access icon free Active learning with label correlation exploration for multi-label image classification

Multi-label image classification has attracted considerable attention in machine learning recently. Active learning is widely used in multi-label learning because it can effectively reduce the human annotation workload required to construct high-performance classifiers. However, annotation by experts is costly, especially when the number of labels in a dataset is large. Inspired by the idea of semi-supervised learning, in this study, the authors propose a novel, semi-supervised multi-label active learning (SSMAL) method that combines automated annotation with human annotation to reduce the annotation workload associated with the active learning process. In SSMAL, they capture three aspects of potentially useful information – classification prediction information, label correlation information, and example spatial information – and they use this information to develop an effective strategy for automated annotation of selected unlabelled example-label pairs. The experimental results obtained in this study demonstrate the effectiveness of the authors' proposed approach.

Inspec keywords: learning (artificial intelligence); image classification; image capture

Other keywords: semisupervised multilabel active learning method; automated annotation; label correlation exploration; machine learning; SSMAL method; human annotation; classification prediction information; label correlation information; example spatial information; multilabel image classification

Subjects: Computer vision and image processing techniques; Image recognition; Knowledge engineering techniques

References

    1. 1)
      • 13. Zhang, X., Cheng, J., Xu, C., et al: ‘Multi-view multi-label active learning for image classification’. Proc. Int. Conf. Multimedia and Expo (ICME), 2009, pp. 258261.
    2. 2)
      • 26. Tsoumakas, G., Spyromitros-Xioufis, E., Vilcek, J., et al: ‘MULAN: a JAVA library for multi-label learning’, J. Mach. Learn. Res., 2011, 12, pp. 24112414.
    3. 3)
      • 6. Boutell, M.R., Luo, J., Shen, X., et al: ‘Learning multi-label scene classification’, Pattern Recogn., 2004, 37, (9), pp. 17571771.
    4. 4)
      • 7. Li, X., Wang, L., Sung, E.: ‘Multilabel SVM active learning for image classification’. Proc. Int. Conf. Image Proc. (ICIP), 2004, vol. 4, pp. 22072210.
    5. 5)
      • 16. Ye, C., Wu, J., Sheng, V.S., et al: ‘Multi-label active learning with label correlation for image classification’. Proc. IEEE Int. Conf. Image Processing (ICIP), 2015, pp. 34373441.
    6. 6)
      • 10. Vasisht, D., Damianou, A., Varma, M., et al: ‘Active learning for sparse Bayesian multilabel classification’. Proc. ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining (KDD), 2014, pp. 472481.
    7. 7)
      • 4. Settles, B.: ‘Active learning literature survey’. Computer Science Technical Report 1648, University of Wisconsin-Madison, USA, 2010.
    8. 8)
      • 29. Li, J., Li, X., Yang, B., et al: ‘Segmentation-based image copy-move forgery detection scheme’, IEEE Trans. Inf. Forens. Secur., 2015, 10, (3), pp. 507518.
    9. 9)
      • 24. Duygulu, P., Barnard, K., de Freitas, J.F., et al: ‘Object recognition as machine translation: Learning a lexicon for a fixed image vocabulary’. Proc. European Conf. Computer Vision, 2002, pp. 97112.
    10. 10)
      • 23. Chua, T.S., Tang, J., Hong, R., et al: ‘NUS-WIDE: areal-world web image database from National University of Singapore’. Proc. ACM Int. Conf. Image Video Retrieval, 2009, pp. 19.
    11. 11)
      • 22. Zhu, X.J.: ‘Semi-supervised learning literature survey’. Computer Sciences Technical Report 1530, University of Wisconsin-Madison, USA, 2008.
    12. 12)
