The use of neural networks for the automatic detection and classification of weak photometric sub-components in early-type galaxies
Weak photometric subcomponents offer an important tool to understand the present structure and past evolution of early-type galaxies. The detection of such structures can be performed either by detailed modeling of the 2-d light distribution or by means of expecially tailored filtering techniques such as, for instance, the Adaptive Laplacian Algorithm. In this paper we present vive application of a Multi Layer Perceptron and a Self-organizing neural nets to the detection and classification of some Laplacian morphologies. In particular we discuss the construction of a training set of images of artificial galaxies and some preliminary results.
KeywordsDust Convolution sinO
Unable to display preview. Download preview PDF.
- Lorenz H. et al. 1994, AA, 277, 321Google Scholar
- Capaccioli et al., 1996, in Proceed, ofs VI Italo-Korean Meeting, Word Scientific Press (Singapore), in pressGoogle Scholar
- de Vaucouleurs G, 1948, Ann. Ap., 11, 247Google Scholar
- D.E. Eumelhart, J.L. McClelland, “Parallel Distributed Processing”, MIT Press, Cambridge, MA (1986).Google Scholar
- T. Kohonen “The Self Organizing Map”, Proceedings of the IEEE, vol 78, n.9 (1990).Google Scholar
- N.R. Pal and S.K. Pal “ A review on image segmentation techniques”, Pattern Recognition, vol. 9, 1277 (1994).Google Scholar