Počet záznamů: 1
Non-separable rotation moment invariants
- 1.0555291 - ÚTIA 2023 RIV GB eng J - Článek v odborném periodiku
Bedratyuk, L. - Flusser, Jan - Suk, Tomáš - Kostková, Jitka - Kautský, J.
Non-separable rotation moment invariants.
Pattern Recognition. Roč. 127, č. 1 (2022), č. článku 108607. ISSN 0031-3203. E-ISSN 1873-5142
Grant CEP: GA ČR GA21-03921S
Institucionální podpora: RVO:67985556
Klíčová slova: Image recognition * Rotation invariants * Non-separable moments * Appell polynomials * Bi-orthogonality * Recurrent relation
Obor OECD: Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Impakt faktor: 8, rok: 2022
Způsob publikování: Omezený přístup
http://library.utia.cas.cz/separaty/2022/ZOI/flusser-0555291.pdf https://www.sciencedirect.com/science/article/pii/S0031320322000887?via%3Dihub
In this paper, we introduce new rotation moment invariants, which are composed of non-separable Appell moments. We prove that Appell polynomials behave under rotation as monomials, which enables easy construction of the invariants. We show by extensive tests that non-separable moments may outperform the separable ones in terms of recognition power and robustness thanks to a better distribution of their zero curves over the image space.
Trvalý link: http://hdl.handle.net/11104/0330272
Počet záznamů: 1