Advertisement

Supervised Approach to Sky and Ground Classification Using Whiteness-Based Features

  • Flávia de Mattos
  • Arlete Teresinha Beuren
  • Bruno Miguel Nogueira de Souza
  • Alceu De Souza BrittoJr.
  • Jacques Facon
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10633)

Abstract

Sky \(\diagup \!\!\!\) ground detection plays an important role in many applications such as unmanned control vehicle, dehazing process, cloud detection, for instance. This paper proposes a supervised sky-ground classification technique to color images. The novelty of the proposal is to evaluate the efficiency of whiteness indexes on the classification task. The strategy of the proposal consists in evaluating the power of whiteness indices in classification task. Eleven whiteness indices are used as features to feed a SVM classifier. Experimental results onto 1200 images and numerical evaluations have highlighted that the combination of five whiteness indices is a interesting strategy to classify the sky and the ground.

Keywords

Sky Ground Segmentation SVM classifier Whiteness index 

References

  1. 1.
    De Croon, G., De Wagter, C., Remes, B., Ruijsink, R.: Sky segmentation approach to obstacle avoidance. In: Proceedings of IEEE Aerospace Conference, pp. 1–16 (2011)Google Scholar
  2. 2.
    Russell, B.C., Torralba, A.T., Murphy, K.P., Freeman, W.T.: LabelMe: a database and web-based tool for image annotation. Int. J. Comput. Vis. 77, 157–173 (2008)CrossRefGoogle Scholar
  3. 3.
    Lipschutz, I., Gershikov, E., Milgrom, B.: New methods for horizon line detection in infrared and visible sea images. Int. J. Comput. Eng. Res. 3, 226–233 (2013)Google Scholar
  4. 4.
    Williams, S., Howard, A.M.: Horizon line estimation in glacial environments using multiple visual cues. In: PIEEE International Conference on Robotics and Automation, ICRA, pp. 5887–5892 (2011)Google Scholar
  5. 5.
    Lin, J.: Factors affecting the perception and measurement of optically brightened white textiles. Ph.D. thesis, Graduate Faculty of North Carolina State University Raleigh, North Carolina (2013)Google Scholar
  6. 6.
    X-Rite: X-Rite: Color iQC and color iMatch color calculations guide, Version 8.0 (2012). http://www.xrite.com/service-support/product-support/formulation-and-qc-software/color-iqc
  7. 7.
    Puebla, C.P.: Whiteness assessment: a primer. Axiphos GmbH, Germany (2006)Google Scholar
  8. 8.
    Joiner, A., Hopkinson, I., Deng, Y., Westland, S.: A review of tooth colour and whiteness. J. Dent. 36, s2–s7 (2008)CrossRefGoogle Scholar
  9. 9.
    Smith, T.C., Frank, E.: Introducing machine learning concepts with WEKA. In: Mathé, E., Davis, S. (eds.) Statistical Genomics. MMB, vol. 1418, pp. 353–378. Springer, New York (2016).  https://doi.org/10.1007/978-1-4939-3578-9_17CrossRefGoogle Scholar
  10. 10.
    Van Geste, T., Suykens, J., Baesens, B.: Benchmarking least squares support vector machine classifiers. Mach. Learn. 54, 5–32 (2004)CrossRefGoogle Scholar
  11. 11.
    Mihail, R.P., Workman, S., Bessinger, Z., Jacobs, N.: Sky segmentation in the wild: an empirical study. Inference Corporation (2016). http://mypages.valdosta.edu/rpmihail/

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  • Flávia de Mattos
    • 1
  • Arlete Teresinha Beuren
    • 2
  • Bruno Miguel Nogueira de Souza
    • 3
  • Alceu De Souza BrittoJr.
    • 1
  • Jacques Facon
    • 4
  1. 1.Pontifícia Universidade Católica do ParanáCuritibaBrazil
  2. 2.UTFPR Universidade Tecnológica Federal do ParanáSanta HelenaBrazil
  3. 3.Universidade Estadual do Norte do ParanáBandeirantesBrazil
  4. 4.Universidade Federal do Espírito SantoSão MateusBrazil

Personalised recommendations