Comparative Performance Analysis of a High Speed 2-D Discrete Wavelet Transform Using Three Different Architectures

Harshitha S, Gopalakrishna K

Abstract: The discrete wavelet transform is used extensively in many applications, such as computer vision, pattern recognition, sub band coding, speech analysis etc. because of its capability of decomposing a signal at multiple resolution levels. It is also used in the field of image processing for texture discrimination, fractal analysis, and image compression. This 2-D DWT requires massive computations and hence there is a need for use of efficient architectures for the implementation of 2-D DWT. This paper focuses on comparing the performance of three types of architectures that is intended to compute the discrete wavelet transform of a 2-D signal at a faster rate. Here we present a comparative study of the computation time for a high speed 2- D discrete wavelet transform implementation using three different architectures: single-processor architecture, parallelprocessor architecture and pipeline architecture. Keywords: Discrete wavelet transform, discrete cosine transform, fractal analysis, texture discrimination.

Title:	Comparative Performance Analysis of a High Speed 2-D Discrete Wavelet Transform Using Three Different Architectures
Author Name:	Harshitha S, Gopalakrishna K
Abstract:	Abstract: The discrete wavelet transform is used extensively in many applications, such as computer vision, pattern recognition, sub band coding, speech analysis etc. because of its capability of decomposing a signal at multiple resolution levels. It is also used in the field of image processing for texture discrimination, fractal analysis, and image compression. This 2-D DWT requires massive computations and hence there is a need for use of efficient architectures for the implementation of 2-D DWT. This paper focuses on comparing the performance of three types of architectures that is intended to compute the discrete wavelet transform of a 2-D signal at a faster rate. Here we present a comparative study of the computation time for a high speed 2- D discrete wavelet transform implementation using three different architectures: single-processor architecture, parallelprocessor architecture and pipeline architecture. Keywords: Discrete wavelet transform, discrete cosine transform, fractal analysis, texture discrimination.
Cite this article:	Harshitha S, Gopalakrishna K , " Comparative Performance Analysis of a High Speed 2-D Discrete Wavelet Transform Using Three Different Architectures " , International Journal of Emerging Trends & Technology in Computer Science (IJETTCS), Volume 3, Issue 4, July - August 2014 , pp. 114-117 , ISSN 2278-6856.
Full Text [PDF] Home