CUDA implementation of fractal image compression
- PDF / 1,973,444 Bytes
- 13 Pages / 595.276 x 790.866 pts Page_size
- 64 Downloads / 179 Views
ORIGINAL RESEARCH PAPER
CUDA implementation of fractal image compression Abir Al Sideiri1,2,3 · Nasser Alzeidi2 · Mayyada Al Hammoshi4 · Munesh Singh Chauhan5 · Ghaliya AlFarsi1 Received: 29 November 2018 / Accepted: 24 June 2019 © Springer-Verlag GmbH Germany, part of Springer Nature 2019
Abstract Fractal coding is a lossy image compression technique, which encodes the image in a way that would require less storage space using the self-similar nature of the image. The main drawback of fractal compression is the high encoding time. This is due to the hard task of finding all fractals during the partition step and the search for the best match of fractals. Lately, GPUs (Graphical Processing Unit) have been exploited to implement fractal image compression algorithms due to their high computational power. The prime aim of this paper is to design and implement a parallel version of the Fisher classification scheme using CUDA to exploit the computational power available in the GPUs. Fisher classification scheme is used to reduce the encoding time of fractal images by limiting the search for the best match of fractals. Encoding time, compression ratio and peak signal-to-noise ratio was used as metrics to assess the correctness and the performance of the developed algorithm. Eight images with different sizes (512 × 512, 1024 × 1024 and 2048 × 2048) have been used for the experiments. The conducted experiments showed that a speedup of 6.4 × was achieved in some images using NVIDIA GeForce GT 660 M GPU. Keywords Fractal image compression · Quad-tree partitioning · GPU · Parallel processing · CUDA
1 Introduction
* Abir Al Sideiri [email protected] Nasser Alzeidi [email protected] Mayyada Al Hammoshi [email protected] Munesh Singh Chauhan [email protected] Ghaliya AlFarsi [email protected] 1
Department of Information Technology, Buraimi University College, Al‑Buraimi, Oman
2
Department of Computer Science, Sultan Qaboos University, Muscat, Oman
3
Department of Systems and Networks, Universiti Tenaga National (UniTen), Kajang, Malaysia
4
School of Computer Information System, Virginia International University, Fairfax, VA, USA
5
Department of Information Technology, College of Applied Science, Salalah, Oman
Due to the advances in information systems and technologies, there is an essential need for efficient data storage and fast data transmission. Digital images possess the characteristic of being data intensive [1, 2]. Thus, storing these images in less memory leads to a direct reduction in data transmissions and storage costs. Therefore, data compression has always been an active area of research to offer solutions for these critical issues. There are two general categories of data compression methods, namely lossless and lossy methods. A lossless method will produce an image that is identical to the original image when decompressed. On the other hand, a lossy method will produce an image that closely resembles the original image. The main drawback of lossless methods is that they cannot achieve v
Data Loading...
