Paper
19 December 2002 Steps towards the development of an automatic classifier for astronomical sources
Author Affiliations +
Abstract
We present the progress we have made in implementing a new kind of automatic classifier for astronomical objects. The developed classifier will work both in the image and time domain and take into account the geometrical and temporal characteristics of the sources. We have first constructed a 2D classifier which is based on a Self Organizing Map. The developed network is able to learn through experience and to discriminate between astronomical objects such as stars, galaxies, saturated objects or blended objects. In order to recognize and classify variable objects, the method had to be improved. We present the next step of classification through our 3D (geometry - time) classifier. The temporal characteristics of the sources are obtained by different analysis of their light curves: time domain, frequency and time-frequency analysis. We add the geometrical and temporal characteristics to obtain a complete classification of the sources. We plan to use the difference image analysis to obtain block of images and analyze them directly through the classifier. Such a complete classification has not yet been realized in the astronomical domain. In general our method works better than other automatic methods and allows a more complete discrimination through astronomical sources.
© (2002) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Carole Thiebaut, Michel Boer, and Sylvie Roques "Steps towards the development of an automatic classifier for astronomical sources", Proc. SPIE 4847, Astronomical Data Analysis II, (19 December 2002); https://doi.org/10.1117/12.461027
Lens.org Logo
CITATIONS
Cited by 1 scholarly publication.
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Astronomy

Wavelets

Stars

Autoregressive models

Galactic astronomy

Time-frequency analysis

Neurons

RELATED CONTENT


Back to Top