Freight Status Classification in Real-World Images Using SIFT and KNN Model

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

Abstract

This paper proposes a unified image classification framework to label railway freights status that includes the Scale-Invariant Feature Transform (SIFT) description through a robust optimization approach. The developed model consists of several computational stages: (a) the SIFT descriptors in each image are extracted; (b) the training features are optimized by using K-Affinity Propagation (K-AP) algorithm; (c) construction of the Expectation-Maximization Principal Component Analysis (EMPCA) is applied for feature compression into low dimensional space; and finally (d) k-nearest neighbor (KNN) is used to register each image to trained classifiers. In this paper we are particularly interested to evaluate the classification performance of proposed algorithm on a diverse dataset of 600 real-world freights images. The experimental results show the effectiveness of proposed feature optimization technique when compared with the performance offered by the same classification schema with different feature descriptors.

Keywords

Image classification Freights status classification SIFT K-AP EMPCA KNN 

References

  1. 1.
    Yu DH, Han JF (2011) The application of digital image processing in freight wagon inspection system. Third Int Conf Comput Tech Dev 3(1):319–327Google Scholar
  2. 2.
    Dong D, Dinh TB, Dinh T, Duc D (2012) Sports video classification using bag of words model. Intelligent information and database systems, vol 7198. Springer, Heidelberg, pp 316–325CrossRefGoogle Scholar
  3. 3.
    Grauman K, Darrel T (2005) The pyramid match kernel: discriminative classification with sets of image features. Proc. IEEE Int. Conf. Computer Vision (ICCV), Beijing, 17–21 October, 2005, pp. 1458–1465Google Scholar
  4. 4.
    Duda R, Hart P (1973) Pattern classification and scene analysis. Wiley, New York, NYMATHGoogle Scholar
  5. 5.
    Lowe DG (2004) Distinctive image features from scale-invariant keypoints. Int J Comput Vis 60(2):91–110CrossRefGoogle Scholar
  6. 6.
    Bay H et al (2006) Surf: speeded up robust features. Computer Vision–ECCV, 7–13 May, Graz, Austria, 2006, pp. 404–417Google Scholar
  7. 7.
    Clarke L, Velthuizen R et al (1993) MRI: stability of three supervised segmentation techniques. Magn Reson Imaging 11(1):95–106CrossRefGoogle Scholar
  8. 8.
    Yu DH, Ghita O, Sutherland A, Whelan PF (2006) Dictionary based lip reading classification. In: China-Ireland international conference on information and communications technologies, Hangzhou, China, 18–19 October, 2006, pp. 226–229Google Scholar
  9. 9.
    Rosten E, Drummond T (2006) Machine learning for high-speed corner detection. Computer Vision–ECCV, 7–13 May, Graz, Austria, 2006, pp. 430–443Google Scholar
  10. 10.
    Rublee E et al (2011) ORB: an efficient alternative to SIFT or SURF. Proc. IEEE Int. Conf. Computer Vision (ICCV), Barcelona, 6–13 November, 2011, pp. 2564–2571Google Scholar
  11. 11.
    Juan L, Gwun O (2009) A comparison of SIFT, PCA-SIFT and SURF. Int J Image Process 65:143–152Google Scholar
  12. 12.
    Roweis S (1998) EM algorithm for PCA and SPCA. Adv Neural Inform Process Syst 10:626–632Google Scholar
  13. 13.
    Frey BJ, Dueck D (2007) Clustering by passing messages between data points. Science 315:972–976CrossRefMATHMathSciNetGoogle Scholar
  14. 14.
    Dueck D. and Frey BJ. (2007). Non-metric affinity propagation for unsupervised image categorization. Proc. IEEE Int. Conf. Computer Vision (ICCV), Rio de Janeiro, 14–21 October, 2007, pp. 1–8Google Scholar
  15. 15.
    Lv G et al (2010) Optimization of training samples with affinity propagation algorithm for multi-class SVM classification. Adv Neural Network 2010:33–41Google Scholar
  16. 16.
    Yang D, Guo P (2009) Improvement of image modeling with affinity propagation algorithm for semantic image annotation. Proceedings of international conference on neural information processing, Bangkok, Thailand, 1–5 December, 2009, pp. 778–787Google Scholar
  17. 17.
    Zhang X et al (2010) K-AP: generating specified k clusters by efficient affinity propagation. Proc. IEEE Int. Conf. data mining (ICDM), Sydney, Australia, 14–17 December, 2010, pp. 1187–1192Google Scholar
  18. 18.
    Sim J, Kim SY, Lee J (2005) Prediction of protein solvent accessibility using fuzzy k-nearest neighbor method. Bioinformatics 21(12):2844–2849CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Dongyang Wang
    • 1
  • Dahai Yu
    • 1
    • 2
  • Junwei Han
    • 1
  • Shujun Li
    • 2
  1. 1.School of AutomationNorthwestern Poly-technical UniversityXi’anChina
  2. 2.Tianjin Optical Electrical GaoSi Communication Engineering Technology Co., LtdTianjinChina

Personalised recommendations