CN105321180A - Target tracking and positioning method and apparatus based on cloud computing - Google Patents
Target tracking and positioning method and apparatus based on cloud computing Download PDFInfo
- Publication number
- CN105321180A CN105321180A CN201510694807.4A CN201510694807A CN105321180A CN 105321180 A CN105321180 A CN 105321180A CN 201510694807 A CN201510694807 A CN 201510694807A CN 105321180 A CN105321180 A CN 105321180A
- Authority
- CN
- China
- Prior art keywords
- point
- registration
- target
- registration point
- track
- 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
Classifications
-
- 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
- G06T2207/30248—Vehicle exterior or interior
Landscapes
- Image Analysis (AREA)
Abstract
The present invention provides a target tracking and positioning method and apparatus based on cloud computing. The method comprises the following steps of: carrying out registering on an angular point of a current frame of a target object and an angular point of a next frame adjacent to the current frame, and acquiring registration points; and selecting a preset number of the registration points, generating a target affine matrix, obtaining a position of a convergence point of the current frame in the next frame by the target affine matrix, performing iteration by the position, and performing tracking and positioning on the target object. The target tracking and positioning method and apparatus based on cloud computing implement accurate positioning on the target object and improve the efficiency of tracking and positioning the target object.
Description
Technical field
The invention belongs to positioning control field, particularly relate to a kind of target following localization method based on cloud computing and device.
Background technology
Along with the fast development of cloud computing technology, make cloud computing more and more wider in every field application, such as: cloud computing application relates to traffic, public security, government etc., but in field of traffic, vehicle tracking is a difficult point always, and the conventional track algorithm adopted at present is Mean-shift track algorithm, adopts Mean-shift track algorithm to carry out iteration and looks for the calculated amount of tracking target larger, and its primary iteration point elects the convergence point of previous frame as, be easy to like this cause following the tracks of unsuccessfully.
Therefore, in the urgent need to provide a kind of fast, the accurate scheme of positioned vehicle in the position of next frame, solve track algorithm calculated amount large, follow the tracks of the problem such as unsuccessfully.
Summary of the invention
The invention provides a kind of target following localization method based on cloud computing and device, to solve the problem.
The invention provides a kind of target following localization method based on cloud computing.Said method comprises the following steps:
Registration is carried out to the angle point of the angle point of destination object present frame next frame adjacent with present frame, obtains registration point;
Select preset number registration point, generate target affine matrix and obtain the convergence point of present frame in the position of next frame by described target affine matrix, and carrying out iteration by described position, track and localization is carried out to destination object.
The present invention also provides a kind of target following locating device based on cloud computing, comprises registration point acquisition module, track and localization module; Wherein, described registration point acquisition module is connected with described track and localization module;
Described registration point acquisition module, for carrying out registration to the angle point of the angle point of destination object present frame next frame adjacent with present frame, obtaining registration point and described registration point is sent to described track and localization module;
Described track and localization module, for selecting preset number registration point, generating target affine matrix and obtaining the convergence point of present frame in the position of next frame by described target affine matrix, and carrying out iteration by described position, carrying out track and localization to destination object.
The present invention also provides a kind of target following locating device based on cloud computing, comprises registration point acquisition module, registration point processing module, track and localization module; Wherein, described registration point acquisition module is connected with described track and localization module by described registration point processing module;
Described registration point acquisition module, for carrying out registration to the angle point of the angle point of destination object present frame next frame adjacent with present frame, obtaining registration point and described registration point is sent to described registration point processing module;
Described registration point processing module, for obtaining slope information between angle point that registration goes out and according to described slope information, determining target registration point and described target registration point is sent to described track and localization module;
Described track and localization module, for selecting preset number registration point, generating target affine matrix and obtaining the convergence point of present frame in the position of next frame by described target affine matrix, and carrying out iteration by described position, carrying out track and localization to destination object.
By following scheme: carry out registration to the angle point of the angle point of destination object present frame next frame adjacent with present frame, obtain registration point; Select preset number registration point, generate target affine matrix and obtain the convergence point of present frame in the position of next frame by described target affine matrix, and carry out iteration by described position, track and localization is carried out to destination object, achieve the accurate location to destination object, improve the track and localization efficiency to destination object.
By following scheme: carry out registration to the angle point of the angle point of destination object present frame next frame adjacent with present frame, after obtaining registration point, also comprise: obtain slope information between the angle point that goes out of registration and according to described slope information, determine target registration point, further increase the accuracy of registration, ensure that the accuracy to destination object track and localization.
