CN102521841A - Multi-target object tracking method - Google Patents
Multi-target object tracking method Download PDFInfo
- Publication number
- CN102521841A CN102521841A CN2011103729255A CN201110372925A CN102521841A CN 102521841 A CN102521841 A CN 102521841A CN 2011103729255 A CN2011103729255 A CN 2011103729255A CN 201110372925 A CN201110372925 A CN 201110372925A CN 102521841 A CN102521841 A CN 102521841A
- Authority
- CN
- China
- Prior art keywords
- moving object
- tracking
- chained list
- characteristic
- following 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
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a multi-target object tracking method, and overcomes the diversity of the implementation framework of the existing target tracking algorithm and the comparison difficulty of the algorithm performance. By inducing and abstracting ordinary steps of target tracking, the multi-target object tracking method realizes unified multi-target object tracking. The invention has the positive effects that with the adoption of abstract models, various tracking algorithms can be compared conveniently, and new feature extraction algorithms are added to the tracking algorithms so as to improve the tracking performance of a multi-target object. The tracking of multiple automobiles by means of image processing in an intelligent traffic video analysis system can be realized by the multi-target object tracking method.
Description
Technical field
The invention belongs to the image processing technique field, be specifically related to a kind of multiple target objects volume tracing method.
Background technology
Target object track algorithm based on Digital Image Processing mainly contains at present: optical flow method, particle filter and CamShift algorithm.Through these algorithms are discovered, it is big that they all have calculated amount, and the inner neither one unified shader of algorithm is unfavorable for the shortcoming of the comparison and the fusion of multiple algorithm.
Summary of the invention
In order to overcome the above-mentioned shortcoming of prior art; The invention provides a kind of multiple target objects volume tracing method; Overcome existing target tracking algorism and realized the diversity of framework; And algorithm performance be difficult for comparatively, through concluding the general step that abstract object is followed the tracks of, realized unified multiple target objects volume tracing.
The technical solution adopted for the present invention to solve the technical problems is: a kind of multiple target objects volume tracing method comprises the steps:
The first step, employing moving object detection algorithm extract the coordinate of the moving object in the digital picture;
Second step, the coordinate of the moving object that extracts is carried out feature extraction;
The 3rd step, characteristic matching:
Traversal target following track chained list; With the characteristic of the total movement object that obtains respectively with target following track chained list in object features compare; And according to comparative result; The successful moving object of coupling is added pairing target following track chained list,, and will mate in the new target following track chained list of unsuccessful moving object adding generation new target following track chained list of characteristic generation of the unsuccessful moving object of coupling;
The 4th step, tracking finish:
When the total movement object in the frame of digital image after all the step, said methods analyst finished by the 3rd, the target following track chained list that deletion does not have moving object to add, the tracking of the pairing moving object of these track chained lists is finished.
For same object, if become two objects after extracting, then carry out the merging of connected component, merge later again that size is little object and filter out, the coordinate of the moving object that will be left then and ratio whole picture are approaching extracts.
Second step, described feature extracting methods was: adopt the various features extraction algorithm to handle the moving object image that extract the front; The characteristic that then various algorithms of different is extracted is carried out characteristic relatively with target object respectively; Respectively success ratio, mortality and mistake matching rate are added up and compared; Be selected at last that power is the highest, mistake matching rate minimum feature extraction algorithm, as the feature extracting method of system.
Compared with prior art, good effect of the present invention is: through abstract model, can more various easily track algorithms, and add new feature extraction algorithm therein, to improve the tracking performance of multiple goal object.The present invention can realize the tracking of a plurality of vehicles through the mode of Flame Image Process in the intelligent transportation video analytic system.
Description of drawings
The present invention will explain through example and with reference to the mode of accompanying drawing, wherein:
Fig. 1 is the schematic flow sheet of the inventive method.
Embodiment
A kind of multiple target objects volume tracing method, as shown in Figure 1, comprise the steps:
The coordinate of the first step, moving object extracts:
Adopt the moving object detection algorithm that the motion parts in the digital picture is extracted, wherein, owing to reasons such as picture noises; Originally be same object, may become two objects after extracting, at this moment just can carry out the merging of connected component; Again undersized object is filtered out after merging; Remaining is exactly institute's moving object proper with ratio whole picture that will detect, can their coordinate be extracted at this moment, so that subsequent analysis.
