Title :
Parallel Framework for Real-time DSP Implementation of EBC Algorithm
Author :
Zuo, Zhen ; Su, Shaojing ; Du, Liebo
Author_Institution :
Coll. of Mechatron. Eng. & Autom., Nat. Univ. of Defense Technol., Changsha
Abstract :
JPEG2000, the state-of-the-art image compression standard, has many advantages, such as high compression efficiency and high image fidelity. However, because of its high computational complexity, real-time hardware implementation of JPEG2000 becomes a big problem when using it in deadline constrained applications. EBC algorithm is the main part of JPEG2000 which comprises more than 50% of the computational complexity. Aiming at the shortcut of the EBC framework which is not suitable for the establishment of DSP software pipeline, a parallel framework for real-time DSP implementation of EBC algorithm is proposed in this paper. The original bit-plane coding module is firstly partitioned into bit-plane scanning module and arithmetic coding module. Then the sign coding is separated from the bit-plane scanning module and implemented ahead. And the multiple loops of the bit-plane scanning module are unrolled with context combining. By these optimizations, the software pipeline of the EBC algorithm can be established which leads to an eight-time faster implementation speed.
Keywords :
computational complexity; data compression; digital signal processing chips; image coding; DSP software pipeline; EBC algorithm; JPEG2000; arithmetic coding module; bit-plane coding module; bit-plane scanning module; high compression efficiency; high computational complexity; high image fidelity; real-time DSP implementation; state-of-the-art image compression standard; Application software; Arithmetic; Computational complexity; Digital signal processing; Hardware; Image coding; Partitioning algorithms; Pipelines; Software algorithms; Transform coding; DSP; EBC algorithm; JPEG2000; pipeline;
Conference_Titel :
Education Technology and Computer Science, 2009. ETCS '09. First International Workshop on
Conference_Location :
Wuhan, Hubei
Print_ISBN :
978-1-4244-3581-4
DOI :
10.1109/ETCS.2009.450