DocumentCode :
624439
Title :
A novel lifting algorithm for Daubechies wavelets using algebraic integers
Author :
Balakrishnan, Pranav ; Wahid, K.
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. of Saskatchewan, Saskatoon, SK, Canada
fYear :
2013
fDate :
5-8 May 2013
Firstpage :
1
Lastpage :
4
Abstract :
The paper presents a lifting based algorithm to compute Daubechies-4 (D4) and Daubechies-6 (D6) Wavelet Transforms using a novel algebraic integer quantization (AIQ) technique. The complex transform filter coefficients are first mapped error-freely by algebraic integers, and then quantized using subband coding algorithm. The performance of the scheme is assessed by using several benchmark images and the results are compared in terms of PSNR and bit rate. Finally, both D4 and D6 architectures have been implemented on FPGA. The results show that the proposed AIQ-based implementations consume fewer resources and perform better than most existing techniques.
Keywords :
data compression; field programmable gate arrays; filtering theory; image coding; matrix algebra; quantisation (signal); wavelet transforms; D4 architectures; D4 wavelet transform; D6 architectures; D6 wavelet transform; Daubechies-4 wavelet transform; Daubechies-6 wavelet transform; FPGA; PSNR; algebraic integers; benchmark images; bit rate; complex transform filter coefficients; error-freely mapping; field programmable gate arrays; lifting-based algorithm; quantization; subband coding algorithm; Computer architecture; Discrete wavelet transforms; Hardware; PSNR; Quantization (signal); Daubechies wavelet transform; Integer transforms; Lifting scheme; image compression;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Electrical and Computer Engineering (CCECE), 2013 26th Annual IEEE Canadian Conference on
Conference_Location :
Regina, SK
ISSN :
0840-7789
Print_ISBN :
978-1-4799-0031-2
Electronic_ISBN :
0840-7789
Type :
conf
DOI :
10.1109/CCECE.2013.6567727
Filename :
6567727
Link To Document :
بازگشت