Second the step, feature extraction is carried out in the moving object that extracts:
Adopt various feature extracting method to handle the moving object image that extract the front; The characteristic that then various feature extracting method is extracted is carried out characteristic relatively with target object respectively; Success ratio, mortality and mistake matching rate are added up and compared; Be selected at last that power is the highest, mistake matching rate minimum feature extraction algorithm, as the system features method for distilling.
The 3rd step, characteristic matching:
Traversal target following track chained list; With the characteristic of the total movement object that obtains respectively with target following track chained list in object features compare; And according to comparative result; The successful moving object of coupling is added pairing target following track chained list,, and will mate in the new target following track chained list of unsuccessful moving object adding generation new target following track chained list of characteristic generation of the unsuccessful moving object of coupling.
The 4th step, tracking finish:
When the total movement object in the frame of digital image after all the step, said methods analyst finished by the 3rd, the target following track chained list that deletion does not have moving object to add, the tracking of the pairing moving object of these track chained lists is finished.
Claims (3)
1. a multiple target objects volume tracing method is characterized in that, comprises the steps:
The first step, employing moving object detection algorithm extract the coordinate of the moving object in the digital picture;
Second step, the coordinate of the moving object that extracts is carried out feature extraction;
The 3rd step, characteristic matching:
Traversal target following track chained list; With the characteristic of the total movement object that obtains respectively with target following track chained list in object features compare; And according to comparative result; The successful moving object of coupling is added pairing target following track chained list,, and will mate in the new target following track chained list of unsuccessful moving object adding generation new target following track chained list of characteristic generation of the unsuccessful moving object of coupling;
The 4th step, tracking finish:
When the total movement object in the frame of digital image after all the step, said methods analyst finished by the 3rd, the target following track chained list that deletion does not have moving object to add, the tracking of the pairing moving object of these track chained lists is finished.
2. multiple target objects volume tracing method according to claim 1; It is characterized in that; For same object,, then carry out the merging of connected component if become two objects after extracting; Again that size is little object filters out after merging, and the coordinate of the moving object that will be left then and ratio whole picture are approaching extracts.
3. multiple target objects volume tracing method according to claim 1; It is characterized in that; Second step, described feature extracting methods was: adopt the various features extraction algorithm to handle the moving object image that extract the front; The characteristic that then various algorithms of different is extracted is carried out characteristic relatively with target object respectively; Respectively success ratio, mortality and mistake matching rate are added up and compare, be selected at last power the highest, miss the minimum feature extraction algorithm of matching rate, as the feature extracting method of system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011103729255A CN102521841A (en) | 2011-11-22 | 2011-11-22 | Multi-target object tracking method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011103729255A CN102521841A (en) | 2011-11-22 | 2011-11-22 | Multi-target object tracking method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102521841A true CN102521841A (en) | 2012-06-27 |
Family
ID=46292745
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011103729255A Pending CN102521841A (en) | 2011-11-22 | 2011-11-22 | Multi-target object tracking method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102521841A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107085823A (en) * | 2016-02-16 | 2017-08-22 | 北京小米移动软件有限公司 | Face image processing process and device |
| WO2018058530A1 (en) * | 2016-09-30 | 2018-04-05 | 富士通株式会社 | Target detection method and device, and image processing apparatus |
| CN110047095A (en) * | 2019-03-06 | 2019-07-23 | 平安科技（深圳）有限公司 | Tracking, device and terminal device based on target detection |
| CN110163124A (en) * | 2019-04-30 | 2019-08-23 | 北京易华录信息技术股份有限公司 | A kind of trajectory track processing system |
| CN111738063A (en) * | 2020-05-08 | 2020-10-02 | 华南理工大学 | Ship target tracking method, system, computer equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101025862A (en) * | 2007-02-12 | 2007-08-29 | 吉林大学 | Video based mixed traffic flow parameter detecting method |
| CN101098465A (en) * | 2007-07-20 | 2008-01-02 | 哈尔滨工程大学 | Moving object detecting and tracing method in video monitor |
| CN101673403A (en) * | 2009-10-10 | 2010-03-17 | 安防制造（中国）有限公司 | Target following method in complex interference scene |
| CN101727570A (en) * | 2008-10-23 | 2010-06-09 | 华为技术有限公司 | Tracking method, track detection processing unit and monitor system |
| CN101840507A (en) * | 2010-04-09 | 2010-09-22 | 江苏东大金智建筑智能化***工程有限公司 | Target tracking method based on character feature invariant and graph theory clustering |
| CN101854516A (en) * | 2009-04-02 | 2010-10-06 | 北京中星微电子有限公司 | Video monitoring system, video monitoring server and video monitoring method |
-
2011
- 2011-11-22 CN CN2011103729255A patent/CN102521841A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101025862A (en) * | 2007-02-12 | 2007-08-29 | 吉林大学 | Video based mixed traffic flow parameter detecting method |
| CN101098465A (en) * | 2007-07-20 | 2008-01-02 | 哈尔滨工程大学 | Moving object detecting and tracing method in video monitor |
| CN101727570A (en) * | 2008-10-23 | 2010-06-09 | 华为技术有限公司 | Tracking method, track detection processing unit and monitor system |
| CN101854516A (en) * | 2009-04-02 | 2010-10-06 | 北京中星微电子有限公司 | Video monitoring system, video monitoring server and video monitoring method |
| CN101673403A (en) * | 2009-10-10 | 2010-03-17 | 安防制造（中国）有限公司 | Target following method in complex interference scene |
| CN101840507A (en) * | 2010-04-09 | 2010-09-22 | 江苏东大金智建筑智能化***工程有限公司 | Target tracking method based on character feature invariant and graph theory clustering |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107085823A (en) * | 2016-02-16 | 2017-08-22 | 北京小米移动软件有限公司 | Face image processing process and device |
| WO2018058530A1 (en) * | 2016-09-30 | 2018-04-05 | 富士通株式会社 | Target detection method and device, and image processing apparatus |
| CN109478333A (en) * | 2016-09-30 | 2019-03-15 | 富士通株式会社 | Object detection method, device and image processing equipment |
| CN110047095A (en) * | 2019-03-06 | 2019-07-23 | 平安科技（深圳）有限公司 | Tracking, device and terminal device based on target detection |
| CN110047095B (en) * | 2019-03-06 | 2023-07-21 | 平安科技（深圳）有限公司 | Tracking method and device based on target detection and terminal equipment |
| CN110163124A (en) * | 2019-04-30 | 2019-08-23 | 北京易华录信息技术股份有限公司 | A kind of trajectory track processing system |
| CN111738063A (en) * | 2020-05-08 | 2020-10-02 | 华南理工大学 | Ship target tracking method, system, computer equipment and storage medium |
| CN111738063B (en) * | 2020-05-08 | 2023-04-18 | 华南理工大学 | Ship target tracking method, system, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Performance and challenges of 3D object detection methods in complex scenes for autonomous driving | |
| CN102521841A (en) | Multi-target object tracking method | |
| CN103871079A (en) | Vehicle tracking method based on machine learning and optical flow | |
| Siam et al. | On-board multiple target detection and tracking on camera-equipped aerial vehicles | |
| CN103905824A (en) | Video semantic retrieval and compression synchronization camera system and method | |
| CN104662560B (en) | A kind of method of video image processing and system | |
| CN102682453A (en) | Moving vehicle tracking method based on multi-feature fusion | |
| CN103971524A (en) | Traffic flow detection method based on machine vision | |
| CN103810718A (en) | Method and device for detection of violently moving target | |
| CN105740835A (en) | Preceding vehicle detection method based on vehicle-mounted camera under night-vision environment | |
| CN103198491A (en) | Indoor visual positioning method | |
| Feng et al. | Learning multi-layer attention aggregation Siamese network for robust RGBT tracking | |
| CN107547865A (en) | Trans-regional human body video frequency object tracking intelligent control method | |
| CN102254396B (en) | Intrusion detection method and device based on video | |
| CN103187083B (en) | A kind of storage means based on time domain video fusion and system thereof | |
| CN105516661A (en) | Master-slave target monitoring system and method in combination of fisheye camera and PTZ camera | |
| Zhang et al. | LiDAR-camera panoptic segmentation via geometry-consistent and semantic-aware alignment | |
| Zhang et al. | An optical flow based moving objects detection algorithm for the UAV | |
| CN113361475A (en) | Multi-spectral pedestrian detection method based on multi-stage feature fusion information multiplexing | |
| Zhang et al. | Opensight: A simple open-vocabulary framework for lidar-based object detection | |
| CN104331908A (en) | Compressed haar character and motion information fused target tracking method | |
| Xu et al. | [Retracted] Technical Research on Moving Target Monitoring and Intelligent Tracking Algorithm Based on Machine Vision | |
| Sun et al. | Research on target recognition and tracking in mobile augmented reality assisted maintenance | |
| Zhou et al. | A relation network design for visible thermal person re-identification | |
| CN101937511B (en) | Rapid image matching method based on stochastic parallel optimization algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20120627 |