Scene Recognition Based on Multi-feature Fusion for Indoor Robot

Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10636)

Abstract

In this paper, a method of scene recognition based on multi feature fusion is proposed to solve the problems of poor accuracy in scene recognition of intelligent home robot. Firstly, the H/I color model is used to extract the color feature from the scene. Secondly, the characteristics of the uniform background are extracted by the DS descriptors, and the scene of great difference is extracted using the SURF descriptors. The extracted feature descriptors are quantized using the “visual bag of words”, and The SURF-DS-BOW model is generated by weighted fusion of the two feature descriptors. Finally, the multi kernel learning support vector machine (MKL-SVM) is used to fuse the color feature and the SURF-DS-BOW model to improve the accuracy of scene recognition. The experimental results show that the recognition rate of the method in indoor scene recognition is 86.4%, which is better than the relevant literature algorithm.

Keywords

Scene recognition MKL-SVM SURF-DS-BOW H/I color model 

References

  1. 1.
    Zhou, B., Lapedriza, A., Xiao, J.: Learning deep features for scene recognition using places database. In: 21st International Conference on Neural Information Processing Systems, pp. 487–495. MIT Press, Sarawak (2014)Google Scholar
  2. 2.
    Oliva, A., Torralba, A.: Building the gist of a scene: the role of global image features in recognition. J. Prog. Brain Res. 155(2), 23–36 (2006)CrossRefGoogle Scholar
  3. 3.
    Sivic, J., Zisserman, A.: Video Google: a text retrieval approach to object matching in videos. In: 9th IEEE International Conference on Computer Vision, p. 1470. IEEE Computer Society, France (2003)Google Scholar
  4. 4.
    Wang, R., Wang, Z., Ma, X.: Indoor scene classification based on the bag-of-words model of local feature information gain. J. IEICE Trans. Inf. Syst. E96.D(4), 984–987 (2013)CrossRefGoogle Scholar
  5. 5.
    Espinace, P., Kollar, T., Soto, A.: Indoor scene recognition through object detection. In: IEEE International Conference on Robotics and Automation, pp. 1406–1413. IEEE Xplore, Anchorage, Alaska, (2010)Google Scholar
  6. 6.
    Yu, J., Hong, C., Tao, D.: Semantic embedding for indoor scene recognition by weighted hypergraph learning. J. Signal Process. 112, 129–136 (2015)CrossRefGoogle Scholar
  7. 7.
    Sural, S., Qian, G., Pramanik, S.: Segmentation and histogram generation using the HSV color space for image retrieval. In: International Conference on Image Processing, Proceedings, vol. 2, pp. II-589–II-592. IEEE, New York (2002)Google Scholar
  8. 8.
    Nister, D., Stewenius, H.: Robust scalable recognition with a vocabulary tree. J. 2(10), 2161–2168 (2006)Google Scholar
  9. 9.
    Quan, Z., Rehman, S.U., Yu, Z.: Face recognition using dense sift feature alignment. J. Chin. J. Electron. 25(6), 1034–1039 (2016)CrossRefGoogle Scholar
  10. 10.
    Bay, H., Ess, A., Tuytelaars, T.: Speeded-up robust features. J. Comput. Vis. Image Underst. 110(3), 404–417 (2008)Google Scholar
  11. 11.
    Lu, X., Li, X., Mou, L.: Semi-supervised multitask learning for scene recognition. J. IEEE Trans. Cybern. 45(9), 1967–1976 (2015)CrossRefGoogle Scholar
  12. 12.
    Lee, J., Lim, J.H., Choi, H., Kim, D.-S.: Multiple kernel learning with hierarchical feature representations. In: Lee, M., Hirose, A., Hou, Z.-G., Kil, R.M. (eds.) ICONIP 2013. LNCS, vol. 8228, pp. 517–524. Springer, Heidelberg (2013). doi: 10.1007/978-3-642-42051-1_64 CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  1. 1.College of Communication EngineeringJilin UniversityChangchunChina

Personalised recommendations