Advertisement

Conditional Random Fields for Urban Scene Classification with Full Waveform LiDAR Data

  • Joachim Niemeyer
  • Jan Dirk Wegner
  • Clément Mallet
  • Franz Rottensteiner
  • Uwe Soergel
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6952)

Abstract

We propose a context-based classification method for point clouds acquired by full waveform airborne laser scanners. As these devices provide a higher point density and additional information like echo width or type of return, an accurate distinction of several object classes is possible. However, especially in dense urban areas correct labelling is a challenging task. Therefore, we incorporate context knowledge by using Conditional Random Fields. Typical object structures are learned in a training step and improve the results of the point-based classification process. We validate our approach with two real-world datasets and by a comparison to Support Vector Machines and Markov Random Fields.

Keywords

Conditional Random Fields 3D Point Cloud Full Waveform LiDAR Urban Classification 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Anguelov, D., Taskar, B., Chatalbashev, V., Koller, D., Gupta, D., Heitz, G., Ng, A.: Discriminative learning of markov random fields for segmentation of 3d scan data. In: Proc. of IEEE CVPR, vol. 2, pp. 169–176 (2005)Google Scholar
  2. Chehata, N., Guo, L., Mallet, C.: Airborne lidar feature selection for urban classification using random forests. IntArchPhRS 38(3), 207–212 (2009)Google Scholar
  3. Frey, B., MacKay, D.: A revolution: Belief propagation in graphs with cycles. In: Advances in Neural Information Processing Systems, vol. 10, pp. 479–485 (1998)Google Scholar
  4. Kumar, S., Hebert, M.: Discriminative random fields. International Journal of Computer Vision 68(2), 179–201 (2006)CrossRefGoogle Scholar
  5. Lafferty, J., McCallum, A., Pereira, F.: Conditional random fields: Probabilistic models for segmenting and labeling sequence data. In: Proc. 18th Int. Conference on Machine Learning, pp. 282–289 (2001)Google Scholar
  6. Lim, E., Suter, D.: Conditional random field for 3d point clouds with adaptive data reduction. In: Int. Conference on Cyberworlds, pp. 404–408 (2007)Google Scholar
  7. Lim, E., Suter, D.: 3d terrestrial lidar classifications with super-voxels and multi-scale conditional random fields. Comp. Aided Design 41(10), 701–710 (2009)CrossRefGoogle Scholar
  8. Mallet, C.: Analysis of Full-Waveform Lidar Data for Urban Area Mapping. Ph.D. thesis, Télécom ParisTech (2010)Google Scholar
  9. Munoz, D., Vandapel, N., Hebert, M.: Directional associative markov network for 3-d point cloud classification. In: Int. Symposium 3DPVT (2008)Google Scholar
  10. Rusu, R., Holzbach, A., Blodow, N., Beetz, M.: Fast geometric point labeling using conditional random fields. In: IEEE IROS, pp. 7–12 (2009)Google Scholar
  11. Shapovalov, R., Velizhev, A., Barinova, O.: Non-associative markov networks for 3d point cloud classification. In: PCV 2010. IntArchPhRS, vol. 38(A), pp. 103–108 (2010)Google Scholar
  12. Vishwanathan, S., Schraudolph, N., Schmidt, M., Murphy, K.: Accelerated training of conditional random fields with stochastic gradient methods. In: 23rd Int. Conference on Machine Learning, pp. 969–976 (2006)Google Scholar
  13. Wagner, W., Ullrich, A., Ducic, V., Melzer, T., Studnicka, N.: Gaussian decomposition and calibration of a novel small-footprint full-waveform digitising airborne laser scanner. ISPRS Journal of Photogrammetry and Remote Sensing 60(2), 100–112 (2006)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Joachim Niemeyer
    • 1
  • Jan Dirk Wegner
    • 1
  • Clément Mallet
    • 2
  • Franz Rottensteiner
    • 1
  • Uwe Soergel
    • 1
  1. 1.Institute of Photogrammetry and GeoInformationLeibniz University HannoverHannoverGermany
  2. 2.Laboratoire MATIS, Institut Géographique NationalUniversité Paris EstSaint-MandéFrance

Personalised recommendations