Title :
On a simplification of non-linear wavelets and its application to image compression/reconstruction
Author :
Nobuhara, H. ; Bede, B. ; Hirota, Kaoru
Author_Institution :
Tokyo Inst. of Technol., Japan
Abstract :
Summary form only given. A simplification of non-linear wavelets, which can be recognized as morphological wavelets, is proposed. The proposed non-linear wavelets can be deduced based on the perfect reconstruction condition with respect to analysis/synthesis operations, and it can be performed with low computational complexity compared with conventional non-linear wavelets. Through image compression/reconstruction experiments using 10 images extracted from the standard image database (SIDBA), it is confirmed that the root mean square of the proposed wavelet is decreased to 86.03% of that of the conventional one.
Keywords :
image coding; image reconstruction; mathematical morphology; wavelet transforms; image compression; image reconstruction; morphological wavelets; nonlinear wavelet simplification; perfect reconstruction condition; wavelet root mean square; Image coding; Image reconstruction;
Conference_Titel :
Nonlinear Signal and Image Processing, 2005. NSIP 2005. Abstracts. IEEE-Eurasip
Conference_Location :
Sapporo
Print_ISBN :
0-7803-9064-4
DOI :
10.1109/NSIP.2005.1502252
