Exploiting Class Hierarchies for Knowledge Transfer in Hyperspectral Data
Obtaining ground truth for hyperspectral data is an expensive task. In addition, a number of factors cause the spectral signatures of the same class to vary with location and/or time. Therefore, adapting a classifier designed from available labeled data to classify new hyperspectral images is difficult, but invaluable to the remote sensing community. In this paper, we use the Binary Hierarchical Classifier to propose a knowledge transfer framework that leverages the information gathered from existing labeled data to classify the data obtained from a spatially separate test area. Experimental results show that in the absence of any labeled data in the new area, our approach is better than a direct application of the old classifier on the new data. Moreover, when small amounts of labeled data are available from the new area, our framework offers further improvements through semi-supervised learning mechanisms.
Unable to display preview. Download preview PDF.
- 1.Blum, A.: On-line algorithms in machine learning. In: Fiat, A. (ed.) Dagstuhl Seminar 1996. LNCS, vol. 1442. Springer, Heidelberg (1998)Google Scholar
- 4.Ham, J., Chen, Y., Crawford, M.M., Ghosh, J.: Investigation of the Random Forest Framework for Classification of Hyperspectral Data. IEEE Trans. Geoscience and Remote Sensing (2005) (page to appear)Google Scholar
- 5.Joachims, T.: Making large-scale SVM learning practical. In: Burges, C., Scholkopf, B., Smola, A. (eds.) Advances in Kernel Methods: Support Vector Learning, pp. 169–184. MIT Press, Cambridge (1999)Google Scholar
- 6.Kelly, M.G., Hand, D.J., Adams, M.N.: The impact of changing populations on classifier performance. In: The Fifth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 67–371 (1999)Google Scholar
- 12.Morgan, J.T.: Adaptive Hierarchical Classifier with Limited Training Data. PhD thesis, Dept. of Mech. Eng., Univ. of Texas at Austin (2002)Google Scholar