A&A Supplement series, Vol 121, March 1997, 575-585
Received March 14; accepted June 11, 1996
K. Bouyoucef, D. Fraix-Burnet, and S. RoquesLaboratoire d'Astrophysique de Toulouse, UMR 5572, GdR-PRC ISIS, Observatoire Midi-Pyrénées, 14 Avenue Edouard Belin, 31400 Toulouse, France
The regularized reconstruction algorithm Interactive Deconvolution with Error Analysis (IDEA) is presented. One of its main advantages is to control the error propagation and to provide an estimation of the quality of the deconvolved image. IDEA is compared with two popular probabilistic methods: Richardson-Lucy and Maximum Entropy. These three methods are applied to Hubble Space Telescope aberrated images: SN 1987A and the jets of M 87 and 3C 66B.
keywords: techniques: image processing -- supernovae: individual: SN 1987A -- galaxies: jets -- galaxies: individual: M 87 -- galaxies: individual: 3C 66B