基于壓縮特征的魚眼視頻目標跟蹤算法研究

李雅倩; 賈璐; 李海濱; 張文明; 張巖松

doi:10.11999/JEIT170745

基于壓縮特征的魚眼視頻目標跟蹤算法研究

doi: 10.11999/JEIT170745

基金項目:

河北省自然科學基金(F2015203212)

計量
- 文章訪問數(shù): 1442
- HTML全文瀏覽量: 180
- PDF下載量: 177
- 被引次數(shù): 0
出版歷程
- 收稿日期: 2017-07-21
- 修回日期: 2018-01-24
- 刊出日期: 2018-05-19

Research on Target Tracking Algorithm from Fisheye Camera Based on Compressive Sensing

Funds:

The Natural Science Foundation of Hebei Province (F2015203212)

摘要

摘要: 該文針對畸變嚴重的魚眼圖像中的目標跟蹤，提出一種能適應尺度變化、姿態(tài)變化以及形狀畸變的魚眼視頻目標跟蹤的方法。該方法首先將灰度特征和相對梯度特征相結合得到目標的高維特征，然后對其平均降維得到目標的壓縮特征。并根據(jù)魚眼成像模型得到投影點的運動特性，確定目標的運動范圍。為了適應尺度變化，在塊匹配運動估計思想的基礎上，對目標跟蹤框的頂點分別進行由粗到精的定位，并在此過程中根據(jù)跟蹤框的尺度相應改變壓縮特征的尺度。實驗結果表明：該算法在目標畸變、尺度變化、姿態(tài)變化以及局部遮擋等情況下，判斷指標均優(yōu)于其他對比算法。
- 目標跟蹤 /
- 魚眼成像模型 /
- 相對梯度特征 /
- 壓縮特征 /
- 塊匹配運動估計
Abstract: For object detection in fisheye images which present serious distortion, an object tracking method is proposed to deal with scale variance, pose change and distortion. Firstly, gray feature and gradient feature are combined to obtain a high dimensional feature of the target, then reduce its dimensionality by averaging to obtain targets compressive feature. According to fisheye imaging model, motion of object point is modeled, and range of motion of target is predicted. In order to adjust to scale variance, corner points are positioned respectively in a coarse to fine manner based on the block matching motion estimation, and the scale of compressed feature is changed along with scale change of object box. Experimental results show that the proposed algorithm is superior to other algorithms in the case of distortion, scale change, pose change and part occlusion.
- Target tracking /
- Fisheye imaging model /
- Gradient feature /
- Compression feature /
- Block matching motion estimation

HTML全文

參考文獻(22)

L Lijun and WU Xuewei. Design of initial structure of fisheye lens[J]. Acta Optica Sinica, 2017, 37(2): 105-114.

呂麗軍, 吳學偉. 魚眼鏡頭初始結構的設計[J]. 光學學報, 2017, 37(2): 105-114.

URBAN S, LEITLOFF J, and HINZ S. Improved wide-angle, fisheye and omnidirectional camera calibration[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2015, 108(8): 72-79. doi: 10.1016/j.isprsjprs.2015.06.005.

PEREZ Y A, LOPEZ N G, and GUERRERO J. A novel hybrid camera system with depth and fisheye cameras[C]. International Conference on Pattern Recognition. Cancun, Mexico, 2017: 2789-2794. doi: 10.1109/ICPR.2016.7900058.

WALLHOFF F, ZOBL M, and RIGOLL G. Face tracking in meeting room scenarios using omnidirectional views[C]. International Conference on Pattern Recognition, Washington, USA, 2004: 933-936. doi: 10.1109/ICPR.2004. 368.

BAKSTEIN H and LEONARDIS A. Catadioptric image- based rendering for mobile robot localization[C]. International Conference on Computer Vision, Rio de Janeiro, Brazil, 2007: 1-6. doi: 10.1109/ICCV.2007.4409199.

CAO Z. Dynamic omni-directional vision localization using a beacon tracker based on particle filter[J]. The International Society for Optical Engineering, 2007, 6764(9): 13-28. doi: 10.1117/12.733862.

BRITO J H, ANGST R, KOSER K, et al. Radial distortion self-calibration[C]. Computer Vision and Pattern Recognition, Washington, USA, 2013: 1368-1375. doi: 10.1109/CVPR. 2013.180.

CHEN X, HUANG K, and TAN T. Object tracking across non-overlapping cameras using adaptive models[C]. International Conference on Computer Vision, Berlin, Germany, 2013: 464-477. doi: 10.1007/978-3-642-37484- 5_38.

DEMONCEAUX C and VASSEUR P. Omnidirectional image processing using geodesic metric[C]. IEEE International Conference on Image Processing, Piscataway, USA, 2009: 221-224. doi: 10.1109/ICIP.2009.5414485.

BAZIN J C, YOON K J, KWEON I, et al. Particle filter approach adapted to catadioptric images for target tracking application[C]. British Machine Vision Conference, London, UK, 2009: 1-11. doi: 10.5244/C.23.37.

TAIANA M, GASPAR J, NASCIMENTO J, et al. 3D tracking by catadioptric vision based on particle filters[C]. RoboCup Robot Soccer World Cup XI, Atlanta, USA, 2007: 77-88. doi: 10.1007/978-3-540-68847-1_7.

WANG X, LI W, WANG C, et al. An improved particle filter tracking algorithm for fisheye camera[C]. Chinese Control and Decision Conference, Yinchuan, China, 2010: 329-332. doi: 10.1109/CCDC.2016.7531004.

WANG W, XU Y, WANG Y, et al. Effective weighted compressive tracking[C]. International Conference on Image and Graphics. Qingdao, China, 2013: 353-357. doi: 10.1109/ ICIG.2013.77.

汪龍. 一種新的基于自適應窗口的壓縮跟蹤算法[J]. 計算機與數(shù)字工程, 2016, 44(9): 1700-1704. doi: 10.3969/j.issn. 1672-9722.2016.09.016.

WANG Long. A new compressive tracking algorithm based on adaptive window[J]. Computer and Digital Engineering, 2016, 44(9): 1700-1704. doi: 10.3969/j.issn.1672-9722.2016. 09.016.

ZHANG K, ZHANG L, and YANG M H. Real-time compressive tracking[C]. European Conference on Computer Vision, Berlin, Germany, 2012: 864-877. doi: 10.1007/978-3- 642-33712-3_62.

EICHENSEER A and KAUP A. A data set providing synthetic and real-world fisheye video sequences[C]. International Conference on Acoustics, Speech and Signal Processing, Shanghai, China, 2016: 1541-1545. doi: 10.1109/ ICASSP.2016.7471935.

ZHANG K, ZHANG L, and YANG M H. Fast compressive tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014, 36(10): 2002-2015. doi: 10.1109/ TPAMI.2014.2315808.

ZHANG K, ZHANG L, YANG M H, et al. Fast tracking via spatio-temporal context learning[C]. Computer Vision and Pattern Recognition, Oregon, USA, 2013: 1-15.

POSSEGGER H, MAUTHNER T, and BISCHOF H. In defense of color-based model-free tracking[C]. Computer Vision and Pattern Recognition, Boston, USA, 2015: 2113-2120. doi: 10.1109/CVPR.2015.7298823.

EVERINGHAM M, GOOL L, WILLIAMS C K, et al. The pascal visual object classes (VOC) challenge[J]. International Journal of Computer Vision, 2010, 88(2): 303-338. doi: 10.1007/s11263-009-0275-4.

施引文獻

資源附件(0)

訪問統(tǒng)計