全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

密集特征加权跟踪算法

DOI: 10.11834/jig.20150509

Keywords: 目标跟踪,尺度自适应,密集特征加权,Mean-shift

Full-Text   Cite this paper   Add to My Lib

Abstract:

目的当前大多数基于Mean-shift的跟踪算法都忽视了目标中密集的特征信息,本文有效利用密集特征信息,来提高跟踪的准确性.方法在目标模型中,常存在一些颜色特征相对聚集,形成一定大小的特征密集区,这些区域的面积或大或小,对人眼视觉跟踪异常重要.这些区域形成的空间结构信息,可以被利用到目标跟踪.提出一种高效的目标模型,通过计算密集特征区域面积,以及密集区质心到目标中心的距离,构建加权系数,通过该系数,来增加目标中分布相对集中的特征的权值,同时削弱离散特征的权值.同时使用零阶矩和目标模型与候选模型之间的相似度系数,估算目标的面积;再使用预测目标面积补偿法,对目标中因使用背景加权法而权重被削弱的特征区域,进行面积补偿;最后使用估算的目标区域面积以及二阶中心距,估算目标尺度和方向的改变.在跟踪过程中,背景如发生较大变化,则对目标模型进行更新.结果本文算法具有很好的尺度适应性,跟踪平均准确率在94.6%以上,得到较当前一些先进方法更好的准确度和效率.结论提出的算法能增加目标模型中不同特征权值间的差异,使得构建的目标模型具有较强区分目标和背景的能力,提高了定位目标的准确性;面积补偿法解决了目标因特征权重被削弱,而导致估算的目标面积小于实际面积的问题.

References

[1]  Kalal Z, Mikolajczyk K, Matas J. Forward-backward error: automatic detection of tracking failures [C]// Proceeding of the 20th International Conference on Computer Vision and Pattern Recognition. Istanbul: IEEE Computer Society, 2010: 2756-2759.
[2]  Kalal Z, Mikolajczyk K, Matas J. Tracking-learning-detection [J]. IEEE Transactions on Pattern Analysis And Machine Intelligence, 2012, 34(7): 1409-1422.
[3]  Zhang K, Zhang L, Yang M H. Real-time compressive tracking[M]//Computer Vision-ECCV 2012. Berlin: Springer, 2012: 864-877.
[4]  Zha Y, Cao T Y, Huang H, et al. Object tracking based on learning a dictionary joint practical filter[J]. Journal of Image and Graphics, 2013, 18(12): 1628-1636[查绎, 曹铁勇, 黄辉, 等. 字典学习联合粒子滤波鲁棒跟踪[J]. 中国图象图形学报, 2013, 18(12): 1628-1636.] [DOI:10.11834/jig.20131211]
[5]  Zhang Q R, Feng X Y. Object tracking based onvisual saliency and particle filter[J]. Journal of Image and Graphics, 2013, 18(5): 515-522[张巧荣, 冯新扬. 利用视觉显著性和粒子滤波的运动目标跟踪[J]. 中国图象图形学报, 2013, 18(5): 515-522.] [DOI:10.11834/jig.20130504]
[6]  Fukunaga K, Hostetler L. The estimation of the gradient of a density function, with applications in pattern recognition[J]. IEEE Transactions on Information Theory, 1975, 21(1): 32-40.
[7]  Cheng Y. Mean shift, mode seeking, and clustering[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1995, 17(8): 790-799.
[8]  Comaniciu D, Ramesh V, Meer P. Real-time tracking of non-id objects using mean shift[C]// Proceedings of Computer Vision and Pattern Recognition, Hilton Head Island,SC: IEEE, 2000, 2: 142-149.
[9]  Comaniciu D, Ramesh V. Mean shift and optimal prediction for efficient object tracking[C]// Proceedings of Image Processing, Vancouver, BC: IEEE, 2000, 3: 70-73.
[10]  Comaniciu D, Meer P. Mean shift: a robust approach toward feature space analysis[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(5): 603-619.
[11]  Vojir T, Noskova J, Matas J. Robust Scale-Adaptive Mean-Shift for Tracking[M]//Image Analysis. Berlin Heidelberg: Springer, 2013: 652-663.
[12]  Comaniciu D, Ramesh V, Meer P. Kernel-based object tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(5): 564-577.
[13]  Jeyakar J, Babu R V, Ramakrishnan K R. Robust object tra-cking with background-weighted local kernels[J]. Computer Vision and Image Understanding, 2008, 112(3): 296-309.
[14]  Li L, Feng Z. An efficient object tracking method based on adaptive nonparametric approach [J]. Optoelectronics Review, 2005, 13(4): 325-330.
[15]  Allen J G, Xu R Y D, Jin J S. Mean shift object tracking for a SIMD computer[C]//Proceedings of the 3rd International Conference on Information Technology and Applications. Washington DC: IEEE, 2005, 1: 692-697.
[16]  Ning J, Zhang L, Zhang D, et al. Robust mean-shift tracking with corrected background-weighted histogram[J]. Computer Vision, IET, 2012, 6(1): 62-69.
[17]  Ning J, Zhang L, Zhang D, et al. Scale and orientation adaptive mean shift tracking[J]. Computer Vision, IET, 2012, 6(1): 52-61.
[18]  Bretzner L, Lindeberg T. Qualitative multi-scale feature hierarchies for object tracking[M]//Scale-Space Theories in Computer Vision. Berlin Heidelberg: Springer, 1999: 117-128.
[19]  Godec M, Roth P M, Bischof H. Hough-based tracking of non-rigid objects[J]. Computer Vision and Image Understanding, 2013, 117(10): 1245-1256.
[20]  Cehovin L, Kristan M, Leonardis A. An adaptive coupled-layer visual model for robust visual tracking [C]// Proceeding of IEEE International Conference on Computer Vision. Barcelona: IEEE, 2011: 1363-1370.
[21]  更多...
[22]  Wang S, Lu H, Yang F. et al. Superpixel tracking[C]//Proceedings of the 13th International Conference on Computer Vision. Barcelona, Spain: IEEE, 2011: 1323-1330.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133