CN104820998A - Human body detection and tracking method and device based on unmanned aerial vehicle mobile platform - Google Patents
Human body detection and tracking method and device based on unmanned aerial vehicle mobile platform Download PDFInfo
- Publication number
- CN104820998A CN104820998A CN201510275245.XA CN201510275245A CN104820998A CN 104820998 A CN104820998 A CN 104820998A CN 201510275245 A CN201510275245 A CN 201510275245A CN 104820998 A CN104820998 A CN 104820998A
- Authority
- CN
- China
- Prior art keywords
- human body
- eigenwert
- target
- target body
- recognizer
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30232—Surveillance
Abstract
The invention discloses a human body detection and tracking method and device based on an unmanned aerial vehicle mobile platform and belongs to the computer vision field. The technical key points of the method comprise a human body detector training step, a target human body recognizer offline training step, a target human body detection step and a human body tracking step, wherein the target human body detection step is characterized by receiving a current video frame shot by an unmanned aerial vehicle, extracting object characteristic value in the current video frame, sending the object characteristic value to a human body detector, the human body detector judging whether a human body is detected according to the characteristic value, and if so, further sending the characteristic value to a target human body recognizer, the target human body recognizer judging whether a target human body is detected according to the characteristic value, and if so, labeling the characteristic value and adding the characteristic value to a tracking list; and the human body tracking step is characterized by predicating position of the target human body in the next video frame according to the coordinate position of the target human body in the current video frame.
Description
Technical field
Invention belongs to computer vision field, is a kind of real-time method for tracking target, long-time, the stable human body tracing method under the complex background under especially a kind of unmanned aerial vehicle platform.
Background technology
Unmanned spacecraft is called for short " unmanned plane ", and english abbreviation is " UAV ", is a kind of unmanned vehicle handled by radio robot or self presetting apparatus.In numerous applications of unmanned plane, target following is especially all a difficult problem urgently to be resolved hurrily to the tracking of human body all the time.This is because human body is generally as less in the optical imaging apparatus visual field of unmanned plane, resolution is not high, and in nonrigid irregular movement, its outward appearance and profile change at any time; And the difficult problem such as tracking target is in complicated ground scenery background, there is poor contrast, blocks, stereoscopic vision effect; In addition there is shake in unmanned plane in flight course, makes tenacious tracking become more difficult.
(application number: 201310666989.5) disclose a kind of UAS to mobile surface targets Kinematic Positioning and method, can export the geographic coordinate information such as the longitude and latitude of tracked mobile surface targets to Chinese patent " UAS and method to mobile surface targets Kinematic Positioning " in real time.The method, by considering the angular altitude and position angle of surely taking aim at platform, in conjunction with the attitude angle of acquisition device, longitude, latitude and elevation information, obtains the positional information of mobile surface targets by a series of coordinate conversion, thus the object realizing location and follow the tracks of.The method still belongs to traditional localization method, all comparatively complicated for the software and hardware implemented, and is unfavorable for application and popularization on small-sized especially consumer level unmanned plane.Chinese patent " a kind of unmanned plane dynamic target tracking of view-based access control model and location " (application number 201310059457.5) discloses a kind of unmanned plane dynamic target tracking and localization method of view-based access control model, can complete the detection to moving target and image trace voluntarily.The tracking that the method adopts traditional MeanShift algorithm and Kalman filter method to realize target, when target occur blocking, the situation such as jump time, the method can not ensure tracking steady in a long-term.
To sum up, in the application of unmanned plane, also do not have at present a kind ofly to carry out for a long time for the especially nonrigid human body target of moving target, the stable and anti-tracking blocked.
Summary of the invention
First technical matters to be solved by this invention is for moving target human body, provides a kind of target body to detect and long-time, tenacious tracking method.
