Recently, L1-norm based robust discriminant feature extraction technique has been attracted much attention in dimensionality reduction. However, most existing approaches solve the column vectors of the optimal projection matrix one by one with a greedy strategy. Moreover, they are not suitable for solving the multi-label image classification. To solve these problems, the authors give a model named L1-norm based discriminant manifold learning in this study. An iterative non-greedy algorithm is proposed to solve the objective and the obtained optimal projection matrix necessarily best optimise the corresponding trace ratio objective function, which is the essential criterion function for general supervised dimensionality reduction. They also analyse the convergence of the authors’ proposed algorithm in detail. Extensive experiments on some databases illustrate the effectiveness of their proposed method.

References

1. 1)
  - 5. Guo, Y.-F., Li, S.-J., Yang, J.-Y., et al: ‘A generalized Foley–Sammon transform based on generalized fisher discriminant criterion and its application to face recognition’, Pattern Recognit. Lett., 2003, 24, (1), pp. 147–158.
2. 2)
  - 1. Jolliffe, I.: ‘Principal component analysis’ (Wiley Online Library, New York, USA., 2002).
3. 3)
  - 2. Fisher, R.A.: ‘The use of multiple measurements in taxonomic problems’, Ann. Eugen., 1936, 7, (2), pp. 179–188.
4. 4)
  - 26. Shu, X., Xu, H., Tao, L.: ‘A least squares formulation of multi-label linear discriminant analysis’, Neurocomputing, 2015, 156, pp. 221–230.
5. 5)
  - 17. Liu, Y., Gao, Q., Gao, X., et al: ‘L2, 1-norm discriminant manifold learning’, IEEE Access, 2018, 6, pp. 40723–40734.
6. 6)
  - 23. Cheng, W., Hüllermeier, E.: ‘Combining instance-based learning and logistic regression for multilabel classification’, Mach. Learn., 2009, 76, (2-3), pp. 211–225.
7. 7)
  - 4. Wang, S.-J., Yang, J., Zhang, N., et al: ‘Tensor discriminant colour space for face recognition’, IEEE Trans Image Process, 2011, 20, (9), pp. 2490–2501.
8. 8)
  - 22. Tsoumakas, G., Katakis, I., Vlahavas, I.: ‘Mining multi-label data’, in Maimon, O., Rokach, L. (Eds.): Data mining and knowledge discovery handbook (Springer, Boston, MA, 2009, 2nd edn.).
9. 9)
  - 15. Wang, H., Yan, S., Xu, D., et al: ‘Trace ratio vs. Ratio trace for dimensionality reduction’. Proc. IEEE Int. Conf. Computer Vision on Pattern Recognition, Minneapolis, MN, USA., 2007, pp. 1–8.
10. 10)
  - 14. Zheng, W., Lin, Z., Wang, H.: ‘L1-norm kernel discriminant analysis via Bayes error bound optimization for robust feature extraction’, IEEE Trans. Neural Netw. Learn. Syst., 2014, 25, (4), pp. 793–805.
11. 11)
  - 3. Friedman, J.H.: ‘Regularized discriminant analysis’, J. Am. Stat. Assoc., 1989, 84, (405), pp. 165–175.
12. 12)
  - 20. Zhang, M.-L., Zhou, Z.-H.: ‘Ml-kNN: a lazy learning approach to multi-label learning’, Pattern Recognit., 2007, 40, (7), pp. 2038–2048.
13. 13)
  - 16. Wang, H., Ding, C., Huang, H.: ‘Multi-label linear discriminant analysis’. European Conf. on Computer Vision, San Francisco, CA, USA., 2010, pp. 126–139.
14. 14)
  - 24. Liu, H., Ma, Z., Zhang, S., et al: ‘Penalized partial least square discriminant analysis with l1-norm for multi-label data’, Pattern Recognit., 2015, 48, (5), pp. 1724–1733.
15. 15)
  - 11. Kwak, N.: ‘Principal component analysis based on L1-norm maximization’, IEEE Trans. Pattern Anal. Mach. Intell., 2008, 30, (9), pp. 1672–1680.
16. 16)
  - 10. Ke, Q., Kanade, T.: ‘Robust l1 norm factorization in the presence of outliers and missing data by alternative convex programming’. Proc. IEEE Conf. on Computer Vision and Pattern Recognition, San Diego, CA, USA., 2005, pp. 739–746.
17. 17)
  - 21. Tsoumakas, G., Vlahavas, I.: ‘Random k-label sets: an ensemble method for multilabel classification’. European Conf. on Machine Learning, Warsaw, Poland, 2007, pp. 406–417.
18. 18)
  - 6. De La Torre, F., Black, M.J.: ‘A framework for robust subspace learning’, Int. J. Comput. Vis., 2003, 54, (1–3), pp. 117–142.
19. 19)
  - 7. Zhang, F., Yang, J., Qian, J., et al: ‘Nuclear norm-based 2-dpca for extracting features from images’, IEEE Trans. Neural Netw. Learn. Syst., 2015, 26, (10), pp. 2247–2260.
20. 20)
  - 9. Ding, C., Zhou, D., He, X., et al: ‘R1-pca: rotational invariant l1-norm principal component analysis for robust subspace factorization’. Proc. of the 23rd Int. Conf. on Machine Learning, Pittsburgh, PA, USA., 2006, pp. 281–288.
21. 21)
  - 8. Gao, Q., Xu, S., Chen, F., et al: ‘R1-2DPCA and face recognition’, IEEE Trans. Cybern., 2019, 49, (4), pp. 1212–1223.
22. 22)
  - 19. Gao, Q., Ma, L., Liu, Y., et al: ‘Angle 2DPCA: a new formulation for 2DPCA’, IEEE Trans. Cybern., 2018, 48, (5), pp. 1672–1678.
23. 23)
  - 12. Zhong, F., Zhang, J.: ‘Linear discriminant analysis based on l1-norm maximization’, IEEE Trans. Image Process., 2013, 22, (8), pp. 3018–3027.
24. 24)
  - 18. Aanas, H., Fisker, R., Astrm, K., et al: ‘Robust factorization’, IEEE. Trans. Pattern Anal. Mach. Intell., 2002, 24, (9), pp. 1215–1225.
25. 25)
  - 25. Zhang, Y., Zhou, Z.-H.: ‘Multilabel dimensionality reduction via dependence maximization’, ACM Trans. Know. Discov. Data (TKDD), 2010, 4, (3), pp. 1–21.
26. 26)
  - 13. Liu, Y., Gao, Q., Miao, S., et al: ‘A non-greedy algorithm for l1-norm lda’, IEEE Trans. Image Process., 2017, 26, (2), pp. 684–695.

L1-norm based discriminant manifold learning for multi-label image classification

References

Related content