A Parallel SRM Feature Extraction Algorithm for Steganalysis Based on GPU Architecture

Author

Kaizhi Chen ; Chenjun Lin ; Shangping Zhong ; Longkun Guo

Author_Institution

Coll. of Math. & Comput. Sci., Fuzhou Univ., Fuzhou, China

fYear

2014

fDate

13-15 July 2014

Firstpage

178

Lastpage

182

Abstract

Based on GPU parallel technology, this paper proposes a parallel SRM feature extraction algorithm to accelerate the extraction of SRM feature for steganalysis of HUGO images. Using the parallel program framework of OpenCL for GPU, we parallelize and implement a serial algorithm and employ some optimization technologies for our parallel program to accelerate the extraction process. The techniques include convolution unrolling, combined memory access, aversion of bank conflicts. The experimental results show that the speed of the proposed parallel extraction algorithm for different size images is 25~55 times faster than the original serial algorithm, and 2~4.2 times faster than running the parallel method on Quad-core CPU.

Keywords

convolution; feature extraction; graphics processing units; image coding; multiprocessing systems; optimisation; parallel architectures; parallel programming; steganography; GPU architecture; GPU parallel technology; HUGO images; OpenCL; bank conflicts; convolution unrolling; image size; memory access; optimization technologies; parallel SRM feature extraction algorithm; parallel program framework; quad-core CPU; serial algorithm; steganalysis; Parallel architectures; Programming; OpenCL; Parallel program; SRM feature; Steganalysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Parallel Architectures, Algorithms and Programming (PAAP), 2014 Sixth International Symposium on

Conference_Location

Beijing

ISSN

2168-3034

Print_ISBN

978-1-4799-3844-5

Type

conf

DOI

10.1109/PAAP.2014.36

Filename

6916460