Human detection based on unmanned plane moving platform provided by the invention and tracking, comprising:
Human body detector training step, uses the eigenwert of human body to train the first model of cognition, obtains human body detector;
Target body recognizer off-line training step, uses the eigenwert of target body to train the second model of cognition, obtains target body recognizer; The tag along sort value that wherein eigenwert of target body is corresponding is true, and the tag along sort value that the eigenwert of non-targeted human body is corresponding is false;
Target body detecting step, receive the current video frame of unmanned plane shooting, extract the eigenwert of the object in current video frame, the eigenwert of described object is sent into human body detector, human body detector judges whether human body to be detected according to described eigenwert, if then further described eigenwert is sent into target body recognizer, target body recognizer judges whether target body to be detected according to described eigenwert; If then described eigenwert carried out marking and add tracking list to;
Human body tracking step, according to the coordinate position target of prediction human body position in next frame of video of target body at current video frame.
Further, target body recognizer on-line study step is also comprised:
When target body recognizer detects target body, then the target body eigenwert of current video frame is sent into online study module;
On-line study module utilizes described eigenwert on-line training target body recognizer.
Further, in target body detecting step, also comprise target body give step for change: when there is no a target body in the frame of video of unmanned plane shooting, utilize Kalman filter may appear at region in next frame frame of video according to the frame of video target of prediction human body photographed before lose objects human body; After next frame of video photographs, the eigenwert of object in region described in advantage distillation, again eigenwert is sent into human body detector, human body detector judges whether human body to be detected according to described eigenwert, if then further described eigenwert is sent into target body recognizer, target body recognizer judges whether target body to be detected according to described eigenwert.
Described human body tracking step comprises further, utilizes the position of optical flow method target of prediction human body in next frame of video.
Further, described eigenwert comprises resemblance value and motion characteristic value.
Present invention also offers a kind of human detection based on unmanned plane moving platform and tracking means, comprising:
Target body detection module, for receiving the current video frame of unmanned plane shooting, extract the eigenwert of the object in current video frame, the eigenwert of described object is sent into human body detector, human body detector judges whether human body to be detected according to described eigenwert, if then further described eigenwert is sent into target body recognizer, target body recognizer judges whether target body to be detected according to described eigenwert; If then described eigenwert carried out marking and add tracking list to;
Human tracking module, for according to the coordinate position target of prediction human body position in next frame of video of target body at current video frame.
Further, described human body detector is that training obtains like this: end user's body characteristics value trains the first model of cognition, obtains human body detector;
Described target body recognizer is that training obtains like this: use the eigenwert of target body to train the second model of cognition, obtain target body recognizer.
Further, also comprise target body recognizer on-line study module, for when target body recognizer detects target body, then utilize the target body eigenwert on-line training target body recognizer of current video frame.
Further, in target body detection module, also comprise target body give submodule for change, for when there is no target body in the frame of video that unmanned plane is taken, utilize Kalman filter may appear at region in next frame frame of video according to the frame of video target of prediction human body photographed before lose objects human body; After next frame of video photographs, the eigenwert of object in region described in advantage distillation, eigenwert is sent into human body detector, human body detector judges whether human body to be detected according to described eigenwert, if then further described eigenwert is sent into target body recognizer, target body recognizer judges whether target body to be detected according to described eigenwert.
Described human tracking module is further used for utilizing the position of optical flow method target of prediction human body in next frame of video.
In sum, owing to have employed technique scheme, the invention has the beneficial effects as follows:
1. the present invention is by off-line training human body detector and target body recognizer, achieves the identification of target body.While tracking target human body, whenever target body being detected, then its eigenwert and label thereof are upgraded target body recognizer as new sample training, constantly improve the training sample of target body recognizer online, the accuracy of further raising human bioequivalence, makes human body tracking more stable.
2. after being blocked in the frame of video photographed at unmanned plane when target body or following the tracks of loss, the position that the present invention adopts Kalman filter to carry out target of prediction human body may to occur in next frame of video, and whether preferential this position of detecting has target body, after target body reappears, the position of target body can be determined fast, thus overcome shake that unmanned plane moving platform brings and the problem such as to block, realize long-time, stable and effectively follow the tracks of.
Accompanying drawing explanation
Examples of the present invention will be described by way of reference to the accompanying drawings, wherein:
Fig. 1 is the process flow diagram of first embodiment of the invention.
