Advertisement

A Semi-local Topological Constraint for Efficient Detection of Near-Duplicate Image Fragments

  • Mariusz Paradowski
  • Andrzej Śluzek
Conference paper
Part of the Advances in Intelligent and Soft Computing book series (AINSC, volume 95)

Abstract

Image matching methods are an important branch of computer vision and have many possible applications, i.e. robotics, navigation, etc. Their goal is to detect near-duplicate images, sub-images or even localize image fragments. The paper addresses the last, and the most difficult problem: simultaneous localization of multiple image fragments in images of unknown content. A complete lack of any a priori knowledge is assumed, including no information of the number of fragments. The presented approach combines low level vision techniques and high level spatial constraints. Photometric properties of the images are represented using key-regions. Spatial consistency is verified using the topology of images. A graph of topologically consistent key-regions is created. It allows efficient localization of entire near-duplicate image fragments.

Keywords

Neighborhood Size Topological Graph High Computational Complexity Topological Constraint Spatial Neighborhood 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Fischler, M., Bolles, R.: Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography. In: Buxton, B.F., Cipolla, R. (eds.) ECCV 1996. LNCS, vol. 1065, pp. 683–695. Springer, Heidelberg (1996)Google Scholar
  2. 2.
    Kannala, J., Salo, M., Heikkila, J.: Algorithms for computing a planar homography from conics in correspondence (2006)Google Scholar
  3. 3.
    Lowe, D.G.: Object recognition from local scale-invariant features. In: Proc. 7th IEEE Int. Conf. Computer Vision, vol. 2, pp. 1150–1157 (1999)Google Scholar
  4. 4.
    Lowe, D.G.: Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision 60(2), 91–110 (2004)CrossRefGoogle Scholar
  5. 5.
    Mikolajczyk, K., Schmid, C.: Scale and affine invariant interest point detectors. International Journal of Computer Vision 60, 63–86 (2004)CrossRefGoogle Scholar
  6. 6.
    Paradowski, M., Śluzek, A.: Automatic visual object formation using image fragment matching. In: Proc. International Multiconference on Computer Science and Information Technology, pp. 97–104 (2010)Google Scholar
  7. 7.
    Paradowski, M., Śluzek, A.: Detection of image fragments related by affine transforms: Matching triangles and ellipses. In: Proc. International Conference on Information Science and Applications, vol. 1, pp. 189–196 (2010)Google Scholar
  8. 8.
    Paradowski, M., Śluzek, A.: Keypoint-based detection of near-duplicate image fragments using image geometry and topology. In: Bolc, L., Tadeusiewicz, R., Chmielewski, L.J., Wojciechowski, K. (eds.) ICCVG 2010. LNCS, vol. 6375, pp. 175–182. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  9. 9.
    Paradowski, M., Śluzek, A.: Matching planar fragments in images using histograms of decomposed affine transforms. In: 2nd review in TPAMI (2010)Google Scholar
  10. 10.
    Schmid, C., Mohr, R.: Object recognition using local characterization and semi-local constraints. Technical report, INRIA (1996)Google Scholar
  11. 11.
    Schmid, C., Mohr, R.: Local grayvalue invariants for image retrieval. IEEE TPAMI 19(5), 530–535 (1997)Google Scholar
  12. 12.
    Yang, D., Sluzek, A.: A low-dimensional local descriptor incorporating tps warping for image matching. Image and Vision Computing (2009), doi:10.1016/j.imavis.2009.12.003Google Scholar
  13. 13.
    Zhao, W.-L., Ngo, C.-W.: Scale-rotation invariant pattern entropy for keypoint-based near-duplicate detection. IEEE Trans. on Image Processing 2, 412–423 (2009)CrossRefMathSciNetGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Mariusz Paradowski
    • 1
    • 2
    • 3
  • Andrzej Śluzek
    • 1
    • 2
    • 3
  1. 1.Institute of InformaticsWroclaw University of TechnologyPoland
  2. 2.School of Computer EngineeringNanyang Technological UniversitySingapore
  3. 3.Faculty of Physics, Astronomy and InformaticsNicolaus Copernicus UniversityPoland

Personalised recommendations