Accompanying drawing explanation
Accompanying drawing described herein is used to provide a further understanding of the present invention, and form a application's part, schematic description and description of the present invention, for explaining the present invention, does not form inappropriate limitation of the present invention.In the accompanying drawings:
Figure 1 shows that the target following localization method processing flow chart based on cloud computing of the embodiment of the present invention 1;
Figure 2 shows that the target following positioning device structure figure based on cloud computing of the embodiment of the present invention 2;
Figure 3 shows that the target following positioning device structure figure based on cloud computing of the embodiment of the present invention 3.
Embodiment
Hereinafter also describe the present invention in detail with reference to accompanying drawing in conjunction with the embodiments.It should be noted that, when not conflicting, the embodiment in the application and the feature in embodiment can combine mutually.
Figure 1 shows that the target following localization method processing flow chart based on cloud computing of the embodiment of the present invention 1, comprise the following steps:
Step 101: carry out registration to the angle point of the angle point of destination object present frame next frame adjacent with present frame, obtains registration point;
Further, carry out registration to the angle point of the angle point of destination object present frame next frame adjacent with present frame, the process obtaining registration point is:
The angle point of destination object present frame, the angle point of next frame adjacent with present frame according to normalized crosscorrelation matching algorithm and NCC (NormalizedCrossCorrelationmethod is extracted by angle point algorithm, normalized crosscorrelation) matching algorithm carries out registration, obtains registration point.
Step 102: slope information between the angle point that acquisition registration goes out also according to described slope information, determines target registration point;
Further, obtain slope information between the angle point that goes out of registration and according to described slope information, determine that the process of target registration point is:
Slope information between the angle point gone out by statistics registration, is generated histogram, and then determines target registration point.
Further, slope information between the angle point gone out by statistics registration, is generated histogram, and then determines that the process of target registration point is:
Slope information between the angle point gone out by statistics registration, generates histogram, if slope exceedes preset range, then corresponding registration point is abnormal registration point, and abnormal registration point is deleted from registration point, then remaining registration point is target registration point.
Wherein, slope information between the angle point gone out by statistics registration, generate histogram, because the angle point of the angle point of present frame and next frame adjacent with present frame is if the point on destination object, so their slope then reflects the movable information of destination object, if described slope exceedes default slope range, then corresponding registration point is abnormal registration point, and from registration point, abnormal registration point is deleted, then remaining registration point is target registration point.
Such scheme improves the accuracy of registration.
Step 103: select preset number registration point, generates target affine matrix and obtains the convergence point of present frame in the position of next frame by described target affine matrix, and carrying out iteration by described position, carrying out track and localization to destination object.
Further, select preset number registration point, the process generating target affine matrix is:
If select three assembly on schedule, then generate initial affine matrix by three assembly on schedule;
Bring residue registration point into described initial affine matrix, obtain the registration point accounting in default error range, if described registration point accounting (such as: (such as: 60%), then described initial affine matrix is optimum matching matrix is also target affine matrix 80%) to be greater than preset ratio.
Further, select preset number registration point, generate target affine matrix and obtain the convergence point of present frame in the position of next frame by described target affine matrix, and carrying out iteration by described position, the process of destination object being carried out to track and localization is:
Select preset number registration point, generate target affine matrix;
Described target affine matrix, by affined transformation, obtains the convergence point of present frame in the position of next frame;
With described position for primary iteration point carries out iteration, track and localization is carried out to destination object.
Wherein, Mean-shift track algorithm is adopted to carry out iteration.
Figure 2 shows that the target following positioning device structure figure based on cloud computing of the embodiment of the present invention 2, comprising: registration point acquisition module 201, track and localization module 202; Wherein, described registration point acquisition module 201 is connected with described track and localization module 202;
Described registration point acquisition module 201, for carrying out registration to the angle point of the angle point of destination object present frame next frame adjacent with present frame, obtaining registration point and described registration point is sent to described track and localization module 202;
Described track and localization module 202, for selecting preset number registration point, generate target affine matrix and obtain the convergence point of present frame in the position of next frame by described target affine matrix, and carrying out iteration by described position, track and localization is carried out to destination object.
Figure 3 shows that the target following positioning device structure figure based on cloud computing of the embodiment of the present invention 3, the basis of Fig. 2 adds " registration point processing module 200 "; Wherein, described registration point acquisition module 201 is connected with described track and localization module 202 by described registration point processing module 200;
Described registration point acquisition module 201, for carrying out registration to the angle point of the angle point of destination object present frame next frame adjacent with present frame, obtaining registration point and described registration point is sent to described registration point processing module 200;
Described registration point processing module 200, for obtaining slope information between angle point that registration goes out and according to described slope information, determining target registration point and described target registration point is sent to described track and localization module 202;
Described track and localization module 202, for selecting preset number registration point, generate target affine matrix and obtain the convergence point of present frame in the position of next frame by described target affine matrix, and carrying out iteration by described position, track and localization is carried out to destination object.
