Pipelined embedded processor of quaternionic m-band wavelets for image multiresolution analysis

Author

Petrovsky, Nicolai A. ; Parfieniuk, Marek ; Petrovsky, Alexander

Author_Institution

Dept. of Comput. Eng., Belarusian State Univ. of Inf. & Radioelectron., Minsk, Belarus

fYear

2013

fDate

15-20 June 2013

Firstpage

196

Lastpage

199

Abstract

Multiresolution analysis of image data based on the novel m-channel linear-phase paraunitary filter banks (LP PUFBs) using quaternion multipliers are presented, which are aimed at finite-precision implementation. In the given paper the universal quaternion multiplier as a kernel of m-band wavelets analysis on combining the CORDIC algorithm with a lifting scheme is considered. Finally, the FPGA-based eight-channel linear phase paraunitary filter bank is validated by its application to image, in particular, to lossless-to-lossy coding.

Keywords

channel bank filters; digital arithmetic; embedded systems; field programmable gate arrays; image coding; image resolution; matrix decomposition; pipeline processing; wavelet transforms; CORDIC algorithm; FPGA-based eight-channel linear phase unitary filter bank; LP-PUFB; finite-precision implementation; image data multiresolution analysis; image multiresolution analysis; lifting scheme; lossless-to-Iossy coding; m-channel linear-phase paraunitary filter banks; pipelined embedded processor; quatemionic m-band wavelet analysis; universal quaternion multiplier; Table lookup; Wavelet transforms;

fLanguage

English

Publisher

ieee

Conference_Titel

Embedded Computing (MECO), 2013 2nd Mediterranean Conference on

Conference_Location

Budva

ISSN

1800-993X

Type

conf

DOI

10.1109/MECO.2013.6601355

Filename

6601355