基于二階空間直方圖的雙核跟蹤
doi: 10.11999/JEIT141321
-
1.
(廣西師范大學(xué)計(jì)算機(jī)科學(xué)與信息工程學(xué)院 桂林 541004) ②(桂林電子科技大學(xué) 桂林 541004)
國(guó)家自然科學(xué)基金(61365009, 61165009),廣西自然科學(xué)基金(2014GXNSFAA118368, 2012GXNSFAA053219)和廣西高校科技項(xiàng)目(2013YB027)
Dual-kernel Tracking Approach Based on Second-order Spatiogram
-
2.
(Guangxi Experiment Center of Information Science, Guilin University of Electronic Technology, Guilin 541004, China)
-
摘要: 針對(duì)核跟蹤算法中的背景信息棄用和空間結(jié)構(gòu)丟失問(wèn)題,該文提出一種基于二階空間直方圖的雙核式目標(biāo)跟蹤算法。該算法以二階空間直方圖為目標(biāo)表示模型,以相似度和對(duì)比度為目標(biāo)判斷準(zhǔn)則,來(lái)建立全新的目標(biāo)函數(shù);并依據(jù)多變量泰勒展開(kāi)和目標(biāo)函數(shù)最大化方法,推導(dǎo)出雙核式目標(biāo)位移公式;最后使用均值漂移程序遞歸地獲得了目標(biāo)的最優(yōu)位置。通過(guò)對(duì)各種條件下運(yùn)動(dòng)目標(biāo)的跟蹤驗(yàn)證了算法的有效性。
-
關(guān)鍵詞:
- 模式識(shí)別 /
- 雙核跟蹤 /
- 空間直方圖 /
- 延森-香農(nóng)散度
Abstract: In order to avoid the loss of background and spatial information in mean shift tracker, a dual-kernel tracking approach based on the second-order spatiogram is proposed. In the method, the second-order spatiogram is employed to represent a target, the similarity and contrast are considered simultaneously when evaluating the target candidate, and they are adaptively integrated into a novel objective function. By performing multi-variable Taylor series expansion and maximization on the objective function, a dual-kernel target location-shift formula is induced. Finally, the optimal target location is gained recursively by applying the mean shift procedure. Experimental evaluations on several image sequences demonstrate the effectiveness of the proposed algorithm.-
Key words:
- Pattern recognition /
- Dual-kernel tracking /
- Spatiogram /
- Jensen-Shannon divergence
-
Zhang S, Yao H, Sun X, et al.. Sparse coding based visual tracking: review and experimental comparison[J]. Pattern Recognition, 2013, 46(7): 1772-1788. Wu Y, Lim J, and Yang M. Online object tracking: a benchmark[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Oregon, Portland, USA, 2013: 2411-2418. Isard M and Black A. Condensation-conditional density propagation for visual tracking[J]. International Journal on Computer Vision, 1998, 29(1): 5-28. 程旭, 李擬珺, 周同池, 等. 稀疏表示的超像素在線(xiàn)跟蹤[J]. 電子與信息學(xué)報(bào), 2014, 36(10): 2393-2399. Cheng Xu, Li Ni-jun, Zhou Tong-chi, et al.. Online tracking via superpixel and sparse representation[J]. Journal of Electronics Information Technology, 2014, 36(10): 2393-2399. Comaniciu D, Ramesh V, and Meer P. Kernel-based object tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(5): 564-577. Leichter I. Mean shift trackers with cross-bin metrics[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(4): 695-706. Avidan S. Support vector tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(8): 1064-1072. Zhang K and Song H. Real-time visual tracking via online weighted multiple instance learning[J]. Pattern Recognition, 2013, 46(1): 397-411. Ross D, Lim J, Lin R, et al.. Incremental learning for robust visual tracking[J]. International Journal of Computer Vision, 2008, 77(3): 125-141. Mei X and Ling H. Robust visual tracking using l1 minimization[C]. Proceedings of the International Conference on Computer Vision, Kyoto, Japan, 2009: 1436-1443. Fouad B, Lynda D, and Hichem S. Improved mean shift integrating texture and color features for robust real time object tracking[J]. The Visual Computer, 2013, 29(3): 155-170. Tomas V, Jana N, and Jiri M. Robust scale-adaptive mean- shift for tracking[J]. Lecture Notes in Computer Science, 2013 (7944): 652-663. Birchfield S and Rangarajan S. Spatiograms versus histograms for region-based tracking[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, SanDiego, California, USA, 2005: 1158-1163. Collins R, Liu Y, and Leordeanum M. Online selection of discriminative tracking features[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(10): 1631-1643. Ning J, Zhang L, Zhang D, et al.. Robust mean shift tracking with corrected background-weighted histogram[J]. IET Computer Vision, 2012, 6(1): 62-69. Conaire C, O'Connor N, and Smeaton A. An improved spatiogram similarity measure for robust object localization [C]. Proceedings of International Conference on Acoustics, Speech, and Signal Processing, Hawaii, USA, 2007: 1069-1072. Lin J. Divergence measures based on the shannon entropy[J]. IEEE Transactions on Information Theory, 1991, 37(1): 145-151. Liu Y, Tong S, and Chen C. Adaptive fuzzy control via observer design for uncertain nonlinear systems with unmodeled dynamics[J]. IEEE Transactions on Fuzzy Systems, 2013, 21(2): 275-288. Zhang C, Jing Z, and Jin B. A dual-kernel-based tracking approach for visual target[J]. SCIENCE CHINA: Information Sciences, 2012, 55(3): 566-576. -
計(jì)量
- 文章訪(fǎng)問(wèn)數(shù): 1161
- HTML全文瀏覽量: 108
- PDF下載量: 562
- 被引次數(shù): 0