access icon free Fuzzy multiclass active learning for hyperspectral image classification

The possibility theory, which is an extension of fuzzy sets and fuzzy logic, has shown considerable potential for solving active learning (AL) problems, particularly for multiclass scenarios’ classification. Hence, two recently proposed fuzzy multiclass AL algorithms (classification ambiguity (CA) and fuzzy C-order ambiguity (FCOA)) are investigated to properly generalise them for classifying hyperspectral images, and two improved versions of the CA and FCOA are proposed. In addition to comparing the performances of the original and improved algorithms, several other state-of-the-art AL methods are evaluated, such as breaking ties, margin sampling, and multi-class level uncertainty, with or without diversity criteria such as angle-based diversity (ABD), clustering-based diversity (CBD), and enhanced clustering-based diversity (ECBD). Tests on two benchmark hyperspectral images confirm that the proposed improved algorithms are superior to and more effective than the original ones.

Inspec keywords: fuzzy set theory; hyperspectral imaging; learning (artificial intelligence); geophysical image processing; image classification; remote sensing; possibility theory

Other keywords: fuzzy sets; fuzzy multiclass active learning; hyperspectral image classification; fuzzy logic; active learning problems; fuzzy multiclass AL algorithms; possibility theory; fuzzy C-order ambiguity

Subjects: Computer vision and image processing techniques; Instrumentation and techniques for geophysical, hydrospheric and lower atmosphere research; Geography and cartography computing; Combinatorial mathematics; Knowledge engineering techniques; Algebra, set theory, and graph theory; Image recognition; Combinatorial mathematics; Geophysical techniques and equipment

References

    1. 1)
      • 7. Settles, B.: ‘Active learning literature survey’, University of Wisconsin, Madison, 2010, vol. 52, no. 55–66, p. 11.
    2. 2)
      • 16. Xu, Z., Yu, K., Tresp, V., et al: ‘Representative sampling for text classification using support vector machines’. European Conf. on Information Retrieval, 2003, pp. 393407.
    3. 3)
      • 2. Foody, G.M., Mathur, A.: ‘Toward intelligent training of supervised image classifications: directing training data acquisition for SVM classification’, Remote Sens. Environ., 2004, 93, (1), pp. 107117, doi: 10.1016/j.rse.2004.06.017.
    4. 4)
      • 18. Frélicot, C., Mascarilla, L., Fruchard, A.: ‘An ambiguity measure for pattern recognition problems using triangular-norms combination’, WSEAS Trans. Syst., 2004, 8, (3), pp. 27102715.
    5. 5)
      • 9. Tuia, D., Volpi, M., Copa, L., et al: ‘A survey of active learning algorithms for supervised remote sensing image classification’, IEEE J. Sel. Top. Signal Process., 2011, 5, (3), pp. 606617, doi: 10.1109/JSTSP.2011.2139193.
    6. 6)
      • 8. Demir, B., Persello, C., Bruzzone, L.: ‘Batch-mode active-learning methods for the interactive classification of remote sensing images’, IEEE Trans. Geosci. Remote Sens., 2011, 49, (3), pp. 10141031, doi: 10.1109/TGRS.2010.2072929.
    7. 7)
      • 17. Dubois, D., Foulloy, L., Mauris, G., et al: ‘Probability-possibility transformations, triangular fuzzy sets, and probabilistic inequalities’, Reliab. Comput., 2004, 10, (4), pp. 273297, doi: 10.1023/B:REOM.0000032115.22510.b5.
    8. 8)
      • 1. Olofsson, P., Foody, G.M., Herold, M., et al: ‘Good practices for estimating area and assessing accuracy of land change’, Remote Sens. Environ., 2014, 148, pp. 4257, doi: 10.1016/j.rse.2014.02.015.
    9. 9)
      • 13. Mascarilla, L., Berthier, M., Frélicot, C.: ‘A k-order fuzzy or operator for pattern classification with k-order ambiguity rejection’, Fuzzy Sets Syst., 2008, 159, (15), pp. 20112029, doi: 10.1016/j.fss.2008.02.019.
    10. 10)
      • 12. Mitra, P., Murthy, C.A., Pal, S.K.: ‘A probabilistic active support vector learning algorithm’, IEEE Trans. Pattern Anal. Mach. Intell., 2004, 26, (3), pp. 413418, doi: 10.1109/TPAMI.2004.1262340.
    11. 11)
      • 10. Persello, C., Boularias, A., Dalponte, M., et al: ‘Cost-sensitive active learning with lookahead: optimizing field surveys for remote sensing data classification’, IEEE Trans. Geosci. Remote Sens., 2014, 52, (10), pp. 66526664, doi: 10.1109/TGRS.2014.2300189.
    12. 12)
      • 14. Luo, T., Kramer, K., Goldgof, D.B., et al: ‘Active learning to recognize multiple types of plankton’, J. Mach. Learn. Res., 2005, 6, pp. 589613.
    13. 13)
      • 6. Liu, A., Jun, G., Ghosh, J.: ‘A self-training approach to cost sensitive uncertainty sampling’, Mach. Learn., 2009, 76, (2–3), pp. 257270, doi: 10.1007/s10994-009-5131-9.
    14. 14)
      • 19. Wang, X.Z., Dong, L.C., Yan, J.H.: ‘Maximum ambiguity-based sample selection in fuzzy decision tree induction’, IEEE Trans. Knowl. Data Eng., 2012, 24, (8), pp. 14911505, doi: 10.1109/TKDE.2011.67.
    15. 15)
      • 5. Samat, A., Gamba, P., Du, P., et al: ‘Active extreme learning machines for quad-polarimetric SAR imagery classification’, Int. J. Appl. Earth Obs. Geoinf., 2015, 35, pp. 305319, doi: 10.1016/j.jag.2014.09.019.
    16. 16)
      • 3. Samat, A., Li, J., Liu, S., et al: ‘Improved hyperspectral image classification by active learning using pre-designed mixed pixels’, Pattern Recognit., 2016, 51, pp. 4358, doi: 10.1016/j.patcog.2015.08.019.
    17. 17)
      • 4. Tuia, D., Ratle, F., Pacifici, F., et al: ‘Active learning methods for remote sensing image classification’, IEEE Trans. Geosci. Remote Sens., 2009, 47, (7), pp. 22182232, doi: 10.1109/TGRS.2008.2010404.
    18. 18)
      • 11. Wang, R., Chow, C.Y., Kwong, S.: ‘Ambiguity-based multiclass active learning’, IEEE Trans. Fuzzy Syst., 2015, 24, (1), pp. 242248, doi: 10.1109/TFUZZ.2015.2451698.
    19. 19)
      • 15. Brinker, K.: ‘Incorporating diversity in active learning with support vector machines’. Proc. 20th Int. Conf. on Machine Learning, Washington, DC, 2003, vol. 3, pp. 5966.
    20. 20)
      • 20. Mura, M.D., Villa, A., Benediktsson, J.A., et al: ‘Classification of hyperspectral images by using extended morphological attribute profiles and independent component analysis’, IEEE Geosci. Remote Sens. Lett., 2011, 8, (3), pp. 542546, doi: 10.1109/LGRS.2010.2091253.
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-ipr.2017.0784
Loading

Related content

content/journals/10.1049/iet-ipr.2017.0784
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading