A new paradigm for the comparison of size functions is presented; it stresses the relevance of the "angular points" of the functions, and gives greater value to the stable ones. A simple example of classification of monograms (88 elements in the training set, 88 in the test set, for 22 classes, with a hit rate of 78%) is given, and a current enhancement of the experiment is described.
Point selection: A new comparison scheme for size functions (with an application to monogram recognition) / Ferri M.; Frosini P.; Lovato A.; Zambelli C.. - STAMPA. - 1351:(1997), pp. 329-337. (Intervento presentato al convegno 3rd Asian Conference on Computer Vision, ACCV 1998 tenutosi a chn nel 1998) [10.1007/3-540-63930-6_138].
Point selection: A new comparison scheme for size functions (with an application to monogram recognition)
Frosini P.;
1997
Abstract
A new paradigm for the comparison of size functions is presented; it stresses the relevance of the "angular points" of the functions, and gives greater value to the stable ones. A simple example of classification of monograms (88 elements in the training set, 88 in the test set, for 22 classes, with a hit rate of 78%) is given, and a current enhancement of the experiment is described.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.