Comparative study of vision tracking methods for tracking of construction site resources

被引:86
作者
Park, Man-Woo [1 ]
Makhmalbaf, Atefe [2 ]
Brilakis, Ioannis [1 ]
机构
[1] Georgia Inst Technol, Sch Civil & Env Engn, Atlanta, GA 30332 USA
[2] Georgia Inst Technol, Coll Architecture, Atlanta, GA 30332 USA
基金
美国国家科学基金会;
关键词
Comparative studies; Tracking; Remote sensing; Image processing; Computer vision; ALGORITHM; CONTOURS;
D O I
10.1016/j.autcon.2011.03.007
中图分类号
TU [建筑科学];
学科分类号
0813 ;
摘要
Vision tracking has significant potential for tracking resources on large scale, congested construction sites, where a small number of cameras strategically placed around the site could replace hundreds of tracking tags. The correlation of vision tracking 2D positions from multiple views can provide the 3D position. However. there are many 2D vision trackers available in the literature, and little information is available on which one is most effective for construction applications. In this paper, a comparative study of various vision tracker categories is carried out, to identify which one is most effective in tracking construction resources. Testing parameters for evaluating categories of trackers are identified, and benefits and limitations of each category are presented. The most promising trackers are tested using a database of construction operations videos. The results indicate the effectiveness of each tracker in relation to each parameter of the test, and the most suitable tracker needed to research effective 3D vision trackers of construction resources. (C) 2011 Elsevier B.V. All rights reserved.
引用
收藏
页码:905 / 915
页数:11
相关论文
共 28 条
[1]  
[Anonymous], INT J IMAGE GRAPHICS
[2]  
[Anonymous], 2005, P IEEE INT C IM PROC
[3]  
[Anonymous], CORRELATING MULTIPLE
[4]  
Brilakis I., 2008, 15 EG ICE WORKSH 200
[5]   Construction site image retrieval based on material cluster recognition [J].
Brilakis, Ioannis K. ;
Soibelman, Lucio ;
Shinagawa, Yoshihisa .
ADVANCED ENGINEERING INFORMATICS, 2006, 20 (04) :443-452
[6]  
Caldas C.H., 2004, P ISARC 2004 IAARC J
[7]   Kernel-based object tracking [J].
Comaniciu, D ;
Ramesh, V ;
Meer, P .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2003, 25 (05) :564-577
[8]   An efficient implementation of Reid's multiple hypothesis tracking algorithm and its evaluation for the purpose of visual tracking [J].
Cox, IJ ;
Hingorani, SL .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1996, 18 (02) :138-150
[9]   Active contours for tracking distributions [J].
Freedman, D ;
Zhang, T .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2004, 13 (04) :518-526
[10]   CONDENSATION - Conditional density propagation for visual tracking [J].
Isard, M ;
Blake, A .
INTERNATIONAL JOURNAL OF COMPUTER VISION, 1998, 29 (01) :5-28