By following scheme: carry out registration to the angle point of the angle point of destination object present frame next frame adjacent with present frame, obtain registration point; Select preset number registration point, generate target affine matrix and obtain the convergence point of present frame in the position of next frame by described target affine matrix, and carry out iteration by described position, track and localization is carried out to destination object, achieve the accurate location to destination object, improve the track and localization efficiency to destination object.
By following scheme: carry out registration to the angle point of the angle point of destination object present frame next frame adjacent with present frame, after obtaining registration point, also comprise: obtain slope information between the angle point that goes out of registration and according to described slope information, determine target registration point, further increase the accuracy of registration, ensure that the accuracy to destination object track and localization.
The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for a person skilled in the art, the present invention can have various modifications and variations.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (10)
1., based on a target following localization method for cloud computing, it is characterized in that, comprise the following steps:
Registration is carried out to the angle point of the angle point of destination object present frame next frame adjacent with present frame, obtains registration point;
Select preset number registration point, generate target affine matrix and obtain the convergence point of present frame in the position of next frame by described target affine matrix, and carrying out iteration by described position, track and localization is carried out to destination object.
2. method according to claim 1, is characterized in that, carries out registration to the angle point of the angle point of destination object present frame next frame adjacent with present frame, and the process obtaining registration point is:
The angle point of destination object present frame, the angle point of next frame adjacent with present frame carry out registration according to normalized crosscorrelation matching algorithm and NCC matching algorithm, acquisition registration point is extracted by angle point algorithm.
3. method according to claim 1, is characterized in that, carries out registration to the angle point of the angle point of destination object present frame next frame adjacent with present frame, after obtaining registration point, also comprises:
Slope information between the angle point that acquisition registration goes out also according to described slope information, determines target registration point.
4. method according to claim 3, is characterized in that, slope information between the angle point that acquisition registration goes out also according to described slope information, determines that the process of target registration point is:
Slope information between the angle point gone out by statistics registration, is generated histogram, and then determines target registration point.
5. method according to claim 4, is characterized in that, slope information between the angle point gone out by statistics registration, is generated histogram, and then determine that the process of target registration point is:
Slope information between the angle point gone out by statistics registration, generates histogram, if slope exceedes preset range, then corresponding registration point is abnormal registration point, and abnormal registration point is deleted from registration point, then remaining registration point is target registration point.
6. method according to claim 1, is characterized in that, selects preset number registration point, and the process generating target affine matrix is:
If select three assembly on schedule, then generate initial affine matrix by three assembly on schedule;
Bring residue registration point into described initial affine matrix, obtain the registration point accounting in default error range, if described registration point accounting is greater than preset ratio, then described initial affine matrix is optimum matching matrix is also target affine matrix.
7. the method according to claim 1 or 6, it is characterized in that, select preset number registration point, generate target affine matrix and obtain the convergence point of present frame in the position of next frame by described target affine matrix, and carry out iteration by described position, the process of destination object being carried out to track and localization is:
Select preset number registration point, generate target affine matrix;
Described target affine matrix, by affined transformation, obtains the convergence point of present frame in the position of next frame;
With described position for primary iteration point carries out iteration, track and localization is carried out to destination object.
8. method according to claim 7, is characterized in that, adopts Mean-shift track algorithm to carry out iteration.
9. based on a target following locating device for cloud computing, it is characterized in that, comprise registration point acquisition module, track and localization module; Wherein, described registration point acquisition module is connected with described track and localization module;
Described registration point acquisition module, for carrying out registration to the angle point of the angle point of destination object present frame next frame adjacent with present frame, obtaining registration point and described registration point is sent to described track and localization module;
Described track and localization module, for selecting preset number registration point, generating target affine matrix and obtaining the convergence point of present frame in the position of next frame by described target affine matrix, and carrying out iteration by described position, carrying out track and localization to destination object.
