基于粒子濾波與樣本加權(quán)的壓縮跟蹤算法

張紅穎; 王賽男; 胡文博

doi:10.11999/JEIT170854

基于粒子濾波與樣本加權(quán)的壓縮跟蹤算法

doi: 10.11999/JEIT170854

基金項(xiàng)目:

天津市自然科學(xué)基金青年基金(12JCQNJC00600)，中央高校基本科研業(yè)務(wù)費(fèi)(3122015C016)，國家自然科學(xué)基金民航聯(lián)合研究基金(U1533203)

計(jì)量
- 文章訪問數(shù): 1366
- HTML全文瀏覽量: 104
- PDF下載量: 126
- 被引次數(shù): 0
出版歷程
- 收稿日期: 2017-09-07
- 修回日期: 2018-01-31
- 刊出日期: 2018-06-19

Compressive Tracking Algorithm Based on Particle Filter and Sample Weighting

Funds:

The Natural Science Foundation of Tianjin (12JCQNJC00600), The Fundamental Research Funds for the Central Universities (3122015C016), The National Natural Science Foundation of China (U1533203)

摘要

摘要: 該文針對壓縮跟蹤算法無法適應(yīng)目標(biāo)尺度的變化以及沒有考慮樣本權(quán)重的問題，提出一種基于粒子濾波與樣本加權(quán)的壓縮跟蹤算法。首先，對壓縮特征進(jìn)行改進(jìn)，提取歸一化矩形特征用于構(gòu)建目標(biāo)表觀模型。然后，引入樣本加權(quán)的思想，根據(jù)正樣本與目標(biāo)之間距離的不同賦予正樣本不同的權(quán)重，提高分類器的分類精度。最后，在粒子濾波的框架下融合尺度不變壓縮特征進(jìn)行動態(tài)狀態(tài)估計(jì)，在粒子預(yù)測階段利用2階自回歸模型對粒子狀態(tài)進(jìn)行估計(jì)與預(yù)測，借助觀測模型對粒子狀態(tài)進(jìn)行更新，并且對粒子進(jìn)行重采樣以防止粒子退化。實(shí)驗(yàn)結(jié)果表明，相比于原始壓縮跟蹤算法，改進(jìn)算法能夠更好地跟蹤目標(biāo)尺度的變化，提高跟蹤的穩(wěn)定性和準(zhǔn)確性。
- 壓縮跟蹤 /
- 粒子濾波 /
- 樣本加權(quán) /
- 分類器
Abstract: To solve the problem that Compressive Tracking (CT) algorithm is unable to adapt to the scale change of the object and ignores the sample weight, an optimized compressive tracking algorithm based on particle filter and sample weighting is presented. Firstly, the compressive feature is improved for building a target apparent model with normalized rectangle features. Then, the thought of sample weighting is utilized. In order to increase the precision of the classifier, different weights are given to the positive samples in accordance with the different distances between the positive samples and the object. Finally, the dynamic state estimation is made under the particle filter frame with integrating the scale invariant feature. At the phase of particle prediction, a second-order autoregressive model is utilized to obtain the estimation and prediction of the particle state. The particle state is updated with the observation model. The particles resampling is used to prevent the degradation of particles. Experimental results demonstrate that the improved algorithm can adapt to the scale change of object, and the accuracy and stability of the compressive tracking algorithm is improved.
- Compressive Tracking (CT) /
- Particle filter /
- Sample weighting /
- Classifier

HTML全文

參考文獻(xiàn)(19)

黎萬義, 王鵬, 喬紅. 引入視覺注意機(jī)制的目標(biāo)跟蹤方法綜述[J]. 自動化學(xué)報(bào), 2013, 40(4): 561-576. doi: 10.3724/SP.J. 1004.2014.00561.

LI Wanyi, WANG Peng, and QIAO Hong. A survey of visual attention based methods for object tracking[J]. Acta Automatica Sinica, 2013, 40(4): 561-576. doi: 10.3724/SP.J. 1004.2014.00561.

MEI Xue and LING Haibin. Robust visual tracking and vehicle classification via spare representation[J]. IEEE Transactions on Pattern Analysis Machine Intelligence, 2011, 33(11): 2259-2272. doi: 10.1109/TPAMI.2011.66.

BABENKO B, YANG Minghsuan, and BELONGIE S. Visual tracking with online multiple instance learning[J]. IEEE Transactions on Pattern Analysis Machine Intelligence, 2009, 33(8): 983-990. doi: 10.1109/CVPR.2009.5206737.

KALAL Z, MIKOLAJCZYK K, and MATAS J. Tracking- learning-detection[J]. IEEE Transactions on Pattern Analysis Machine Intelligence, 2012, 34(7): 1409-1422. doi: 10.1109/TPAMI.2011.239.

MAO Jiansen and QU Yufu. Tracking of variable scale object based on compressive sensing[J]. Chinese Journal of Liquid Crystals Displays, 2016, 31(6): 497-505. doi: 10.3788/ YJYXS20163105.0497.

ESLAHI N and AGHAGOLZADEH A. Compressive sensing image restoration using adaptive curvelet thresholding and nonlocal sparse regularization[J]. IEEE Transactions on Image Processing, 2016, 25(7): 3126-3140. doi: 10.1109/TIP. 2016.2562563.

ZHANG Kaihua, ZHANG Lei, and YANG Minghsuan. Real-time compressive tracking[C]. Proceedings of the 12th European Conference on Computer Vision, Florence, Italy, 2012: 864-877. doi: 10.1007/978-3-642-33712-3_62.

ZHANG Lei, WANG Yanjie, and HE Shuwen. Real-time compressive tracking method based on phase congruency[J]. Acta Photonica Sinica, 2014, 43(8): 124-131 doi: 10.3788/ gzxb20144308.0810003.

鄭超, 陳杰, 殷松峰, 等. 改進(jìn)的協(xié)同訓(xùn)練框架下壓縮跟蹤[J]. 電子與信息學(xué)報(bào), 2016, 38(7): 1624-1630. doi: 10.11999/ JEIT151001.

ZHENG Chao, CHEN Jie, YIN Songfeng, et al. Optimized compressive tracking in co-training framework[J]. Journal of Electronics Information Technology, 2016, 38(7): 1624-1630. doi: 10.11999/JEIT151001.

崔燦, 王民鋼, 李立, 等. 改進(jìn)壓縮特征的實(shí)時(shí)壓縮跟蹤算法[J]. 計(jì)算機(jī)工程與應(yīng)用, 2017, 53(15): 210-216. doi: 10.3778/ j.issn.1002-8331.1603-0060.

CUI Can, WANG Mingang, LI Li, et al. Real-time compressive tracking with advanced compression features[J]. Computer Engineering and Applications, 2017, 53(15): 210-216. doi: 10.3778/j.issn.1002-8331.1603-0060.

GHIRMIA T. Distributed particle filter for object tracking: with reduced sensor communications[J]. Sensors, 2016, 16(9): 1454-1464. doi: 10.3390/s16091454.

ADAM A, RIVLIN E, and SHIMSHONI I. Robust fragments- based tracking using the integral histogram[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. New York, USA, 2006: 798-805. doi: 10.1109/ CVPR.2006.256.

NING J, SHI W, YANG S, et al. Visual tracking based on distribution fields and online weighted multiple instance learning[J]. Image Vision Computing, 2013, 31(11): 853-863. doi: 10.1016/j.imavis.2013.09.003.

ORON S, BAARHILLEL A, and LEVI D. Locally orderless tracking[J]. International Journal of Computer Vision, 2015, 111(2): 213-228. doi: 10.1109/CVPR.2012.6247895.

GRABNER H, GRABNER M, and BISCHOF H. Real-time tracking via on-line boosting[C]. Proceedings of the British Machine Vision Conference, Edinburgh, UK, 2006, 1: 47-56. doi: 10.5244/C.20.6.

WU Y, LIM J, and YANG M H. Online object tracking: A benchmark [C]. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Portland, USA, 2013: 2411-2418. doi: 10.1109/CVPR.2013.312.

相關(guān)文章

施引文獻(xiàn)

資源附件(0)

訪問統(tǒng)計(jì)