A combined multiresolution approach for faint source extraction from inrared astronomical raw images sequence

摘要: Faint sources detection is one of the major issues during reconstruction an astronomical science image from a raw data sequence. This problem consequence limit infrared instruments as well number cosmic ray impacts (glitches) that leads to false detection. Astronomical images contain many objects with isotropic structures (e.g. point sources) but also plenty anisotropic information filamentary structures). The wavelet transform usually applied separate all these signal constituents in each pixel, then map built represent associated noise before applying source algorithm. Wavelets are adapted singularities (discontinuities), however, they have orientation selectivity. Therefore, do not smooth curves) effectively. paper presents combined approach conlourlet-wavelet for faint extraction raw- sequences. While contourlet representation provides oriented support efficient approximation structures, geometry effectively captured by separable wavelets. novel has been tested on real and simulated images, stemming space observatory database