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...

利用Hough变换的匹配对提纯

DOI: 10.11834/jig.20150804

Keywords: 匹配对提纯,Hough变换,误匹配,参数空间,投票

Full-Text   Cite this paper   Add to My Lib

Abstract:

目的针对传统的匹配对提纯算法存在容错性差、效率低等问题,提出了一种利用Hough变换的匹配对提纯算法。方法假设正确的匹配对一致性地服从一个变换模型。首先,为两幅图像的变换关系选择一个合适的数学模型,利用Hough变换确定模型方程参数的解。然后检验原始匹配对,保留符合模型方程的匹配对,从而达到提纯的目的。结果与传统的RANSAC(randomsampleconsensus)等算法相比,本文算法具有更高的容错率、召回率与更优的运行效率,且是稳定的。实验结果表明,在误配率低于85%时算法能完全剔除误匹配,且误配率高达95%时依然有50%的可能性成功剔除误匹配。结论把Hough变换引入到匹配对提纯的应用中,该算法在所选模型准确或近似准确的情况下能鲁棒地提纯匹配对。由于模型方程参数个数决定参数空间维数,维数高导致投票及搜索最大值点的时间、空间复杂度大,因此该算法适用于模型参数较少(不大于4)的情况。

References

[1]  Huber P J. Robust Statistics[M].New York: John Wiley & Sons, 1981:43-55.
[2]  Stewart C V.Robust parameter estimation in computer vision[J]. SIAM Reviews, 1999,41(3):513-537.[DOI:10.1137/S0036144598345802]
[3]  Rousseeuw P J. Least median of squares regression[J]. Journal of the American Statistical Association,1984,79:871-880.[DOI: 10.1080/01621459.1984.10477105]
[4]  Torr P H S, Zisserman A. MELSAC: a new robust estimator with application to estimating image geometry[J]. Computer Vision and Image Understanding, 2000, 78(1):138-156.[DOI: 10.1006/cviu.1999.0832]
[5]  Fischler M A, Bolles R C. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography[J]. Communications of the ACM,1981,24(6):381-395. [DOI:10.1145/358669.358692]
[6]  Wang Y W, Xu T F, Wang J H. Improved matching point purification algorithm mRANSAC[J].Journal of Southeast University, 2013,(A01):163-167.[王亚伟,许廷发,王吉晖. 改进的匹配点提纯算法mRANSAC[J].东南大学学报:自然科学版, 2013,(A01):163-167.][DOI:10.3969/j.issn.1001-0505. 2013. S1. 034]
[7]  Lowe D G. Distinctive image features from scale-invari-ant key points[J]. International Journal of Computer Vision, 2004, 60(2): 91-110. [DOI:10. 1023/B:VISI. 0000029664. 99615. 94]
[8]  Morel J M, Yu G. ASIFT: a new framework for fully affine invariant image comparison[J]. SIAM Journal on Image Sciences, 2009, 2(2):438-469.[DOI:10.1137/080732730]
[9]  Nister D. Preemptive RANSAC for live structure and motion estimation[C]//Proceedings of the 9th International Conference on Computer Vision. Nice, France:IEEE,2003:199-206.[DOI:10.1109/ICCV.2003.1238341]
[10]  Chum O, Matas J. Matching with PROSAC-progressive sample consensus[C]//Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition. SanDiego,CA,USA: IEEE, 2005:220-226. [DOI:10.1109/CVPR. 2005.221]
[11]  Matas J, Chum O. Randomized RANSAC with Td-d test[J]. Image and Vision Computing, 2004,22(10):837-842.[DOI: 10.1016/j.imavis.2004.02.009]
[12]  Chum O, Matas J. Optimal randomize RANSAC[J]. Pattern Analysis and Machine Intelligence, 2008,30(8):1472-1482.[DOI:10.1109/TPAMI.2007.70787]
[13]  Zhong J R,Du Q C, Liu Y, et al.Robust method of skew correction based on feature points matching[J]. Journal of Image and Graphics,2013,18(7):738-745.] [钟金荣,杜奇才,刘荧,等. 特征提取和匹配的图像倾斜校正[J].中国图象图形学报, 2013,18(7):738-745. [DOI:10.11834/jig.20130703]
[14]  Li X, Hu Z. Rejecting mismatches by correspondence function[J]. International Journal of Computer Vision, 2010, 89(1):1-17. [DOI:10.1007/s11263-010-0318-x]
[15]  Guil N, Villalba J, Zapata E L. A fast hough transform for segment detection[J]. IEEE Transactions on Image Processing, 1995,4(11): 1541-1548.[DOI: 10.1109/83.469935]
[16]  Hough P C C. Method and means for recognizing complex patterns: U.S.Patent,3069654[P].1962-12-18.
[17]  Guil N, Zapata E L. Lower order circle and ellipse hough transform[J].Pattern Recognition, 1997,30(10):1729-1744.[DOI: 10.1016/S0031-3203(96)00191-4]
[18]  Shapiro V A. On the hough transform of multi-level pictures[J]. Pattern Recognition, 1996,29(4):589-602.[DOI: 10.1016/0031-3203(95)00116-6]

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133