From Eigenspots to Fisherspots — Latent Spaces in the Nonlinear Detection of Spot Patterns in a Highly Varying Background
We present a scheme for the development of a spot detection procedure which is based on the learning of latent linear features from a training data set. Adapting ideas from face recognition to this low level feature extraction task, we suggest to learn a collection of filters from representative data that span a subspace which allows for a reliable distinction of a spot vs. the heterogeneous background; and to use a non-linear classifier for the actual decision. Comparing different subspace projections, in particular principal component analysis, partial least squares, and linear discriminant analysis, in conjunction with subsequent classification by random forests on a data set from archaeological remote sensing, we observe a superior performance of the subspace approaches, both compared with a standard template matching and a direct classification of local image patches.
Unable to display preview. Download preview PDF.
- BELHUMEUR, P.N. et al. (1996): Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projections. Proc ECCV.Google Scholar
- BORGA, M. (1997): A Unified Approach to PCA, PLS, MLR and CCA. Technical Report, University of Linkoping, Sweden.Google Scholar
- BREIMAN, L. (2001): Random Forests. Machine Learning, 45, 532.Google Scholar
- MOON, T.K. and STIRLING, W.C. (2000): Mathematical Methods and Algorithms for Signal Processing. Prentice Hall, New York.Google Scholar
- RAHNENFUEHRER, J. and BOZINOV, D. (2004): Hybrid Clustering for Microarray Image Analysis. BMC Bioinformatics, 5, online.Google Scholar
- SHERRATT, A. (2004): Spotting Tells from Space: Antiquity, 77, online.Google Scholar