基于中智加权相似度量的尺度自适应视觉目标跟踪算法

doi:10.11959/j.issn.1000-0801.2018176

电信科学 ›› 2018, Vol. 34 ›› Issue (5): 50-62.doi: 10.11959/j.issn.1000-0801.2018176

基于中智加权相似度量的尺度自适应视觉目标跟踪算法

胡珂立¹(),范恩¹,叶军¹,沈士根¹,谷宇章²

¹ 绍兴文理学院，浙江绍兴 312000
² 中国科学院上海微系统与信息技术研究所，上海 200050

修回日期:2018-05-07 出版日期:2018-05-01 发布日期:2018-05-30
作者简介:胡珂立（1989-），男，博士，绍兴文理学院计算机科学与工程系讲师，主要研究方向为图像处理、目标跟踪、中智理论。|范恩（1982-），男，博士，绍兴文理学院计算机科学与工程系讲师，主要研究方向为传感器数据融合、目标跟踪、中智理论。|叶军（1959-），男，绍兴文理学院电子与信息工程系教授，主要研究方向为中智理论。|沈士根（1974-），男，博士，绍兴文理学院计算机科学与工程系教授，主要研究方向为无线传感器网络、移动互联网、博弈论。|谷宇章（1976-），男，博士，中科院上海微系统与信息技术研究所物联网系统技术实验室副研究员，主要研究方向为图像处理、仿生机器人视觉、立体视觉感知。

A scale adaptive visual object tracking algorithm based on weighted neutrosophic similarity coefficient

Keli HU¹(),En FAN¹,Jun YE¹,Shigen SHEN¹,Yuzhang GU²

¹ Shaoxing University,Shaoxing 312000,China
² Shanghai Institute of Microsystem and Information Technology,Chinese Academy of Sciences,Shanghai 200050,China

Revised:2018-05-07 Online:2018-05-01 Published:2018-05-30
Supported by:
浙江省公益性技术应用研究计划基金资助项目(61603258);The National Natural Science Foundation of China(61603258);浙江省公益性技术应用研究计划基金资助项目(61703280);The National Natural Science Foundation of China(61703280);浙江省公益性技术应用研究计划基金资助项目(2016C31082);The Public Welfare Technology Application Research Project of Zhejiang Province(2016C31082)

摘要/Abstract

摘要：

中智集理论是传统模糊理论的拓展，它能够表述现实生活中的不确定性信息。面对不同问题，中智框架下的真（truth）、不确定（indeterminacy）、假（falsity）隶属度所占权重可能不同。提出了一种分量加权的余弦相似度量，并将其引入均值漂移视觉跟踪算法中。首先基于3σ理论和目标/背景相似度两种属性提出了相应的真、不确定、假量测，然后利用加权余弦相似度量构建权值向量，同时提出了基于中智加权余弦相似度量的尺度更新算法，综合提升均值漂移跟踪性能。实验结果表明，提出的视觉跟踪算法能较好克服相似背景、光照变化、尺度变化等挑战。

关键词: 中智集, 加权相似度量, 目标跟踪

Abstract:

The weight of the truth,indeterminacy,and falsity membership under the neutrosophic framework may be different when dealing with different problems.Due to this,a component weighted cosine similarity coefficient was proposed,and it was introduced into the mean shift tracking algorithm.Firstly,the corresponding methods for calculating the membership of the truth,indeterminacy,and falsity were proposed based on the theory of 3σ,as well as the similarity between the features of the corresponding area of the object and background.Then the weighted cosine similarity coefficient was used to construct the weight vector.In addition,a weighted cosine similarity coefficient based scale updating method was proposed.The experimental results demonstrate that the modified visual tracking algorithm performs well,even when there exists challenges like similar background,illumination or scale variation.

Key words: neutrosophic set, weighted similarity coefficient, object tracking

中图分类号:

TP391

胡珂立,范恩,叶军,沈士根,谷宇章. 基于中智加权相似度量的尺度自适应视觉目标跟踪算法[J]. 电信科学, 2018, 34(5): 50-62.

Keli HU,En FAN,Jun YE,Shigen SHEN,Yuzhang GU. A scale adaptive visual object tracking algorithm based on weighted neutrosophic similarity coefficient[J]. Telecommunications Science, 2018, 34(5): 50-62.

图/表 13

表1

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

表2

表3

参考文献 31

[1]	赵利平, 林涛, 周开伦 ,等. 一种多色度格式级联编码的AVS2全色度图像编码算法[J]. 电信科学, 2018,34(4): 57-67.
	ZHAO L P , LIN T , ZHOU K L ,et al. A multi-chroma format cascaded coding method for full-chroma image in AVS2[J]. Telecommunications Science, 2018,34(4): 57-67.
[2]	唐彪, 金炜, 符冉迪 ,等. 多稀疏表示分类器决策融合的人脸识别[J]. 电信科学, 2018,34(4): 31-40.
	FU B , JIN W , FU R D ,et al. Face recognition using decision fusion of multiple sparse representation-based classifiers[J]. Telecommunications Science, 2018,34(4): 31-40.
[3]	颜文, 金炜, 符冉迪 . 结合 VLAD 特征和稀疏表示的图像检索[J]. 电信科学, 2016,32(12): 80-85.
	YAN W , JIN W , FU R D . Image retrieval based on the feature of VLAD and sparse representation[J]. Telecommunications Science, 2016,32(12): 80-85.
[4]	YILMAZ A , JAVED O , SHAH M . Object tracking:a survey[J]. ACM Computing Surveys, 2006,38(4):13.
[5]	WU Y , LIM J , YANG M H . Online object tracking:a benchmark[C]// IEEE Conference on Computer Vision and Pattern Recognition (CVPR),June 23-28,2013,Portland,OR,USA. Piscataway:IEEE Press, 2013: 2411-2418.
[6]	SMEULDERS A W M , CHU D M , CUCCHIARA R ,et al. Visual tracking:an experimental survey[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014,36(7): 1442-1468.
[7]	COMANICIU D , RAMESH V , MEER P . Real-time tracking of non-rigid objects using mean shift[C]// IEEE Conference on Computer Vision and Pattern Recognition (CVPR),June 13-15,2000,Hilton Head Island,SC,USA. Piscataway:IEEE Press, 2000: 142-149.
[8]	COMANICIU D , RAMESH V , MEER P . Kernel-based object tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003,25(5): 564-577.
[9]	LEICHTER I . Mean shift trackers with cross-bin metrics[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011,34(4): 695-706.
[10]	ZHOU H , YUAN Y , SHI C . Object tracking using SIFT features and mean shift[J]. Computer Vision and Image Understanding, 2009,113(3): 345-352.
[11]	BOUSETOUANE F , DIB L , SNOUSSI H . Improved mean shift integrating texture and color features for robust real time object tracking[J]. The Visual Computer, 2013,29(3): 155-170.
[12]	VOJIR T , NOSKOVA J , MATAS J . Robust scale-adaptive mean-shift for tracking[J]. Pattern Recognition Letters, 2014(49): 250-258.
[13]	COLLINS R T , . Mean-shift blob tracking through scale space[C]// IEEE Conference on Computer Vision Pattern Recognition (CVPR),June 16-22,2003,Madison,WI,USA. Piscataway:IEEE Press, 2003: 234-234.
[14]	SMARANDACHE F . Neutrosophy:neutrosophic probability,set and logic[M]. Rehoboth: American Research PressPress, 1998:105.
[15]	WANG H , SMARANDACHE F , ZHANG Y Q ,et al. Single valued neutrosophic sets[J]. Multispace and Multistructure, 2010(4): 410-413.
[16]	YE J . Vector similarity measures of simplified neutrosophic sets and their application in multicriteria decision making[J]. International Journal of Fuzzy Systems, 2014,16(2): 204-211.
[17]	GUO Y,?ENGüR A . A novel image segmentation algorithm based on neutrosophic similarity clustering[J]. Applied Soft Computing Journal, 2014(25): 391-398.
[18]	GUO Y , XIA R , ?ENGüR A ,et al. A novel image segmentation approach based on neutrosophic C-means clustering and indeterminacy filtering[J]. Neural Computing and Applications, 2016: 1-11.
[19]	HU K , YE J , FAN E ,et al. A novel object tracking algorithm by fusing color and depth information based on single valued neutrosophic cross-entropy[J]. Journal of Intelligent and Fuzzy Systems, 2017,32(3): 1775-1786.
[20]	GUO Y , SENGUR A . A novel 3D skeleton algorithm based on neutrosophic cost function[J]. Applied Soft Computing Journal, 2015(36): 210-217.
[21]	崔西希, 吴成茂 . 核空间中智模糊聚类及图像分割应用[J]. 中国图象图形学报, 2016,21(10): 1316-1327.
	CUI X X , WU C M . Neutrosophic C-means clustering in kernel space and its application in image segmentation[J]. Journal of Image and Graphics, 2016,21(10): 1316-1327.
[22]	张桂梅, 王大雷 . 结合 LPG＆PCA 的中智学图像分割[J]. 中国图象图形学报, 2014,19(5): 693-700.
	ZHANG G M , WANG D L . Neutrosophic image segmentation approach integrated LPG＆PCA[J]. Journal of Image and Graphics, 2014,19(5): 693-700.
[23]	郑肇葆, 潘励, 郑宏 . 中智逻辑图像分割方法的研究与分析[J]. 武汉大学学报(信息科学版), 2015,40(2): 143-146.
	ZHENG Z B , PAN L , ZHENG H . Research and analysis of neutrosophic logic image segmentation (NLIS) method[J]. Geomatics and Information Science of Wuhan University, 2015,40(2): 143-146.
[24]	YE J . Single-valued neutrosophic similarity measures based on cotangent function and their application in the fault diagnosis of steam turbine[J]. Soft Computing, 2015,21(3): 817-825.
[25]	MA Y X , WANG J Q , WANG J ,et al. An interval neutrosophic linguistic multi-criteria group decision-making method and its application in selecting medical treatment options[J]. Neural Computing and Applications, 2016: 1-21.
[26]	YE J , FU J . Multi-period medical diagnosis method using a single valued neutrosophic similarity measure based on tangent function[J]. Computer Methods ＆ Programs in Biomedicine, 2016(123): 142-149.
[27]	ANTER A M , HASSANIEN A E , ELSOUD M A A ,et al. Neutrosophic sets and fuzzy C-means clustering for improving CT liver image segmentation[J]. Advances in Intelligent Systems and Computing, 2014(303): 193-203.
[28]	GUO Y , SENGUR A . NECM:neutrosophic evidential C-means clustering algorithm[J]. Neural Computing and Applications, 2015,26(3): 561-571.
[29]	GUO Y , SENGUR A . NCM:Neutrosophic c-means clustering algorithm[J]. Pattern Recognition, 2015,48(8): 2710-2724.
[30]	HU K , FAN E , YE J ,et al. Neutrosophic similarity score based weighted histogram for robust mean-shift tracking[J]. Information, 2017,8(4):122.
[31]	YE J . Multicriteria decision-making method using the correlation coefficient under single-valued neutrosophic environment[J]. International Journal of General Systems, 2013,42(4): 386-394.

序列名称	跟踪目标	主要挑战	总帧数
David	人脸	IPR,OPR,BC	537
Singer	人	IV,SV,OCC,OPR	351
Bolt	人	DEF,BC	293
Walking	人	SV,OCC,DEF	412
Dog	狗	SV,DEF,OPR	127
Car24	车	IV,SV,BC	3 059
Car1	车	IV,SV,MB,FM,BC,LR	1 020
Car2	车	IV,SV,MB,FM,BC	913
Jump	人	SV,OCC,DEF,MB,FM,IPR,OPR	122
Coke	饮料瓶	IV,OCC,FM,IPR,OPR,BC	291
Lemming	玩偶	IV,SV,OCC,FM,OPR	1 336
Faceocc1	人脸	OCC	892
Faceocc2	人脸	IV,OCC,IPR,OPR	812
Jumping	人	MB,FM	313
Redteam	车	SV,OCC,IPR,OPR,LR	1 918

挑战	BC	FM	MB	DEF	IV	IPR	LR	OCC	OPR	SV	综合
WNEUTSMS	0.303	0.477	0.399	0.450	0.371	0.445	0.290	0.469	0.466	0.415	0.442
NEUTMS	0.279	0.485	0.452	0.459	0.354	0.435	0.253	0.461	0.460	0.387	0.434
ASMS	0.253	0.513	0.407	0.403	0.338	0.417	0.349	0.455	0.453	0.445	0.408
KMS	0.242	0.366	0.265	0.366	0.329	0.342	0.199	0.438	0.382	0.337	0.360

挑战	BC	FM	MB	DEF	IV	IPR	LR	OCC	OPR	SV	综合
WNEUTSMS	0.253	0.391	0.377	0.470	0.319	0.372	0.250	0.450	0.424	0.426	0.411
NEUTMS	0.187	0.390	0.325	0.363	0.248	0.349	0.279	0.379	0.374	0.311	0.366
ASMS	0.282	0.364	0.126	0.327	0.321	0.297	0.377	0.375	0.358	0.397	0.351
KMS	0.161	0.218	0.081	0.325	0.265	0.238	0.087	0.397	0.298	0.257	0.291

基于中智加权相似度量的尺度自适应视觉目标跟踪算法

A scale adaptive visual object tracking algorithm based on weighted neutrosophic similarity coefficient

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献 31

相关文章 3

Metrics

推荐阅读 0

[1]	刘述木,杨建,黎远松. WSN中基于自适应预测聚类的多组群目标的跟踪方法[J]. 电信科学, 2016, 32(7): 68-75.
[2]	朱金芳,张晓莺,范一鸣. 基于APF的二元WSN运动目标跟踪算法[J]. 电信科学, 2012, 28(5): 109-114.
[3]	朱向军,毛剑飞,鲍卫兵. 基于达芬奇技术的智能视频监控终端的开发[J]. 电信科学, 2011, 27(2): 100-104.