一種目標(biāo)響應(yīng)自適應(yīng)的通道可靠性跟蹤算法

王鵬; 孫夢宇; 王海燕; 李曉艷; 呂志剛

doi:10.11999/JEIT190569

一種目標(biāo)響應(yīng)自適應(yīng)的通道可靠性跟蹤算法

doi: 10.11999/JEIT190569 cstr: 32379.14.JEIT190569

王鵬^{1, 2, 3},
孫夢宇^2, ,,
王海燕⁴,
李曉艷²,
呂志剛²

1.
西北工業(yè)大學(xué)航海學(xué)院西安 710072
2.
西安工業(yè)大學(xué)電子信息工程學(xué)院西安 710021
3.
海洋聲學(xué)信息感知工業(yè)和信息化部重點(diǎn)實(shí)驗(yàn)室(西北工業(yè)大學(xué)) 西安 710072
4.
陜西科技大學(xué)電子信息與人工智能學(xué)院西安 710021

基金項(xiàng)目: 國家自然科學(xué)基金(61271362)，國家重點(diǎn)研發(fā)計(jì)劃(2016YFC1400200)，陜西省科技廳重點(diǎn)研發(fā)計(jì)劃(2019GY-022、2019GY-066), 2019年西安市未央?yún)^(qū)科技計(jì)劃項(xiàng)目(201923)

詳細(xì)信息

作者簡介:
王鵬：男，1978年生，教授，研究方向?yàn)闄C(jī)器視覺、模式識別、圖像處理

孫夢宇：男，1993年生，碩士生，研究方向?yàn)槟繕?biāo)跟蹤

王海燕：男，1965年生，教授，研究方向?yàn)楝F(xiàn)代信號檢測與現(xiàn)代信息處理

李曉艷：女，1982年生，講師，研究方向?yàn)槟繕?biāo)檢測、目標(biāo)識別

呂志剛：男，1978年生，副教授，研究方向?yàn)槟Ｊ阶R別

通訊作者:
孫夢宇　1215200684@qq.com

中圖分類號: TN911.73; TP391
計(jì)量
- 文章訪問數(shù): 3699
- HTML全文瀏覽量: 1181
- PDF下載量: 99
- 被引次數(shù): 0
出版歷程
- 收稿日期: 2019-07-29
- 修回日期: 2020-03-25
- 網(wǎng)絡(luò)出版日期: 2020-04-03
- 刊出日期: 2020-08-18

An Object Tracking Algorithm with Channel Reliability and Target Response Adaptation

Peng WANG^{1, 2, 3},
Mengyu SUN^{2
, ,},
Haiyan WANG⁴,
Xiaoyan LI²,
Zhigang Lü²

1.
School of Marine Science and Technology, Northwestern Polytechnical University, Xi’an 710072, China
2.
School of Electronic Information Engineering, Xi’an Technological University, Xi’an 710021, China
3.
Key Laboratory of Ocean Acoustics and Sensing (Northwestern Polytechnical University), Ministry of Industry and Information Technology, Xi’an 710072, China
4.
School of Electronic Information and Artificial Intelligence, Shaanxi University of Science and Technology, Xi’an 710021, China

Funds: The National Natural Science Foundation of China (61271362),The National Key Research and Development Project (2016YFC1400200), The Key Science and Technology Program of Shaanxi Province (2019GY-022, 2019GY-066), Weiyang District of Xi’an 2019 Science and Technology Program (201923)

摘要

摘要:
為解決基于時(shí)空正則項(xiàng)的目標(biāo)跟蹤算法(STRCF)在目標(biāo)短時(shí)遮擋時(shí)定位精度低和目標(biāo)旋轉(zhuǎn)時(shí)尺度估計(jì)不準(zhǔn)確的問題，該文提出了一種目標(biāo)響應(yīng)自適應(yīng)的通道可靠性跟蹤算法。該算法在目標(biāo)模型訓(xùn)練時(shí)加入了目標(biāo)響應(yīng)正則項(xiàng)，通過在求解過程中更新理想目標(biāo)響應(yīng)函數(shù)，使得目標(biāo)被短時(shí)遮擋后可重新跟蹤目標(biāo)；加入通道可靠性評價(jià)各特征通道的可靠性，提高了模型對目標(biāo)的表達(dá)能力；將目標(biāo)圖像轉(zhuǎn)換至對數(shù)極坐標(biāo)系下訓(xùn)練尺度濾波器，提高在目標(biāo)旋轉(zhuǎn)時(shí)的尺度估計(jì)精度。實(shí)驗(yàn)結(jié)果表明，該文所提算法較STRCF在平均中心位置誤差中降低了28.54個(gè)像素，在平均重疊率中提高了22.8%，在OTB2015數(shù)據(jù)集下成功率曲線下面積較STRCF提高了1.5%。
- 目標(biāo)跟蹤 /
- 相關(guān)濾波 /
- 目標(biāo)響應(yīng)自適應(yīng) /
- 通道可靠性 /
- 尺度濾波
Abstract:
In order to solve the problems of lower precision of target location in short-term occlusion and inaccurate of scale estimation of target in rotation by Spatial-Temporal Regularized Correlation Filters (STRCF), an object tracking algorithm with channel reliability and target response adaptation is proposed in this paper. In this algorithm, target response regularization is added to train target model. By updating the ideal target response function in the process of solving model, the target can be tracked again after being occluded for a short time. The reliability of each feature channel is evaluated by coefficient of channel reliability, which can improves the model's expression of the target. Scale filters can be trained in log-polar coordinates to improve the accuracy of scale estimation when target is rotating. The experimental results show that the proposed algorithm reduces 28.54 pixels in the average center position error and improves the average overlap rate by 22.8% compared with STRCF.
- Object tracking /
- Correlation filter /
- Target response adaptation /
- Channel reliability /
- Scale correlation filter

HTML全文

圖 1 本文算法流程

下載: 全尺寸圖片幻燈片

圖 2 STRCF與本文算法所得的響應(yīng)圖對比

下載: 全尺寸圖片幻燈片

圖 3 4種算法在5個(gè)視頻序列下的測試結(jié)果

下載: 全尺寸圖片幻燈片

圖 4 5個(gè)視頻序列中心位置誤差跟蹤結(jié)果圖

下載: 全尺寸圖片幻燈片

圖 5 5個(gè)視頻序列重疊率結(jié)果圖

下載: 全尺寸圖片幻燈片

圖 6 OTB2015數(shù)據(jù)集測試結(jié)果

下載: 全尺寸圖片幻燈片

表 1 視頻序列名稱及屬性

視頻序列名稱	視頻屬性
MountainBike	旋轉(zhuǎn)、背景干擾
Basketball	遮擋、旋轉(zhuǎn)、光照變化、形變、背景干擾
Panda	遮擋、旋轉(zhuǎn)、尺寸變化、形變、超出視野、低分辨率
Girl2	遮擋、旋轉(zhuǎn)、尺寸變化、形變、快速移動(dòng)
KiteSurf	遮擋、旋轉(zhuǎn)、光照變化

下載: 導(dǎo)出CSV

表 2 平均中心位置誤差(像素)/平均重疊率

算法名稱	MoutainBike	Basketball	Panda	Girl2	KiteSurf
SAMF-AT	8.85/0.67	22.91/0.49	36.15/0.23	99.65/0.35	58.09/0.35
SRDCF	9.30/0.67	10.08/0.57	45.06/0.17	182.33/0.21	59.18/0.35
STRCF	10.42/0.65	14.06/0.36	10.12/0.39	77.86/0.48	66.73/0.36
本文算法	9.98/0.68	6.30/0.75	7.00/0.51	10.51/0.70	2.70/0.74

下載: 導(dǎo)出CSV

參考文獻(xiàn)(15)

王旭東, 王屹煒, 閆賀. 背景抑制直方圖模型的連續(xù)自適應(yīng)均值漂移跟蹤算法[J]. 電子與信息學(xué)報(bào), 2019, 41(6): 1480–1487. doi: 10.11999/JEIT180588

WANG Xudong, WANG Yiwei, and YAN He. Continuously adaptive mean-shift tracking algorithm with suppressed background histogram model[J]. Journal of Electronics &Information Technology, 2019, 41(6): 1480–1487. doi: 10.11999/JEIT180588

黃立勤, 朱飄. 車載視頻下改進(jìn)的核相關(guān)濾波跟蹤算法[J]. 電子與信息學(xué)報(bào), 2018, 40(8): 1887–1894. doi: 10.11999/JEIT171109

HUANG Liqin and ZHU Piao. Improved kernel correlation filtering tracking for vehicle video[J]. Journal of Electronics &Information Technology, 2018, 40(8): 1887–1894. doi: 10.11999/JEIT171109

LI Yang and ZHU Jianke. A scale adaptive kernel correlation filter tracker with feature integration[C]. European Conference on Computer Vision, Zurich, Switzerland, 2014: 254–265. doi: 10.1007/978-3-319-16181-5_18.

DANELLJAN M, KHAN F S, FELSBERG M, et al. Adaptive color attributes for real-time visual tracking[C]. 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, USA, 2014: 1090–1097. doi: 10.1109/CVPR.2014.143.

DANELLJAN M, H?GER G, KHAN F S, et al. Learning spatially regularized correlation filters for visual tracking[C]. 2015 International Conference on Computer Vision, Santiago, Chile, 2015: 4310–4318. doi: 10.1109/iccv.2015.490.

BIBI A, MUELLER M, and GHANEM B. Target response adaptation for correlation filter tracking[C]. The 14th European Conference on Computer Vision, Amsterdam, The Netherlands, 2016: 419–433. doi: 10.1007/978-3-319-46466-4_25.

WANG Ning, ZHOU Wengang, TIAN Qi, et al. Multi-cue correlation filters for robust visual tracking[C]. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, 2018: 4844–4853.doi: 10.1109/CVPR.2018.00509.

TANG Ming, YU Bin, ZHANG Fan, et al. High-speed tracking with multi-kernel correlation filters[C]. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, 2018: 4874–4833. doi: 10.1109/CVPR.2018.00512.

CHOI J, CHANG H J, FISCHER T, et al. Context-aware deep feature compression for high-speed visual tracking[C]. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, 2018: 479–488. doi: 10.1109/CVPR.2018.00057.

LI Feng, TIAN Cheng, ZUO Wangmeng, et al. Learning spatial-temporal regularized correlation filters for visual tracking[C]. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, 2018: 4904–4913. doi: 10.1109/CVPR.2018.00515.

BOLME D S, BEVERIDGE J R, DRAPER B A, et al. Visual object tracking using adaptive correlation filters[C]. 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, San Francisco, USA, 2010: 2544–2550. doi: 10.1109/cvpr.2010.5539960.

LUKE?I? A, VOJí? T, ZAJC L C, et al. Discriminative correlation filter tracker with channel and spatial reliability[J]. International Journal of Computer Vision, 2018, 126(7): 671–688. doi: 10.1007/s11263-017-1061-3

LI Yang, ZHU Jianke, HOI S C H, et al. Robust estimation of similarity transformation for visual object tracking[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2019, 33(1): 8666–8673. doi: 10.1609/aaai.v33i01.33018666

WU Yi, LIM J, and YANG M H. Object tracking benchmark[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(9): 1834–1848. doi: 10.1109/TPAMI.2014.2388226

WU Yi, LIM J, and YANG M H. Online object tracking: A benchmark[C]. 2013 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)