Fig. 2 is the process flow diagram of second embodiment of the invention.
Embodiment
All features disclosed in this instructions, or the step in disclosed all methods or process, except mutually exclusive feature and/or step, all can combine by any way.
Arbitrary feature disclosed in this instructions, unless specifically stated otherwise, all can be replaced by other equivalences or the alternative features with similar object.That is, unless specifically stated otherwise, each feature is an example in a series of equivalence or similar characteristics.
As Fig. 1, first embodiment of the invention comprises the following steps:
Object selection: follow the tracks of before starting, need to choose target body.The mode of choosing can be manually choose or automatically choose.Manually choose is adopt the mode of manually punctuating to determine target to be tracked for unmanned plane passback video image; Automatically choose without the need to manual intervention, the rule by human detection, recognizer and setting is chosen automatically by machine, and selection operation can realize on unmanned plane, also can realize at remote terminal.It should be noted that, Object selection is not the steps necessary of object detecting and tracking, can also choose by realize target by other means.
Human body detector is trained: use the eigenwert of human body to train the first model of cognition, obtain human body detector.
In the present embodiment, the eigenwert of human body extracts from the multitude of video frame that unmanned plane photographs.Characteristics of human body's value can use gradient orientation histogram (the H O G) method proposed by people such as Dalal to extract.The method, by the gradient of topography and the description of directional spreding situation, extracts the gradient orientation histogram of image local area as detection feature.Eigenwert can comprise resemblance value (as size, profile, color) and motion characteristic value (as walk, run, stand, bend over, squat down).
What training method adopted is support vector machines.
Target body recognizer off-line training: use the eigenwert of target body to train the second model of cognition, obtain target body recognizer.The acquisition methods of the eigenwert of the target body in this step is the same with previous step, and the model of cognition that this step uses also can the same with previous step.
Target body detects: the current video frame receiving unmanned plane shooting, extract the eigenwert of the object in current video frame, the eigenwert of described object is sent into human body detector, human body detector judges whether human body to be detected according to described eigenwert, if then further described eigenwert is sent into target body recognizer, target body recognizer judges whether target body to be detected according to described eigenwert; If then described eigenwert carried out marking and add tracking list to.
Concrete, when similarity is greater than setting threshold value, the similarity that the eigenwert of the object in human body detector calculating current video frame and characteristics of human body are worth, then thinks that described object is human body.Target body recognizer calculates the similarity of described object features value and target body eigenwert, thinks that described object is target body when similarity is greater than setting threshold value.Described threshold value can be 60% or more.
Human body tracking: according to the coordinate position target of prediction human body position in next frame of video of target body at current video frame.In a specific embodiment, the tracking in this step can adopt optical flow method, and the method utilizes image sequence to carry out estimated position velocity field about time (t) and the grey scale change in space (x, y).The method advantage is to detect self-movement target, does not need the information knowing scene in advance, and can be used for the situation of camera motion.
What photograph due to unmanned plane is the video of mobile context, and the state of target body is all different at every frame, along with the carrying out followed the tracks of, feature at first may be not suitable for continuing the model as following the tracks of, therefore need on-line study, constantly introduce new eigenwert, upgrade target body recognizer.
Therefore, target body recognizer on-line training step is also introduced in other embodiments:
When target body recognizer detects target body, then the target body eigenwert of current video frame is sent into online study module; On-line study module utilizes described eigenwert on-line training target body recognizer.
During on-line study, in order to ensure the validity of feature value vector and prevent feature value vector from increasing without limitation with tracing process, can following measures be adopted: the weight calculating each eigenwert in feature value vector, and set weight threshold to reject the eigenwert of low weight.
