高级搜索

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于改进局部不变特征的兴趣点匹配

张良 王海丽 吴仁彪

张良, 王海丽, 吴仁彪. 基于改进局部不变特征的兴趣点匹配[J]. 电子与信息学报, 2009, 31(11): 2620-2625. doi: 10.3724/SP.J.1146.2008.01440
引用本文: 张良, 王海丽, 吴仁彪. 基于改进局部不变特征的兴趣点匹配[J]. 电子与信息学报, 2009, 31(11): 2620-2625. doi: 10.3724/SP.J.1146.2008.01440
Zhang Liang, Wang Hai-li, Wu Ren-biao. Matching of Interesting Points Based on Improved SIFT Algorithm[J]. Journal of Electronics & Information Technology, 2009, 31(11): 2620-2625. doi: 10.3724/SP.J.1146.2008.01440
Citation: Zhang Liang, Wang Hai-li, Wu Ren-biao. Matching of Interesting Points Based on Improved SIFT Algorithm[J]. Journal of Electronics & Information Technology, 2009, 31(11): 2620-2625. doi: 10.3724/SP.J.1146.2008.01440

基于改进局部不变特征的兴趣点匹配

doi: 10.3724/SP.J.1146.2008.01440
基金项目: 

天津市自然科学基金(06YFJMJC00700,07JCYBJC13500),国家自然科学基金(60605008,60736009,60776807)和中国民航大学科研启动基金(QD04Q09)资助课题

Matching of Interesting Points Based on Improved SIFT Algorithm

  • 摘要: 该文提出了一种适用于目标跟踪的局部特征点检测与匹配方法,在尺度不变特征(Scale Invariant Feature Transform, SIFT)算法基础上进行了多方面的改进。在高斯差分尺度空间仅检测局部极大值,提高算法的稳定性;采用基于圆形邻域统计梯度方向直方图,来确定兴趣点的主方向和描述子,避免了图像旋转的运算代价;最后采用最近邻与次近邻之比来对96维的描述子进行匹配。所提方法在有效地提高匹配准确率的同时,大大提高了运算速度, 适用于对实时性要求较高的场合。
  • Lowe D. Distinctive image features from scale-invariantkeypoints[J].International Journal of Computer Vision.2004,60(2):91-110[2]Luo Jun, Ma Y, Takikawa E, Lao S, Kawade M, and LuBao-Liang. Person-specific SIFT features for facerecognition[C]. IEEE International Conference on Acoustics,Speech and Signal Processing, Honolulu, Hawaii, USA, April,2007, 2(II): 593-596.[3]Hu Xue-long, Tang Ying-cheng, and Zhang Zheng-hua. Videoobject matching based on SIFT algorithm[C]. InternationalConference on Neural Networks and Signal Processing,Zhenjiang, China, June, 2008: 412-415.Yang Zhan-Long and Guo Bao-Long. Image mosaic based onSIFT[C]. Intelligent Information Hiding and MultimediaSignal Processing, Harbin, China, August, 2008: 1422-1425.Gao Ke, Lin Shou-xun, Zhang Yong-dong, Tang Sheng, andRen Hua-min. Attention model based SIFT keypointsfiltration for image retrieval[C]. 7th IEEE/ACISInternational Conference on Computer and InformationScience, Portland, Oregon, USA, May, 2008: 191-196.[4]Re Y and Sukthankar R. PCA-SIFT: A more distinctiverepresentation for local image descriptors[C]. Proceedings ofthe IEEE Conference on Computer Vision and PatternRecognition, Washington, DC, USA, June, 2004, 2: 506-513.[5]Dalal N and Triggs B. Histograms of oriented gradients forhuman detection[C]. Proceedings of the IEEE Conference onComputer Vision and Pattern Recognition, San Diego, CA,USA, June, 2005: 886-893.[6]Lazebnik S, Schmid C, and Ponce J. A sparse texturerepresentation using local affine regions[J].IEEETransactions on Pattern Analysis an Machine Intelligence.2005, 27(8):1265-1278[7]Bay H, Tuytelaars T, and Gool Van J L. SURF: Speeded UpRobust Features[C]. European Conference on ComputerVision, Graz, Austria, May, 2006: 404-417.[8]Mikolajczyk K and Schmid C. A performance evaluation oflocal descriptors[J].IEEE Transactions on Pattern Analysisan Machine Intelligences.2005, 27(10):1615-1630[9]Stein A and Hebert M. Incorporating background invarianceinto feature-based object recognition[C]. IEEE Workshops onApplication of Computer Vision, Breckenridge, CO, USA,Jan, 2005, 1: 37-44.[10]Abdel-Hakim A E and Farag A A. CSIFT: A SIFT descriptorwith color invariant characteristics[C]. Proceedings ofComputer Vision and Pattern Recognition Conference, NewYork, USA, 2006: 1978-1983.[11]Ancuti C and Bekaert P. SIFT-CCH: Increasing the SIFTdistinctness by color co-occurrence histograms[C].Proceedings of the 5th International Symposium on imageand Signal Processing and Analysis, Istanbul, Turkey, 2007:130-135.Cheung W and Hamarneh G. n-SIFT: n-dimensional scaleinvariant feature transform for matching medical images[C].Proceedings of the Fourth IEEE International Symposium onBiomedical Imaging: From Nano to Macro, Washington D.C.,USA, 2007: 720-723.Kisku, D R, Rattani A, Grosso E, and Tistarelli M. Faceidentification by SIFT-based complete graph topology[C].IEEE Workshop on Automatic Identification AdvancedTechnologies, Alghero, Italy, June, 2007: 63-68.
  • 加载中
计量
  • 文章访问数:  3225
  • HTML全文浏览量:  88
  • PDF下载量:  1440
  • 被引次数: 0
出版历程
  • 收稿日期:  2008-11-03
  • 修回日期:  2009-04-20
  • 刊出日期:  2009-11-19

目录

    /

    返回文章
    返回