      • 11. Huang, S.J., Jin, R., Zhou, Z.H.: ‘Active learning by querying informative and representative examples’, IEEE Trans. Pattern Anal. Mach. Intell., 2014, 36, (10), pp. 19361949.
    13. 13)
      • 14. Zhang, B., Wang, Y., Wang, W.: ‘Batch mode active learning for multi-label image classification with informative label correlation mining’. Proc. IEEE Workshop on Applications of Computer Vision (WACV), 2012.
    14. 14)
      • 30. Gu, B., Sheng, V.S., Li, S.: ‘Bi-parameter space partition for cost-sensitive SVM’. Proc. 24th Int. Conf. Artificial Intelligence, 2015, pp. 35323539.
    15. 15)
      • 8. Singh, M., Curran, E., Cunningham, P.: ‘Active learning for multi-label image annotation’. Proc. Irish Conf. Artificial Intelligence and Cognitive Science (AICS), 2008, pp. 173182.
    16. 16)
      • 20. Jiao, Y., Zhao, P.P., Wu, J., et al: ‘Active multi-label learning with optimal label subset selection’. Proc. Int. Conf. Advanced Data Mining and Applications (ADMA), 2014, pp. 523534.
    17. 17)
      • 19. Zhao, S.Q., Wu, J., Sheng, V.S., et al: ‘Weak labeled multi-label active learning for image classification’. Proc. 23rd ACM Int. Conf. Multimedia (ACM MM), 2015, pp. 11271130.
    18. 18)
      • 2. Chen, B., Shu, H., Coatrieux, G., et al: ‘Color image analysis by quaternion-type moments’, J. Math. Imaging Vis., 2015, 51, (1), pp. 124144.
    19. 19)
      • 9. Li, X., Guo, Y.: ‘Active learning with multi-label SVM classification’. Proc. Int. Joint Conf. Artificial Intelligence (IJCAI), 2013, pp. 14791485.
    20. 20)
      • 15. Wang, H., Huang, H., Ding, C.: ‘Multi-label feature transform for image classifications’. Proc. European Conf. Computer Visition (ECCV), 2010, pp. 793806.
    21. 21)
      • 5. Wang, M., Hua, X.S.: ‘Active learning in multimedia annotation and retrieval: a survey’, ACM Trans. Intell. Syst. Techol., 2011, 2, (2), pp. 121.
    22. 22)
      • 18. Ye, C., Wu, J., Sheng, V.S., et al: ‘Multi-label active learning with chi-square statistics for image classification’. Proc. Int. Conf. Multimedia Retrieval (ICMR), 2015, pp. 583586.
    23. 23)
      • 1. Wen, X., Shao, L., Xue, Y., et al: ‘A rapid learning algorithm for vehicle classification’, Inf. Sci., 2015, 295, (1), pp. 395406.
    24. 24)
      • 17. Wu, J., Sheng, V.S., Zhang, J., et al: ‘Multi-label active learning for image classification’. Proc. Int. Conf. Image Processing (ICIP), 2014, pp. 52275231.
    25. 25)
      • 3. Zhang, M., Zhou, Z.: ‘A review on multi-label learning algorithms’, IEEE Trans. Know. Data Eng., 2013, 26, (8), pp. 18191837.
    26. 26)
      • 28. Zheng, Y., Jeon, B., Xu, D., et al: ‘Image segmentation by generalized hierarchical fuzzy C-means algorithm’, J. Intell. Fuzzy Syst., 2015, 28, (2), pp. 961973.
    27. 27)
      • 25. Spyromitros, E., Tsoumakas, G., Vlahavas, I.P.: ‘An empirical study of lazy multilabel classification algorithms’, LNCS (LNAI), 2008, 5138, pp. 401406.
    28. 28)
      • 21. Huang, S.J., Chen, S., Zhou, Z.H.: ‘Multi-label active learning: query type matters’. Proc. Int. Conf. Artificial Intelligence (AAAI), 2015, pp. 946952.
    29. 29)
      • 31. Gu, B., Sheng, V.S.: ‘A robust regularization path algorithm for ν-support vector classification’, IEEE Trans. Neural Netw. Learn. Syst., 2017, 28, (5), pp. 12411248.
    30. 30)
      • 12. Qi, G.J., Hua, X.S., Rui, Y., et al: ‘Two-dimensional active learning for image classification’. Proc. IEEE Int. Conf. Computer Vision and Pattern Recognition (CVPR), 2008, pp. 18.
    31. 31)
      • 27. Xia, Z., Wang, X., Sun, X., et al: ‘Steganalysis of LSB matching using differences between non-adjacent pixels’, Multimed Tools Appl., 2016, 75, (4), pp. 19471962.
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-cvi.2016.0243
Loading

Related content

content/journals/10.1049/iet-cvi.2016.0243
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading