Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Steganalysis method for detecting embedded coefficients of discrete-wavelet image transformation into high-frequency domains

 
: Sivachev, Alexey V.; Bashmakov, Daniil A.; Mikhailishenko, Olga V.; Korobeynikov, Anatoliy G.; Rieke, Roland

:

You, I.:
Mobile Internet Security. Second International Symposium, MobiSec 2017 : Jeju Island, Republic of Korea, October 19-22, 2017, Revised Selected Papers
Singapore: Springer Nature Singapore, 2019 (Communications in computer and information science 971)
ISBN: 978-981-13-3731-4 (Print)
ISBN: 978-981-13-3732-1 (Online)
ISBN: 978-981-13-3733-8
pp.83-95
International Symposium on Mobile Internet Security (MobiSec) <2, 2017, Jeju Island/Korea>
English
Conference Paper
Fraunhofer SIT ()

Abstract
The paper deals with a research aimed at providing the highest performance of detection of an embedding into a wavelet-domain image, in particular in LH and HL subbands, due to the combined use of several methods proposed by the authors. The paper discusses various methods for enhancing detection of the embedding into a wavelet-domain image as proposed by the authors for their possible combined use in order to ensure the best possible performance of detecting the embedding into a wavelet-domain image. These methods use the features of the wavelets transform, interrelations of various domains of the coefficients obtained by the wavelet transform of the image, or peculiar features or frequency domain of images. By the results of this research, a steganalysis method is proposed, based on the combined use of the above-described methods for increasing the steganalysis efficiency, which allows providing a better performance of detection of a wavelet-domain image embedding compared with existing methods of the steganalysis. The proposed methods of increasing the efficiency of steganalysis will improve the effectiveness of the steganalysis of information embedded in the LH and HL subbands by 47% in comparison with the already existing methods. The proposed method is based on combining them and allows you to get an additional increase in efficiency by 13%. The results of the given research can be useful for the experts, dealing with the information security problems while detecting and counteracting with the hidden data channel, based on the use of steganography, including in the mobile Internet. The obtained results can be used in the development of steganalysis systems.

: http://publica.fraunhofer.de/documents/N-531184.html