |
| WO2018053877A1 (en) * | 2016-09-26 | 2018-03-29 | 深圳市大疆创新科技有限公司 | Control method, control device, and delivery system |
| US11724805B2 (en) | 2016-09-26 | 2023-08-15 | SZ DJI Technology Co., Ltd. | Control method, control device, and carrier system |
| US11092975B2 (en) | 2016-09-26 | 2021-08-17 | SZ DJI Technology Co., Ltd. | Control method, control device, and carrier system |
| CN107423667A (en) * | 2017-04-12 | 2017-12-01 | 杭州奥腾电子股份有限公司 | A kind of method of prediction barrier on car body |
| CN109902543A (en) * | 2017-12-11 | 2019-06-18 | 北京京东尚科信息技术有限公司 | Target trajectory estimation method, device and Target Tracking System |
| CN108230371A (en) * | 2017-12-29 | 2018-06-29 | 厦门市美亚柏科信息股份有限公司 | Tracking target velocity Forecasting Methodology and storage medium based on holder |
| CN108230370A (en) * | 2017-12-29 | 2018-06-29 | 厦门市美亚柏科信息股份有限公司 | Tracking target velocity Forecasting Methodology and storage medium based on holder |
| CN108230370B (en) * | 2017-12-29 | 2020-08-04 | 厦门市美亚柏科信息股份有限公司 | Tracking target speed prediction method based on holder and storage medium |
| CN110415277A (en) * | 2019-07-24 | 2019-11-05 | 中国科学院自动化研究所 | Based on light stream and the multi-target tracking method of Kalman filtering, system, device |
| CN110415277B (en) * | 2019-07-24 | 2022-03-08 | 中国科学院自动化研究所 | Multi-target tracking method, system and device based on optical flow and Kalman filtering |
| CN110956647A (en) * | 2019-11-02 | 2020-04-03 | 上海交通大学 | System and method for dynamically tracking object behaviors in video based on behavior dynamic line model |
| CN110956647B (en) * | 2019-11-02 | 2021-06-22 | 上海交通大学 | System and method for dynamically tracking object behaviors in video based on behavior dynamic line model |
| CN111079525B (en) * | 2019-11-05 | 2023-05-30 | 阿里巴巴集团控股有限公司 | Image processing method, device, system and storage medium |
| CN111079525A (en) * | 2019-11-05 | 2020-04-28 | 阿里巴巴集团控股有限公司 | Image processing method, apparatus, system and storage medium |
| CN111382784B (en) * | 2020-03-04 | 2021-11-26 | 厦门星纵智能科技有限公司 | Moving target tracking method |
| CN111382784A (en) * | 2020-03-04 | 2020-07-07 | 厦门脉视数字技术有限公司 | Moving target tracking method |
| CN112650298B (en) * | 2020-12-30 | 2021-08-17 | 广东工业大学 | Unmanned aerial vehicle tracking landing method and system |
| CN112650298A (en) * | 2020-12-30 | 2021-04-13 | 广东工业大学 | Unmanned aerial vehicle tracking landing method and system |
| CN116991182A (en) * | 2023-09-26 | 2023-11-03 | 北京云圣智能科技有限责任公司 | Unmanned aerial vehicle holder control method, device, system, computer device and medium |
| CN116991182B (en) * | 2023-09-26 | 2023-12-22 | 北京云圣智能科技有限责任公司 | Unmanned aerial vehicle holder control method, device, system, computer device and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102043964A (en) | Tracking algorithm and tracking system for taking-off and landing of aircraft based on tripod head and camera head | |
| Yundong et al. | Multi-block SSD based on small object detection for UAV railway scene surveillance | |
| CN110782481B (en) | Unmanned ship intelligent decision-making method and system | |
| CN110738846A (en) | Vehicle behavior monitoring system based on radar and video group and implementation method thereof | |
| Liu et al. | Insulator detection in aerial images based on faster regions with convolutional neural network | |
| CN104966045A (en) | Video-based airplane entry-departure parking lot automatic detection method | |
| CN102447835A (en) | Non-blind-area multi-target cooperative tracking method and system | |
| CN106331636A (en) | Intelligent video monitoring system and method of oil pipelines based on behavioral event triggering | |
| CN103440491A (en) | Intensive pedestrian flow rate real-time detection method based on color feature | |
| CN105516661B (en) | Principal and subordinate's target monitoring method that fisheye camera is combined with ptz camera | |
| Van Phat et al. | Deep4air: A novel deep learning framework for airport airside surveillance | |
| Zhao et al. | Dynamic object tracking for self-driving cars using monocular camera and lidar | |
| Ma et al. | Visual target detection for energy consumption optimization of unmanned surface vehicle | |
| Wu et al. | A new multi-sensor fusion approach for integrated ship motion perception in inland waterways | |
| Cheng et al. | Moving Target Detection Technology Based on UAV Vision | |
| CN114419444A (en) | Lightweight high-resolution bird group identification method based on deep learning network | |
| Liu et al. | Multi-scale personnel deep feature detection algorithm based on Extended-YOLOv3 | |
| CN109359545A (en) | A kind of collaboration monitoring method and apparatus under complicated low latitude environment | |
| CN112613668A (en) | Scenic spot dangerous area management and control method based on artificial intelligence | |
| CN112542800A (en) | Method and system for identifying transmission line fault | |
| US20230267753A1 (en) | Learning based system and method for visual docking guidance to detect new approaching aircraft types | |
| CN116993681A (en) | Substation inspection defect detection method and system | |
| Xingxin et al. | Adaptive auxiliary input extraction based on vanishing point detection for distant object detection in high-resolution railway scene | |
| CN111145551A (en) | Intersection traffic planning system based on CNN detection follows chapter rate | |
| CN115311617A (en) | Method and system for acquiring passenger flow information of urban rail station area |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20110504 |