Title :
CORDIC architecture based 2-D DCT and IDCT for image compression
Author :
Sayali R. Bhaisare;Aniket V. Gokhale;Pravin K. Dakhole
Author_Institution :
Department of Electronic Engineering, Yeshwantrao Chavan College of Engineering, an autonomous institute affiliated to Rashtrasant Tukadoji Maharaj Nagpur University, maharashtra India
fDate :
4/1/2015 12:00:00 AM
Abstract :
Generally, 2-D DCT/IDCT (Two dimensional discrete cosine transform and its inverse) are widely used in many image processing systems. In this paper, efficient architectures are proposed. These architectures have parallel and pipelined structures which are used to implement 8×8 DCT/IDCT processors. These processors involve two 8-point DCT/IDCT processors along with a dual-bank of SRAM (128 words) and a coefficient ROM (6 words), a control unit and two multiplexers. Here, CORDIC arithmetic is used to design the kernel arithmetic unit (AU). The proposed architectures for 2-D DCT/IDCT not only reduce the power consumption but also simplify the hardware with a very high performance.
Keywords :
"Discrete cosine transforms","Program processors","Image processing","Computer architecture","Random access memory","Multiplexing","Kernel"
Conference_Titel :
Communications and Signal Processing (ICCSP), 2015 International Conference on
DOI :
10.1109/ICCSP.2015.7322759