In this paper we propose a novel approach towards shape matching for image retrieval. The system takes advantages of generalized Hough transform, as it works well in detecting arbitrary shapes even in the presence of gaps and in handling rotation, scaling and shift variations, and solves the heavy computational aspect by introducing a preliminary automatic selection of the appropriate contour points to consider in the matching phase. The numerical simulations and comparisons have confirmed the effectiveness and the efficiency of the method proposed.
Generalized Hough Transform for Shape Matching
DI RUBERTO, CECILIA
2012-01-01
Abstract
In this paper we propose a novel approach towards shape matching for image retrieval. The system takes advantages of generalized Hough transform, as it works well in detecting arbitrary shapes even in the presence of gaps and in handling rotation, scaling and shift variations, and solves the heavy computational aspect by introducing a preliminary automatic selection of the appropriate contour points to consider in the matching phase. The numerical simulations and comparisons have confirmed the effectiveness and the efficiency of the method proposed.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.