10. based on a target following locating device for cloud computing, it is characterized in that, comprise registration point acquisition module, registration point processing module, track and localization module; Wherein, described registration point acquisition module is connected with described track and localization module by described registration point processing module;
Described registration point acquisition module, for carrying out registration to the angle point of the angle point of destination object present frame next frame adjacent with present frame, obtaining registration point and described registration point is sent to described registration point processing module;
Described registration point processing module, for obtaining slope information between angle point that registration goes out and according to described slope information, determining target registration point and described target registration point is sent to described track and localization module;
Described track and localization module, for selecting preset number registration point, generating target affine matrix and obtaining the convergence point of present frame in the position of next frame by described target affine matrix, and carrying out iteration by described position, carrying out track and localization to destination object.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510694807.4A CN105321180A (en) | 2015-10-21 | 2015-10-21 | Target tracking and positioning method and apparatus based on cloud computing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510694807.4A CN105321180A (en) | 2015-10-21 | 2015-10-21 | Target tracking and positioning method and apparatus based on cloud computing |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105321180A true CN105321180A (en) | 2016-02-10 |
Family
ID=55248495
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510694807.4A Pending CN105321180A (en) | 2015-10-21 | 2015-10-21 | Target tracking and positioning method and apparatus based on cloud computing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105321180A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108961304A (en) * | 2017-05-23 | 2018-12-07 | 阿里巴巴集团控股有限公司 | Identify the method for sport foreground and the method for determining target position in video in video |
| CN111352712A (en) * | 2020-02-25 | 2020-06-30 | 程瑞萍 | Cloud computing task tracking processing method and device, cloud computing system and server |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090141936A1 (en) * | 2006-03-01 | 2009-06-04 | Nikon Corporation | Object-Tracking Computer Program Product, Object-Tracking Device, and Camera |
| CN102456225A (en) * | 2010-10-22 | 2012-05-16 | 深圳中兴力维技术有限公司 | Video monitoring system and moving target detecting and tracking method thereof |
-
2015
- 2015-10-21 CN CN201510694807.4A patent/CN105321180A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090141936A1 (en) * | 2006-03-01 | 2009-06-04 | Nikon Corporation | Object-Tracking Computer Program Product, Object-Tracking Device, and Camera |
| CN102456225A (en) * | 2010-10-22 | 2012-05-16 | 深圳中兴力维技术有限公司 | Video monitoring system and moving target detecting and tracking method thereof |
Non-Patent Citations (1)
| Title |
|---|
| 代兵云: ""基于角点和surf特征的目标尺度自适应研究"", 《万方数据知识服务平台》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108961304A (en) * | 2017-05-23 | 2018-12-07 | 阿里巴巴集团控股有限公司 | Identify the method for sport foreground and the method for determining target position in video in video |
| CN111352712A (en) * | 2020-02-25 | 2020-06-30 | 程瑞萍 | Cloud computing task tracking processing method and device, cloud computing system and server |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104048659B (en) | The conversion method and system of map coordinates system | |
| CN106485732B (en) | A kind of method for tracking target of video sequence | |
| CN102279406B (en) | Fence identification method using global positioning system (GPS) to position tracks | |
| CN106886748B (en) | TLD-based variable-scale target tracking method applicable to unmanned aerial vehicle | |
| CN108279670B (en) | Method, apparatus and computer readable medium for adjusting point cloud data acquisition trajectory | |
| CN105957041B (en) | A kind of wide-angle lens infrared image distortion correction method | |
| CN113012197B (en) | Binocular vision odometer positioning method suitable for dynamic traffic scene | |
| CN108154149B (en) | License plate recognition method based on deep learning network sharing | |
| CN106878951A (en) | User trajectory analysis method and system | |
| EP1061748A3 (en) | Video object segmentation using active contour modelling with global relaxation | |
| CN113409459A (en) | Method, device and equipment for producing high-precision map and computer storage medium | |
| CN109493385A (en) | Autonomic positioning method in a kind of mobile robot room of combination scene point line feature | |
| CN107742304A (en) | Method and device for determining movement track, mobile robot and storage medium | |
| CN106547014B (en) | The generation method of crystal localization method and look-up table | |
| CN104007763A (en) | Method for fixed electronic nose nodes and mobile robot to search for smell source cooperatively | |
| CN110209184A (en) | A kind of unmanned plane barrier-avoiding method based on binocular vision system | |
| CN116879870B (en) | Dynamic obstacle removing method suitable for low-wire-harness 3D laser radar | |
| CN105043354A (en) | System utilizing camera imaging to precisely position moving target | |
| CN103593856A (en) | Method and system for tracking single target | |
| CN111679303A (en) | Comprehensive positioning method and device for multi-source positioning information fusion | |
| Zhang et al. | Cloud motion tracking system using low-cost sky imager for PV power ramp-rate control | |
| CN105279371B (en) | A kind of traverse measurement system POS precision ameliorative way based on control point | |
| CN105321180A (en) | Target tracking and positioning method and apparatus based on cloud computing | |
| CN113971697B (en) | Air-ground cooperative vehicle positioning and orientation method | |
| Ahmadi et al. | HDPV-SLAM: Hybrid depth-augmented panoramic visual SLAM for mobile mapping system with tilted LiDAR and panoramic visual camera |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160210 |