Second embodiment
See Fig. 2, the present embodiment also add target body and gives step for change in the target body detecting step of the first embodiment: when target body because after being blocked or disappearing in the frame of video that exceeds camera coverage scope and photograph at unmanned plane, utilizes Kalman filter may appear at region in next frame frame of video according to the frame of video target of prediction human body photographed before lose objects human body; After next frame of video photographs, the eigenwert of object in region described in advantage distillation, again eigenwert is sent into human body detector, human body detector judges whether human body to be detected according to described eigenwert, if then further described eigenwert is sent into target body recognizer, target body recognizer judges whether target body to be detected according to described eigenwert.The benefit done like this is, can give target body for change fast, be conducive to the tenacious tracking of target body after lose objects human body from next frame of video.
The present invention is not limited to aforesaid embodiment.The present invention expands to any new feature of disclosing in this manual or any combination newly, and the step of the arbitrary new method disclosed or process or any combination newly.
Claims (10)
1., based on human detection and the tracking of unmanned plane moving platform, it is characterized in that, comprising:
Human body detector training step, uses the eigenwert of human body to train the first model of cognition, obtains human body detector;
Target body recognizer off-line training step, uses the eigenwert of target body to train the second model of cognition, obtains target body recognizer;
Target body detecting step, receive the current video frame of unmanned plane shooting, extract the eigenwert of the object in current video frame, the eigenwert of described object is sent into human body detector, human body detector judges whether human body to be detected according to described eigenwert, if then further described eigenwert is sent into target body recognizer, target body recognizer judges whether target body to be detected according to described eigenwert; If then described eigenwert carried out marking and add tracking list to;
Human body tracking step, according to the coordinate position target of prediction human body position in next frame of video of target body at current video frame.
2. a kind of human detection based on unmanned plane moving platform according to claim 1 and tracking, is characterized in that, also comprises target body recognizer on-line study step:
When target body recognizer detects target body, then the target body eigenwert of current video frame is sent into online study module;
On-line study module utilizes described eigenwert on-line training target body recognizer.
3. a kind of human detection based on unmanned plane moving platform according to claim 2 and tracking, it is characterized in that, in target body detecting step, also comprise target body give step for change: target body occurs in tracing process when losing, utilize Kalman filter may appear at region in next frame frame of video according to the frame of video target of prediction human body photographed before lose objects human body; And in the follow-up frame of video photographed of unmanned plane the eigenwert of object in region described in advantage distillation, again eigenwert is sent into human body detector, human body detector judges whether human body to be detected according to described eigenwert, if then further described eigenwert is sent into target body recognizer, target body recognizer judges whether target body to be detected according to described eigenwert.
4. a kind of human detection based on unmanned plane moving platform according to claim 1 and tracking, is characterized in that, described human body tracking step comprises further, utilizes the position of optical flow method target of prediction human body in next frame of video.
5. a kind of human detection based on unmanned plane moving platform according to claim 1 and tracking, it is characterized in that, described eigenwert comprises resemblance value and motion characteristic value.
6., based on human detection and the tracking means of unmanned plane moving platform, it is characterized in that, comprising:
Target body detection module, for receiving the current video frame of unmanned plane shooting, extract the eigenwert of object in current video frame, the eigenwert of described object is sent into human body detector, human body detector judges whether human body to be detected according to described eigenwert, if then further described eigenwert is sent into target body recognizer, target body recognizer judges whether target body to be detected according to described eigenwert; If then described eigenwert carried out marking and add tracking list to;
Human tracking module, for according to the coordinate position target of prediction human body position in next frame of video of target body at current video frame.
7. a kind of human detection based on unmanned plane moving platform according to claim 6 and tracking means, is characterized in that,
Described human body detector is that training obtains like this: use the eigenwert of human body to train the first model of cognition, obtain human body detector;
Described target body recognizer is that training obtains like this: use the eigenwert of target body to train the second model of cognition, obtain target body recognizer.
8. a kind of human detection based on unmanned plane moving platform according to claim 7 and tracking means, it is characterized in that, also comprise target body recognizer on-line study module, for when target body recognizer detects target body, utilize the target body eigenwert on-line training target body recognizer of current video frame.
