Title :
Invariants to convolution with circularly symmetric PSF
Author :
Flusser, Jan ; Zitova, Barbara
Author_Institution :
Inst. of Information Theory & Autom., Acad. of Sci. of the Czech Republic, Prague, Czech Republic
Abstract :
We introduce a new class of moment-based features invariant to rotation and to convolution with an unknown point-spread function having circular symmetry. Unlike the invariants published earlier, they comprise both even and odd order moments, which increases their discrimination power and robustness.
Keywords :
convolution; object recognition; optical transfer function; circular symmetry; convolution invariants; moment-based features; object recognition; point-spread function; Automation; Cameras; Convolution; Degradation; Image sensors; Information theory; Layout; Pattern recognition; Radiometry; Robustness;
Conference_Titel :
Pattern Recognition, 2004. ICPR 2004. Proceedings of the 17th International Conference on
Print_ISBN :
0-7695-2128-2
DOI :
10.1109/ICPR.2004.1333967
