Abstract
A novel approach for automatic mine detection using SONAR data is proposed in this paper relying on a probabilistic based fusion method to classify SONAR instances as mine or mine-like object. The proposed semi-supervised algorithm minimizes some target functions, which fuse context identification, multi-algorithm fusion criteria and a semi-supervised learning term. Our optimization purpose is to learn contexts as compact clusters in subspaces of the high-dimensional feature space through probabilistic feature discrimination and semi-supervised learning. The semi-supervised clustering component appoints degree of typicality to each data sample in order to identify and reduce the influence of noise points and outliers. Then, the approach yields optimal fusion parameters for each context. The experiments on synthetic datasets and standard SONAR dataset illustrate that our semi-supervised local fusion outperforms individual classifiers and unsupervised local fusion.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Peyvandi, H., Farrokhrooz, M., Roufarshbaf, H., Park, S.-J.: SONAR systems and underwater signal processing: classic and modern approaches. In: Kolev, N.Z. (ed.) SONAR systems, pp. 173–206. InTech, Hampshire (2011)
Walsh, N.E., Walsh, W.S.: Rehabilitation of landmine victims: the ultimate challenge
Zamora, G.: Detecting Land Mine, November 2014. http://www.nmt.edu/mainpage/news/landmine.html
Miles, D.: Confronting the Land Mine Threat, November 2014. http://www.dtic.mil/afps/news/9806192.html
Dye, D.: High frequency sonar components of normal and hearing impaired dolphins. Master’s thesis, Naval postgraduate school, Monterey, CA (2000)
Miles, D.: DOD Advances Countermine Technology, November 2014. http://www.dtic.mil/afps/news/9806193.html
Lv, C., Wang, S., Tan, M., Chen, L.: UA-MAC: an underwater acoustic channel access method for dense mobile underwater sensor networks. Int. J. Distrib. Sens. Netw. 2014, 10 p (2014). Article ID 374028
Ben Ismail, M.M., Bchir, O.: Insult detection in social network comments using probabilistic based fusion approach. In: Lee, R. (ed.) Computer and Information Science. SCI, vol. 566, pp. 15–25. Springer, Heidelberg (2015)
Ben Ismail, M.M., Bchir, O., Emam, A.Z.: Endoscopy video summarization based on unsupervised learning and feature discrimination. In: IEEE Visual Communications and Image Processing, VCIP 2013, Malaysia (2013)
http://archive.ics.uci.edu/ml/datasets/Connectionist+Bench+(ines+vs.+Rocks)
Ben Abdallah, A.C., Frigui, H., Gader, P.D.: Adaptive local fusion with fuzzy integrals. IEEE Trans. Fuzzy Syst. 20(5), 849–864 (2012)
Ben Ismail, M.M., Frigui, H.: Unsupervised clustering and feature weighting based on generalized dirichlet mixture modeling. Inf. Sci. 274, 35–54 (2014). doi:10.1016/j.ins.2014.02.146
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Saydali, S., Parvin, H., Safaei, A.A. (2016). Classifier Ensemble by Semi-supervised Learning: Local Aggregation Methodology. In: Kofroň, J., Vojnar, T. (eds) Mathematical and Engineering Methods in Computer Science. MEMICS 2015. Lecture Notes in Computer Science(), vol 9548. Springer, Cham. https://doi.org/10.1007/978-3-319-29817-7_11
Download citation
DOI: https://doi.org/10.1007/978-3-319-29817-7_11
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-29816-0
Online ISBN: 978-3-319-29817-7
eBook Packages: Computer ScienceComputer Science (R0)