基于SVDD的三维目标多视点视图建模

丁昊; 李旭东; 赵慧洁

基于SVDD的三维目标多视点视图建模

北京航空航天大学精密光机电一体化技术教育部重点实验室, 北京 100191

基金项目: 国家自然科学基金资助项目(60802044)

详细信息

中图分类号: TP391
计量
- 文章访问数: 1677
- HTML全文浏览量: 93
- PDF下载量: 540
- 被引次数: 0
出版历程
- 收稿日期: 2011-07-18
- 网络出版日期: 2012-11-30

Method of multi-view modeling for 3D target based on SVDD

Precision Opto-mechatronics Technology Key Laboratory of Education Ministry, Beijing University of Aeronautics and Astronautics, Beijing 100191, China

摘要

摘要: 同一目标在不同观察视点下成像后外形可能有较大差异,因此三维目标多视点视图建模是目标识别的关键.针对该问题,提出了基于支持向量数据描述(SVDD, Support Vector Data Description)方法对目标特征进行描述.在视点球面上均匀采样获取目标全姿态图像,以SVDD方法求取在高维空间内包含尽可能多目标特征向量的最小超球体相关参数,得到数量较少的支持向量将作为目标多视点视图的最佳模型.对多类目标不同姿态的图像(每类2592帧),以规则化不变矩描述目标外形特征,进行了建模实验,并通过识别实验验证了所提方法的有效性和可行性.
- 多视点建模 /
- 支持向量数据描述 /
- 三维目标识别 /
- 规则化不变矩
Abstract: Popular 3D target recognition approaches based on image continue to struggle with challenge—viewpoint sensitivity. Multi-view modeling technique for 3D target offers promise for this challenge. A method of modeling 3D target based on support vector data description (SVDD) that could obtain a tight description covering most of the target feature data was proposed. Target’s images were captured with uniform grid on the viewing sphere and characterized by feature vectors. The support vectors representing characteristic views were obtained by applying SVDD to optimize the parameters of the minimal hyper-sphere which covers as many feature vectors as possible. The experiments were conducted by applying the proposed method to an image set (each target includes 2592 images) characterized by normalized moment invariants. The results show that the proposed method is effective and feasibility.
- multi-view modeling /
- support vector data description (SVDD) /
- 3D target recognition /
- normalized moment invariants

HTML全文

参考文献(19)

[1]	Tax D,Duin R.Support vector data description [J].Machine Learning,2004,54(1):45-66
[2]	彭辉,黄士科,陶琳,等.基于Zernike矩的三维目标多视点特性视图建模[J].红外与激光工程,2005,34(3):292-296 Peng Hui,Huang Shike,Tao Lin,et al.Multi-view modeling of 3D target based on Zernike moments [J].Infrared and Laser Engineering,2005,34(3):292-296(in Chinese)
[3]	Koenderink J,Doom A.The internal representation of solid shape with respect to vision [J].Biological Cybernetics,1979,32(4):211-216
[4]	Gigus Z,Malik J.Computing the aspect graph for line drawings of polyhedral objects [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1990,12(2):113-122
[5]	Bowyer K,Sallam M,Eggert D,et al.Computing the generalized aspect graph for objects with moving parts [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1993,15(6):605-610
[6]	Roy S,Chaudhury S,Banerjee S.Aspect graph construction with noisy feature detectors [J].IEEE Transactions on Systems,Man,and Cybernetics,Part B:Cybernetics,2003,33(2):340-351
[7]	Yu X,Ma H,You S,et al.A solution to efficient viewpoint space partition in 3D object recognition [C]//International Conference on Image and Graphics.Xi-an,Shaanxi,China:IEEE Computer Society,2009:565-570
[8]	Cyr C,Kimia B.3D object recognition using shape similarity-based aspect graph [C]// International Conference on Computer Vision.New York,USA:IEEE Computer Society,2001:254-261
[9]	Belongie S,Malik J,Puzicha J.Shape matching and object recognition using shape contexts [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(4):509-522
[10]	Pang B,Ma H.An effective way of 3D model representation in recognition system [C]// International Conference on Multimedia and Signal Processing.Guilin,Guangxi,China:IEEE Computer Society,2011:107-111
[11]	Tsang I W,Kwok J T,Cheung P M.Core vector machines:fast SVM training on very large data sets [J].Journal of Machine Learning Research,2005,6:363-392
[12]	Joachims T,Finley T,Yu C J.Cutting-plane training of structural SVMs [J].Machine Learning,2009,77(1):27-59
[13]	Dung D N,Matsumoto K,Takishima Y,et al.Condensed vector machines:learning fast machine for large data [J].IEEE Transactions on Neural Network,2010,21(12):1903-1914
[14]	张劲锋,蔡伟.基于组合不变矩的空间目标识别 [C]//全国第十二届空间及运动体控制技术学术会议论文集.桂林:中国自动化学会空间及运动体控制专业委员会,2006:490-494 Zhang Jinfeng,Cai Wei. Space object recognition based on combined moment invariants [C]//The 12th National Conference on Space and Vehicle Control Technology.Guilin:China Association of Automation, 2006:490-494(in Chinese)
[15]	Flusser J,Suk T.Pattern recognition by affine moment invariants [J].Pattern Recognition,1993,26(1):167-174
[16]	HU M K.Visual pattern recognition by moment invariants [J].IRE Transactions on Information Theory,1962,8:179-187
[17]	张天序.成像自动目标识别[M].武汉:湖北科学技术出版社,2005:18-84 Zhang Tianxu.Automated recognition of imaged targets [M].Wuhan:Hubei Science and Technology Press,2005:18-84(in Chinese)
[18]	Mu T,Nandi A K.Multiclass classification based on extended support vector data description [J].IEEE Transactions on Systems,Man,and Cybernetics,Part B:Cybernetics,2009,39(5):1206-1216
[19]	Yang L,Ma W,Tian B.New multi-class classification method based on the SVDD model [C]//International Symposium on Neural Networks.Guilin,Guangxi,China:Springer,2011:103-112