Weakly Supervised Learning for Airplane Detection in Remote Sensing Images

Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 246)


In contrast to the conventional approaches to learn geo-target classifier using fully supervised learning techniques which heavily rely on the artificial annotation in the training set of remote sensing images (RSIs), this paper attempts to develop a weakly supervised learning (WSL) approach for airplane detection in RSIs with cluttered background. The framework includes a novel WSL method to train airplane classifier using the training images with weak labels and an efficient detection scheme to localize the airplanes. The proposed WSL mainly consists of three components: the negative mining based training set initialization, the updating process for both the positive and negative training set, and the classifier evaluation mechanism that can efficiently terminate the updating process for the best performance. Comprehensive experiments on a large number of RSIs and comparisons with state-of-the-art fully supervised models demonstrate the effectiveness and efficiency of the proposed work.


Weakly supervised learning Negative mining Airplane detection 



This work is supported by graduate starting seed fund of Northwestern Polytechnical University under grant Z2013105.


  1. 1.
    Tuia D, Pacifici F, Kanevski M, Emery WJ (2009) Classification of very high spatial resolution imagery using mathematical morphology and support vector machines. IEEE Trans Geosci Rem Sens 47(11):3866–3879CrossRefGoogle Scholar
  2. 2.
    Deselaers T, Alexe B, Ferrari V (2012) Weakly supervised localization and learning with generic knowledge. Int J Comput Vis 100(3):275–293CrossRefMathSciNetGoogle Scholar
  3. 3.
    Deselaers T, Alexe B, Ferrari V (2010) Localizing objects while learning their appearance. ECCV Part IV. LNCS 6314:452–466Google Scholar
  4. 4.
    Pandey M, Lazebnik S (2011) Scene recognition and weakly supervised object localization with deformable part-based model. ICCV, Barcelona, 6–13 November, 2011, pp. 1307–1314Google Scholar
  5. 5.
    Siva P, Xiang T (2011) Weakly supervised object detector learning with model drift detection. ICCV, Barcelona, 6–13 November, 2011, pp. 343–350Google Scholar
  6. 6.
    Achanta R, Hemami S et al (2009) Frequency-tuned salient region detection. IEEE conference on computer vision and pattern recognition, Miami, FL, 20–25 June, 2009, pp. 1597–1604Google Scholar
  7. 7.
    Hou X, Zhang L (2007) Saliency detection: a spectral residual approach. IEEE conference on computer vision and pattern recognition, Minneapolis, MN, 17–22 June, 2007, pp. 1–8Google Scholar
  8. 8.
    Harel J, Koch C, Perona P (2007) Graph-based visual saliency. Advances in neural information processing systems. MIT Press, Cambridge, MA, pp 545–552Google Scholar
  9. 9.
    Judd T, Ehinger K, Durand F, Torralba A (2009) Learning to predict where humans look. Proc. IEEE 12th international conference on computer vision, Kyoto, 29 September–2 October, 2009, pp. 2106–2133.Google Scholar
  10. 10.
    Han B, Zhu H, Ding Y. (2011) Bottom-up saliency based on weighted sparse coding residual. ACM International Conference on Multimedia, Scottsdale, Arizona, pp. 1117–1120Google Scholar
  11. 11.
    Achanta R, Estrada F, Wils P, Süsstrunk S (2008) Salient region detection and segmentation. Int Conf Comput Vis Syst 5008:66–75CrossRefGoogle Scholar
  12. 12.
    Siva P, Chris R, Tao X (2012) In defence of negative mining for annotating weakly labelled data. In ECCV. Springer, Berlin, pp 594–608Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Dingwen Zhang
    • 1
  • Jianfeng Han
    • 2
  • Dahai Yu
    • 1
    • 3
  • Junwei Han
    • 1
  1. 1.School of AutomationNorthwestern Polytechnical UniversityXi’anChina
  2. 2.Computer Department, Information Engineering SchoolTianjin University of CommerceTianjinChina
  3. 3.Tianjin Optical Electrical GaoSi Communication Engineering Technology Co., LtdTianjinChina

Personalised recommendations