全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

基于支撑描述的SIFT匹配方法

DOI: 10.13700/j.bh.1001-5965.2013.0382, PP. 685-689

Full-Text   Cite this paper   Add to My Lib

Abstract:

为减少局部结构相似等因素导致的图像匹配错误,提出一种基于支持描述的匹配判定方法.利用尺度不变特征变换(SIFT,ScaleInvariantFeatureTransform)算法获得初始匹配集,选取其中稳定性较高的特征点对建立支撑特征集;根据支撑特征点的分布,对初始匹配集的剩余特征点对进行支撑描述,并根据所生成支撑描述符的相似程度,判定剩余特征点对是否为正确匹配.经判定正确的匹配特征点对被加入支撑特征集,使支撑特征集动态扩展,保证了支撑特征点的分布密度及支撑描述的准确性.实验结果表明,该方法能够在保留正确匹配的同时,消除90%以上的错误匹配,有效提高正确匹配率.

References

[1]  Piccinini P, Prati A,Cucchiara R.Real-time object detection and localization with SIFT-based clustering[J].Image and Vision Computing,2012,30(8):573-587
[2]  Mikolajczyk K, Schmid C.A performance evaluation of local descriptors[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(10):1615-1630
[3]  Ha S W, Moon Y H.Multiple object tracking using sift features and location matching[J].International Journal of Smart Home,2011,5(4):17-26
[4]  赵龙,肖军波. 一种改进的运动目标抗遮挡跟踪算法[J].北京航空航天大学学报,2013,39(4):517-520 Zhao Long,Xiao Junbo.Improved algorithm of tracking moving objects under occlusions[J].Journal of Beijing University of Aeronautics and Astronautics,2013,39(4):517-520(in Chinese)
[5]  田越,张永梅, 李波.遥感图像的快速配准方法[J].北京航空航天大学学报,2008,34(11):1356-1359 Tian Yue,Zhang Yongmei,Li Bo.Fast remote sensing image registration algorithm[J].Journal of Beijing University of Aeronautics and Astronautics,2008,34(11):1356-1359(in Chinese)
[6]  Mortensen E N, Deng H,Shapiro L.A sift descriptor with global context[C]//Cordelia Schmid.IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Los Alamitos,CA:IEEE Computer Society,2005,1:184-190
[7]  蔡晓东,叶培建. 基于特征点集的匹配算法应用于卫星姿态确定[J].北京航空航天大学学报,2006,32(2):171-175 Cai Xiaodong,Ye Peijian.Image matching algorithm based on feature point set for satellite attitude calculation[J].Journal of Beijing University of Aeronautics and Astronautics,2006, 32(2): 171-175(in Chinese)
[8]  Ha S W, Moon Y H.Multiple object tracking using sift features and location matching[J].International Journal of Smart Home,2011,5(4):17-26
[9]  赵龙,肖军波. 一种改进的运动目标抗遮挡跟踪算法[J].北京航空航天大学学报,2013,39(4):517-520 Zhao Long,Xiao Junbo.Improved algorithm of tracking moving objects under occlusions[J].Journal of Beijing University of Aeronautics and Astronautics,2013,39(4):517-520(in Chinese)
[10]  田越,张永梅, 李波.遥感图像的快速配准方法[J].北京航空航天大学学报,2008,34(11):1356-1359 Tian Yue,Zhang Yongmei,Li Bo.Fast remote sensing image registration algorithm[J].Journal of Beijing University of Aeronautics and Astronautics,2008,34(11):1356-1359(in Chinese)
[11]  Lowe D G. Distinctive image features from scale-invariant key points[J].International Journal of Computer Vision,2004, 60(2): 91-110
[12]  Mikolajczyk K, Schmid C.A performance evaluation of local descriptors[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(10):1615-1630
[13]  Chen J H, Chen C S,Chen Y S.Fast algorithm for robust template matching with M-estimators[J].IEEE Transactions on Signal Processing,2003,51(1):230-243
[14]  Lowe D G. Distinctive image features from scale-invariant key points[J].International Journal of Computer Vision,2004, 60(2): 91-110
[15]  Chen J H, Chen C S,Chen Y S.Fast algorithm for robust template matching with M-estimators[J].IEEE Transactions on Signal Processing,2003,51(1):230-243
[16]  Sidibe D, Montesinos P,Janaqi S.Fast and robust image matching using contextual information and relaxation[C]//Avenida D.Proceedings of 2nd International Conference on Computer Vision Theory and Applications.Esquerdo,Setubal,Portugal:INSTICC Press,2007:68-75
[17]  Sidibe D, Montesinos P,Janaqi S.Fast and robust image matching using contextual information and relaxation[C]//Avenida D.Proceedings of 2nd International Conference on Computer Vision Theory and Applications.Esquerdo,Setubal,Portugal:INSTICC Press,2007:68-75
[18]  Mortensen E N, Deng H,Shapiro L.A sift descriptor with global context[C]//Cordelia Schmid.IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Los Alamitos,CA:IEEE Computer Society,2005,1:184-190

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133