Naslov
Robustly Adaptive Wavelet Filter Bank Using L1 Norm

Autori
Ana Sović ; Damir Seršić

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proceedings IWSSIP 2011 / Branka Zovko-Cihlar, Narcis Behlilović, Mesud Hadžialić - Sarajevo : Faculty of Electrical Engineering, University Sarajevo, B40 d.o.o., 2011, 9-12

ISBN
987-9958-9966-1-0

Skup
18th International Conference on Systems, Signals and Image Processing

Mjesto i datum
Sarajevo, Bosna i Hercegovina, 16.06.2011. - 18.06.2011

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Adaptive wavelets; vanishing moments; minimum L1 criterion; LAD adaptation

Sažetak
Sparse representation of signals is the key for many applications, such as denoising, compression, or compressive sensing. In this paper, we propose an original adaptive wavelet filter bank that, for a class of signals, provides better compaction of information. Previously reported 1D and 2D point-wise adaptive wavelets were based on minimization of the L2 error norm. Now, we introduce minimum of the L1 norm on a sliding window as the adaptation criterion. Its main advantages are robustness to outliers and sparser representation of the input data. The proposed algorithm was tested on synthetic signals. It shows significant improvement over known methods, which is paid with somewhat increased numerical complexity. Still, there is some room for improvements, by further development of the adaptive criterion and its efficient realization.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika, Računarstvo

POVEZANOST RADA