Approximation power of directionlets
dc.contributor.author | Velisavljević, Vladan | en_GB |
dc.contributor.author | Beferull-Lozano, Baltasar | en_GB |
dc.contributor.author | Vetterli, Martin | en_GB |
dc.contributor.author | Dragotti, Pier Luigi | en_GB |
dc.date.accessioned | 2013-05-30T13:01:41Z | |
dc.date.available | 2013-05-30T13:01:41Z | |
dc.date.issued | 2005 | |
dc.identifier.citation | Velisavljevic, V.; Beferull-Lozano, B.; Vetterli, M. and Dragotti, P.-L. (2005) 'Approximation power of directionlets', IEEE International Conference on Image Processing (ICIP 2005, Genova, Italy, 14 September. Genova: IEEE, vol.1, pp.I-741 | en_GB |
dc.identifier.isbn | 0780391349 | |
dc.identifier.doi | 10.1109/ICIP.2005.1529857 | |
dc.identifier.uri | http://hdl.handle.net/10547/293061 | |
dc.description.abstract | In spite of the success of the standard wavelet transform (WT) in image processing, the efficiency of its representation is limited by the spatial isotropy of its basis functions built in only horizontal and vertical directions. One-dimensional (1-D) discontinuities in images (edges and contours), which are very important elements in visual perception, intersect too many wavelet basis functions and reduce the sparsity of the representation. To capture efficiently these anisotropic geometrical structures, a more complex multi-directional (M-DIR) and anisotropic transform is required. We present a new lattice-based perfect reconstruction and critically sampled anisotropic M-DIR WT (with the corresponding basis functions called directionlets) that retains the separable filtering and simple filter design from the standard two-dimensional (2-D) WT and imposes directional vanishing moments (DVM). Further-more, we show that this novel transform has non-linear approximation efficiency competitive to the other previously proposed over-sampled transform constructions. | |
dc.language.iso | en | en |
dc.publisher | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC | en_GB |
dc.relation.url | http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=1529857 | en_GB |
dc.subject | filtering theory | en_GB |
dc.subject | image reconstruction | en_GB |
dc.subject | image representation | en_GB |
dc.subject | image sampling | en_GB |
dc.subject | wavelet transforms | en_GB |
dc.title | Approximation power of directionlets | en |
dc.type | Conference papers, meetings and proceedings | en |
html.description.abstract | In spite of the success of the standard wavelet transform (WT) in image processing, the efficiency of its representation is limited by the spatial isotropy of its basis functions built in only horizontal and vertical directions. One-dimensional (1-D) discontinuities in images (edges and contours), which are very important elements in visual perception, intersect too many wavelet basis functions and reduce the sparsity of the representation. To capture efficiently these anisotropic geometrical structures, a more complex multi-directional (M-DIR) and anisotropic transform is required. We present a new lattice-based perfect reconstruction and critically sampled anisotropic M-DIR WT (with the corresponding basis functions called directionlets) that retains the separable filtering and simple filter design from the standard two-dimensional (2-D) WT and imposes directional vanishing moments (DVM). Further-more, we show that this novel transform has non-linear approximation efficiency competitive to the other previously proposed over-sampled transform constructions. |
This item appears in the following Collection(s)
-
Centre for Wireless Research (CWR)
The Centre for Wireless Research brings together expertise in the areas of mobile and wireless sensor networks. The breadth and depth of the expertise make the Centre rich with research and innovation potential.