9. a kind of human detection based on unmanned plane moving platform according to claim 6 or 7 and tracking means, it is characterized in that, in target body detection module, also comprise target body give submodule for change, for when there is no target body in the frame of video that unmanned plane is taken, utilize Kalman filter may appear at region in next frame frame of video according to the frame of video target of prediction human body photographed before lose objects human body; After next frame of video photographs, the eigenwert of object in region described in advantage distillation, eigenwert is sent into human body detector, human body detector judges whether human body to be detected according to described eigenwert, if then further described eigenwert is sent into target body recognizer, target body recognizer judges whether target body to be detected according to described eigenwert.
10. a kind of human detection based on unmanned plane moving platform according to claim 6 and tracking means, is characterized in that, described human tracking module is further used for utilizing the position of optical flow method target of prediction human body in next frame of video.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510275245.XA CN104820998B (en) | 2015-05-27 | 2015-05-27 | A kind of human testing based on unmanned motor platform and tracking and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510275245.XA CN104820998B (en) | 2015-05-27 | 2015-05-27 | A kind of human testing based on unmanned motor platform and tracking and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104820998A true CN104820998A (en) | 2015-08-05 |
| CN104820998B CN104820998B (en) | 2019-11-26 |
Family
ID=53731283
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510275245.XA Active CN104820998B (en) | 2015-05-27 | 2015-05-27 | A kind of human testing based on unmanned motor platform and tracking and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104820998B (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105184257A (en) * | 2015-09-08 | 2015-12-23 | 北京航空航天大学 | Target detection method and device |
| CN105898343A (en) * | 2016-04-07 | 2016-08-24 | 广州盈可视电子科技有限公司 | Video live broadcasting method and device and terminal video live broadcasting method and device |
| CN105912989A (en) * | 2016-04-01 | 2016-08-31 | 成都云图秀色科技有限公司 | Flight instruction generation system based on image identification and method |
| CN106155093A (en) * | 2016-07-22 | 2016-11-23 | 王威 | A kind of robot based on computer vision follows the system and method for human body |
| WO2017045326A1 (en) * | 2015-09-17 | 2017-03-23 | 深圳市十方联智科技有限公司 | Photographing processing method for unmanned aerial vehicle |
| CN106548170A (en) * | 2016-11-04 | 2017-03-29 | 湖南科技学院 | A kind of vehicles peccancy image processing method and system |
| CN106598071A (en) * | 2016-12-20 | 2017-04-26 | 北京小米移动软件有限公司 | Following-type flight control method, following-type flight control device, and unmanned aerial vehicle |
| CN107622273A (en) * | 2016-07-13 | 2018-01-23 | 深圳雷柏科技股份有限公司 | A kind of target detection and the method and apparatus of identification |
| CN107748860A (en) * | 2017-09-01 | 2018-03-02 | 中国科学院深圳先进技术研究院 | Method for tracking target, device, unmanned plane and the storage medium of unmanned plane |
| CN108089695A (en) * | 2016-11-23 | 2018-05-29 | 纳恩博（北京）科技有限公司 | A kind of method and apparatus for controlling movable equipment |
| CN108717522A (en) * | 2018-04-18 | 2018-10-30 | 上海交通大学 | A kind of human body target tracking method based on deep learning and correlation filtering |
| CN109034171A (en) * | 2018-07-24 | 2018-12-18 | 北京易泊时代科技有限公司 | Unlicensed vehicle checking method and device in video flowing |
| CN109829898A (en) * | 2019-01-17 | 2019-05-31 | 柳州康云互联科技有限公司 | One kind is for measurement detection system and method neural network based in internet detection |
