Back to Volume
Paper: Detection of Spectra in Objective Prism Images Using Neural Networks
Volume: 61, Astronomical Data Analysis Software and Systems III
Page: 335
Authors: Smareglia, R.; Pasian, F.
Abstract: Some methods for the automatic detection of spectra in objective prism images have already been developed, but none of them is completely satisfactory, and an interactive session including a visual check is always necessary. In this paper, a neural network approach is proposed to solve this problem. A multi--layer feed forward neural network has been developed to distinguish single, multiple or overlapped spectra, and some encouraging results are shown. Future developments of this work include the use of neural networks for the detection of objective prism spectra in more complex situations, and their classification with an accuracy of about 2 spectral sub-classes.
Back to Volume