Title :
Oblique multiresolution analysis
Author :
Tira, Zyed ; Duvaut, PatricDk ; Brunel, Pierre ; Georgel, Bruno
Author_Institution :
ETIS-ENSEA, Cergy-Pontoise, France
Abstract :
We present a new type of multiresolution, called oblique multiresolution (OMR), necessary in the decision context. All the fundamental properties of OMR are given and justified. The aim is to develop a multiresolution analysis which provide a redundant decomposition for use in the detection and classification application. The signal decomposition is computed with pyramidal algorithms. We give a necessary-sufficient conditions to reconstruct the signal from its oblique multiresolution decomposition and show how to compute reconstructing filters
Keywords :
filtering theory; signal detection; signal reconstruction; signal resolution; wavelet transforms; decision; fundamental properties; necessary-sufficient conditions; oblique multiresolution; oblique multiresolution decomposition; orthonormal wavelet bases; reconstructing filters; redundant decomposition; signal classification; signal detection; signal reconstruction; Density estimation robust algorithm; Filters; Image coding; Image reconstruction; Multiresolution analysis; Ordinary magnetoresistance; Pattern recognition; Signal detection; Signal resolution; Sufficient conditions;
Conference_Titel :
Time-Frequency and Time-Scale Analysis, 1994., Proceedings of the IEEE-SP International Symposium on
Conference_Location :
Philadelphia, PA
Print_ISBN :
0-7803-2127-8
DOI :
10.1109/TFSA.1994.467332