| CN109978815A (en) * | 2017-12-14 | 2019-07-05 | 欧姆龙株式会社 | Detection system, information processing unit, evaluation method and storage medium |
| CN110298866A (en) * | 2019-06-06 | 2019-10-01 | 武汉易科空间信息技术股份有限公司 | Ground object tracking and system based on unmanned plane image technology |
| WO2019218139A1 (en) * | 2018-05-15 | 2019-11-21 | 合刃科技（武汉）有限公司 | Object recognition device |
| CN111263946A (en) * | 2018-05-15 | 2020-06-09 | 合刃科技（武汉）有限公司 | Object recognition method and computer-readable storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102881022A (en) * | 2012-07-20 | 2013-01-16 | 西安电子科技大学 | Concealed-target tracking method based on on-line learning |
| CN103268616A (en) * | 2013-04-18 | 2013-08-28 | 北京工业大学 | Multi-feature multi-sensor method for mobile robot to track moving body |
| CN103310466A (en) * | 2013-06-28 | 2013-09-18 | 安科智慧城市技术（中国）有限公司 | Single target tracking method and achievement device thereof |
| CN103324950A (en) * | 2012-03-22 | 2013-09-25 | 中国科学院计算技术研究所 | Human body reappearance detecting method and system based on online study |
-
2015
- 2015-05-27 CN CN201510275245.XA patent/CN104820998B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103324950A (en) * | 2012-03-22 | 2013-09-25 | 中国科学院计算技术研究所 | Human body reappearance detecting method and system based on online study |
| CN102881022A (en) * | 2012-07-20 | 2013-01-16 | 西安电子科技大学 | Concealed-target tracking method based on on-line learning |
| CN103268616A (en) * | 2013-04-18 | 2013-08-28 | 北京工业大学 | Multi-feature multi-sensor method for mobile robot to track moving body |
| CN103310466A (en) * | 2013-06-28 | 2013-09-18 | 安科智慧城市技术（中国）有限公司 | Single target tracking method and achievement device thereof |
Non-Patent Citations (1)
| Title |
|---|
| 黄立: "基于在线学习的视频检测与跟踪技术研究", 《万方数据 学位论文》 * |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105184257A (en) * | 2015-09-08 | 2015-12-23 | 北京航空航天大学 | Target detection method and device |
| CN105184257B (en) * | 2015-09-08 | 2018-08-07 | 北京航空航天大学 | Object detection method and device |
| WO2017045326A1 (en) * | 2015-09-17 | 2017-03-23 | 深圳市十方联智科技有限公司 | Photographing processing method for unmanned aerial vehicle |
| CN105912989A (en) * | 2016-04-01 | 2016-08-31 | 成都云图秀色科技有限公司 | Flight instruction generation system based on image identification and method |
| CN105898343B (en) * | 2016-04-07 | 2019-03-12 | 广州盈可视电子科技有限公司 | A kind of net cast, terminal net cast method and apparatus |
| CN105898343A (en) * | 2016-04-07 | 2016-08-24 | 广州盈可视电子科技有限公司 | Video live broadcasting method and device and terminal video live broadcasting method and device |
| CN107622273A (en) * | 2016-07-13 | 2018-01-23 | 深圳雷柏科技股份有限公司 | A kind of target detection and the method and apparatus of identification |
| CN106155093A (en) * | 2016-07-22 | 2016-11-23 | 王威 | A kind of robot based on computer vision follows the system and method for human body |
| CN106548170A (en) * | 2016-11-04 | 2017-03-29 | 湖南科技学院 | A kind of vehicles peccancy image processing method and system |
| CN106548170B (en) * | 2016-11-04 | 2019-11-19 | 湖南科技学院 | A kind of violation vehicle image processing method and system |
| CN108089695B (en) * | 2016-11-23 | 2021-05-18 | 纳恩博（北京）科技有限公司 | Method and device for controlling movable equipment |
| CN108089695A (en) * | 2016-11-23 | 2018-05-29 | 纳恩博（北京）科技有限公司 | A kind of method and apparatus for controlling movable equipment |
| CN106598071A (en) * | 2016-12-20 | 2017-04-26 | 北京小米移动软件有限公司 | Following-type flight control method, following-type flight control device, and unmanned aerial vehicle |
| CN106598071B (en) * | 2016-12-20 | 2019-10-11 | 北京小米移动软件有限公司 | Flight control method and device, the unmanned plane of trailing type |
| CN107748860A (en) * | 2017-09-01 | 2018-03-02 | 中国科学院深圳先进技术研究院 | Method for tracking target, device, unmanned plane and the storage medium of unmanned plane |
| CN109978815A (en) * | 2017-12-14 | 2019-07-05 | 欧姆龙株式会社 | Detection system, information processing unit, evaluation method and storage medium |
