Title :
A new integer energy activity measure for parallel bit plane coding
Author :
Mhedhbi, Imen ; Hachicha, Khalil ; Garda, Patrick
Author_Institution :
LIP6, UPMC Univ. Paris 06, Paris, France
Abstract :
We propose a reduced complexity Adaptive scanning algorithm (LPEAM) coupled with parallel bit plane coding for HENUC image compression. Typically, optimization of data memory layout and address register assignment allows reducing both execution time and code size of programs. The average efficiency improvement of LPEAM is around 50% over the original coder while keeping a good quality with an SSIM up to 0.985. In addition, parallel encoding is operating at higher processing speed up to 5 times more than the basis solution.
Keywords :
image coding; optimisation; HENUC image compression; LPEAM; adaptive scanning algorithm; code size; data memory layout optimisation; execution time; new integer energy activity measurement; parallel bit plane coding; parallel encoding; Acceleration; Adaptation models; Complexity theory; Encoding; Image coding; Multithreading; Prediction algorithms; EAM; HENUC; SSIM; accelaration; adaptive scanning; bit plane; complexity; memory; multithreading;
Conference_Titel :
New Circuits and Systems Conference (NEWCAS), 2014 IEEE 12th International
Conference_Location :
Trois-Rivieres, QC
DOI :
10.1109/NEWCAS.2014.6934043
