CN106023246A - Spatiotemporal context tracking method based on local sensitive histogram - Google Patents
Spatiotemporal context tracking method based on local sensitive histogram Download PDFInfo
- Publication number
- CN106023246A CN106023246A CN201610294447.3A CN201610294447A CN106023246A CN 106023246 A CN106023246 A CN 106023246A CN 201610294447 A CN201610294447 A CN 201610294447A CN 106023246 A CN106023246 A CN 106023246A
- Authority
- CN
- China
- Prior art keywords
- target
- tracking
- local sensitivity
- space
- time
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a spatiotemporal context tracking method based on a local sensitive histogram. The method provided by the invention provides guidance for target tracking in a video image. The method is characterized in that (1) a light-invariant feature is extracted by calculating the local sensitive histogram; and (2) gray-level features in spatiotemporal context tracking are combined to acquire better features. According to the invention, influences can be reduced when illumination is drastically changed, a target is rotated in a plane and the target is blocked; and a great tracking effect is acquired.
Description
Technical field:
The present invention relates to computer vision field, a kind of method that target in video image is tracked.
Background technology:
Target following is a highly important branch in this field of computation vision, is paid close attention to widely by scholar both domestic and external because of its critical role and studies.Meanwhile, target following has a wide range of applications in actual applications, multiple fields such as such as video monitoring, man-machine interaction, vehicle monitoring.
Target following is the important step of intelligent monitoring, and intelligent video monitoring and national defense safety can be played directive function by the research to target following.
Summary of the invention:
It is an object of the invention to as building a kind of method for tracking target having in higher using value, simple video image.
A kind of based on local sensitivity histogrammic space-time context track algorithm, it is based on Bayesian frame, utilize biological vision characteristic, in conjunction with bottom gray feature, illumination invariant feature is extracted based on local sensitivity rectangular histogram, the statistical correlation model setting up target and background realizes following the tracks of, and makes tracking hour offset less and will not be with losing target.The experiment of different video sequence is being shown, compare with multi-instance learning algorithm based on local sensitivity histogrammic space-time contextual algorithms, all show reasonable tracking effect when rotating or block in illumination variation, plane and errors of centration is less, there is higher robustness.
Concrete technical scheme is as follows:
(1) target location needing to follow the tracks of in the first frame is initialized;
(2) calculate local sensitivity rectangular histogram, extract illumination invariant feature;
(3) target location confidence map c (x) is set up, and design conditions probability P (x | c (z), o);
(4) context prior model P (c (z) | o) and spatial context model are calculated respectively
(5) it is calculated the target location of t+1 frame, and updates space-time context model;
(6) target location of present frame is exported
(7) next frame is processed, until having processed all of picture frame.
The invention has the beneficial effects as follows:
1, one is set up based on local sensitivity histogrammic space-time context tracking;
2, the present invention is simple, is capable of stable tracking, is widely used, intelligent video monitoring is had to the value referring to and applying, play directive function in illumination generation acute variation, objective plane in the case of rotating and block for a long time.
The target following be applicable to video image of the present invention, can be that security protection department provides reference, field of intelligent monitoring is played directive function.
Accompanying drawing illustrates:
Fig. 1 is the present invention tracking result figure to Basketball video sequence;
Fig. 2 is the error curve diagram that Basketball video sequence is followed the tracks of by the present invention;
Detailed description of the invention:
Further illustrate the flesh and blood of the present invention below in conjunction with the accompanying drawings with example, but present disclosure is not limited to this.
Embodiment 1:
Obtain Basketball sequence of video images, initialize and follow the tracks of target, manually choose target.Calculate the normalization local sensitivity rectangular histogram of image, generate illumination invariant feature, then set up target location confidence map c (x), and design conditions probability P (x | c (z), o);Calculate context prior model P (c (z) | o) and spatial context model respectivelyIt is calculated the target location of t+1 frame, and updates space-time context model;The target location of output present frameNext frame being processed, terminating to follow the tracks of until having processed all picture frames.
Claims (3)
1. one kind based on local sensitivity histogrammic space-time context tracking.The present invention is to the mesh in video image
Directive function is played in mark tracking, provides reference to video monitoring department.It is characterized in that:
(1) calculate local sensitivity rectangular histogram, extract illumination invariant feature;
(2) combine gray feature and utilize biological vision information, based on Bayesian frame, set up target with
The statistical correlation model of background obtains target location, realizes following the tracks of.
One the most according to claim 1 is based on local sensitivity histogrammic space-time context tracking, and it is special
Levy and be: calculate local sensitivity rectangular histogram, extract illumination invariant feature, so that target characteristic is more steady
Fixed, especially illumination acute variation, block and rotate in objective plane for a long time time track algorithm more
Robust.
One the most according to claim 1 is based on local sensitivity histogrammic space-time context tracking, and it is special
Levy and be: based on utilizing Bayesian frame, in conjunction with gray feature, set up the statistical correlation mould in time and space
Type, confidence map takes the position of maximum and is target location, takes this position and as final position thus realizes following the tracks of.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610294447.3A CN106023246A (en) | 2016-05-05 | 2016-05-05 | Spatiotemporal context tracking method based on local sensitive histogram |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610294447.3A CN106023246A (en) | 2016-05-05 | 2016-05-05 | Spatiotemporal context tracking method based on local sensitive histogram |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106023246A true CN106023246A (en) | 2016-10-12 |
Family
ID=57082275
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610294447.3A Pending CN106023246A (en) | 2016-05-05 | 2016-05-05 | Spatiotemporal context tracking method based on local sensitive histogram |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106023246A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107610159A (en) * | 2017-09-03 | 2018-01-19 | 西安电子科技大学 | Infrared small object tracking based on curvature filtering and space-time context |
CN108447078A (en) * | 2018-02-28 | 2018-08-24 | 长沙师范学院 | The interference of view-based access control model conspicuousness perceives track algorithm |
CN110738685A (en) * | 2019-09-09 | 2020-01-31 | 桂林理工大学 | space-time context tracking method with color histogram response fusion |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104537692A (en) * | 2014-12-30 | 2015-04-22 | 中国人民解放军国防科学技术大学 | Key point stabilization tracking method based on time-space contextual information assisting |
CN104715494A (en) * | 2015-03-30 | 2015-06-17 | 江南大学 | Video target compressing and tracking method based on partial sensitive histogram |
CN104933542A (en) * | 2015-06-12 | 2015-09-23 | 临沂大学 | Logistics storage monitoring method based computer vision |
CN104933735A (en) * | 2015-06-30 | 2015-09-23 | 中国电子科技集团公司第二十九研究所 | A real time human face tracking method and a system based on spatio-temporal context learning |
-
2016
- 2016-05-05 CN CN201610294447.3A patent/CN106023246A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104537692A (en) * | 2014-12-30 | 2015-04-22 | 中国人民解放军国防科学技术大学 | Key point stabilization tracking method based on time-space contextual information assisting |
CN104715494A (en) * | 2015-03-30 | 2015-06-17 | 江南大学 | Video target compressing and tracking method based on partial sensitive histogram |
CN104933542A (en) * | 2015-06-12 | 2015-09-23 | 临沂大学 | Logistics storage monitoring method based computer vision |
CN104933735A (en) * | 2015-06-30 | 2015-09-23 | 中国电子科技集团公司第二十九研究所 | A real time human face tracking method and a system based on spatio-temporal context learning |
Non-Patent Citations (2)
Title |
---|
KAIHUA ZHANG等: "Fast Visual Tracking via Dense Spatio-temporal Context Learning", 《PROCEEDINGS OF THE 2014 EUROPEAN CONFERENCE ON COMPUTER VISION》 * |
SHENGFENG HE 等: "Visual Tracking via Locality Sensitive Histograms", 《PROCEEDINGS OF THE 2013 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107610159A (en) * | 2017-09-03 | 2018-01-19 | 西安电子科技大学 | Infrared small object tracking based on curvature filtering and space-time context |
CN108447078A (en) * | 2018-02-28 | 2018-08-24 | 长沙师范学院 | The interference of view-based access control model conspicuousness perceives track algorithm |
CN108447078B (en) * | 2018-02-28 | 2022-06-10 | 长沙师范学院 | Interference perception tracking algorithm based on visual saliency |
CN110738685A (en) * | 2019-09-09 | 2020-01-31 | 桂林理工大学 | space-time context tracking method with color histogram response fusion |
CN110738685B (en) * | 2019-09-09 | 2023-05-05 | 桂林理工大学 | Space-time context tracking method integrating color histogram response |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Bissacco et al. | Fast human pose estimation using appearance and motion via multi-dimensional boosting regression | |
Mei et al. | Robust visual tracking and vehicle classification via sparse representation | |
JP4677046B2 (en) | Fast human pose estimation using appearance and motion via multidimensional boost regression | |
He et al. | Infrared target tracking via weighted correlation filter | |
Zhang et al. | Adaptive local movement modeling for robust object tracking | |
Fang et al. | Efficient and robust fragments-based multiple kernels tracking | |
Nie et al. | Single/cross-camera multiple-person tracking by graph matching | |
CN106023246A (en) | Spatiotemporal context tracking method based on local sensitive histogram | |
Fan et al. | Discriminative multi-task objects tracking with active feature selection and drift correction | |
Zhong et al. | Structured partial least squares for simultaneous object tracking and segmentation | |
Walia et al. | A novel approach of multi-stage tracking for precise localization of target in video sequences | |
Li et al. | Visual tracking by proto-objects | |
Yang et al. | On-road vehicle tracking using keypoint-based representation and online co-training | |
Yin et al. | Research on real-time object tracking by improved CamShift | |
Quan et al. | Robust object tracking using enhanced random ferns | |
Zhang et al. | A dynamic detection and data association method based on probabilistic models for visual SLAM | |
Hu et al. | Fusing target information from multiple views for robust visual tracking | |
CN110781803B (en) | Human body posture recognition method based on extended Kalman filter | |
Liu et al. | An unsupervised automatic change detection approach based on visual attention mechanism | |
Wang et al. | Collaborative model based uav tracking via local kernel feature | |
JP2019174868A (en) | Object tracking apparatus, method and program | |
Yu et al. | An improved deep-learning monocular visual SLAM method based on local features | |
Zhang | Sparse Visual Localization in GPS-Denied Indoor Environments | |
Qiao et al. | A kernel density estimation model for moving object detection | |
Zhao et al. | Fast facial feature tracking with multi-cue particle filter |
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 |
Application publication date: 20161012 |
|
WD01 | Invention patent application deemed withdrawn after publication |