| CN108717522A (en) * | 2018-04-18 | 2018-10-30 | 上海交通大学 | A kind of human body target tracking method based on deep learning and correlation filtering |
| WO2019218139A1 (en) * | 2018-05-15 | 2019-11-21 | 合刃科技（武汉）有限公司 | Object recognition device |
| CN111263946A (en) * | 2018-05-15 | 2020-06-09 | 合刃科技（武汉）有限公司 | Object recognition method and computer-readable storage medium |
| CN111316321A (en) * | 2018-05-15 | 2020-06-19 | 合刃科技（武汉）有限公司 | Object recognition device |
| CN109034171A (en) * | 2018-07-24 | 2018-12-18 | 北京易泊时代科技有限公司 | Unlicensed vehicle checking method and device in video flowing |
| CN109034171B (en) * | 2018-07-24 | 2022-10-04 | 北京易泊时代科技有限公司 | Method and device for detecting unlicensed vehicles in video stream |
| CN109829898A (en) * | 2019-01-17 | 2019-05-31 | 柳州康云互联科技有限公司 | One kind is for measurement detection system and method neural network based in internet detection |
| CN109829898B (en) * | 2019-01-17 | 2023-05-16 | 柳州康云互联科技有限公司 | Measurement detection system and method based on neural network for Internet detection |
| CN110298866A (en) * | 2019-06-06 | 2019-10-01 | 武汉易科空间信息技术股份有限公司 | Ground object tracking and system based on unmanned plane image technology |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104820998B (en) | 2019-11-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104820998A (en) | Human body detection and tracking method and device based on unmanned aerial vehicle mobile platform | |
| US10634500B2 (en) | Aircraft and obstacle avoidance method and system thereof | |
| US10580206B2 (en) | Method and apparatus for constructing three-dimensional map | |
| US20170300759A1 (en) | Automated multiple target detection and tracking system | |
| US9875579B2 (en) | Techniques for enhanced accurate pose estimation | |
| CN102436738B (en) | Traffic monitoring device based on unmanned aerial vehicle (UAV) | |
| CN104794468A (en) | Human face detection and tracking method based on unmanned aerial vehicle mobile platform | |
| CN106204595B (en) | A kind of airdrome scene three-dimensional panorama monitoring method based on binocular camera | |
| CN104881650A (en) | Vehicle tracking method based on unmanned aerial vehicle (UAV) dynamic platform | |
| CN109141442B (en) | Navigation method based on UWB positioning and image feature matching and mobile terminal | |
| US20170024904A1 (en) | Augmented reality vision system for tracking and geolocating objects of interest | |
| US20150138310A1 (en) | Automatic scene parsing | |
| CN107610157B (en) | Unmanned aerial vehicle target tracking method and system | |
| US10976163B2 (en) | Robust vision-inertial pedestrian tracking with heading auto-alignment | |
| CN112050810B (en) | Indoor positioning navigation method and system based on computer vision | |
| CN104978390A (en) | Context-aware object detection in aerial photographs/videos using travel path metadata | |
| CN103488801B (en) | A kind of airport target detection method based on geographical information space database | |
| CN109584264B (en) | Unmanned aerial vehicle vision guiding aerial refueling method based on deep learning | |
| CN114092668A (en) | Virtual-real fusion method, device, equipment and storage medium | |
| WO2020243256A1 (en) | System and method for navigation and geolocation in gps-denied environments | |
| CN112198884A (en) | Unmanned aerial vehicle mobile platform landing method based on visual guidance | |
| Lukashevich et al. | The new approach for reliable UAV navigation based on onboard camera image processing | |
| Lee et al. | A feasibility study on smartphone localization using image registration with segmented 3d building models based on multi-material classes | |
| Vadlamani et al. | A novel indoor positioning system for unprepared firefighting scenarios | |
| KR102603467B1 (en) | Space formation and positioning error correction system and method through digital twin-linked augmented reality camera |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |