作者: I. El Rube , N. Alajlan , M. Kamel , M. Ahmed , G. Freeman
DOI: 10.1007/11559573_52
关键词:
摘要: In this paper, a multiscale representation and retrieval method for 2D shapes is introduced. First, the are represented using area of triangles formed by shape boundary points. Then, Wavelet Transform (WT) used smoothing decomposing boundaries into levels. At each scale level, triangle-area (TAR) image corresponding Maxima-Minima lines obtained. The resulting TAR (MTAR) more robust to noise, less complex, selective than similar methods such as curvature scale-space (CSS). proposed tested compared CSS MPEG-7 CE-shape-1 dataset. results show that MTAR outperforms test.