← Back to Volume 8

Universal Steganalysis for image Based on Genetic Algorithm and Grey-SVC

Yuehong Wu, Shen Zhuang, Lihong Ma, Zixian Feng
pp.211-216 DOI: 10.62051/ajeb9466 August 9, 2024
Download PDF
Abstract: The isolated samples can produce some effect on distinguishing the best classifying plane, which becomes one of causes of less performance of universal steganalysis that uses Support Vector Machines (SVM) as classifier. This paper proposes a new universal steganalysis algorithm for image based on Genetic Algorithm (GA) and Grey Support Vector Machines (GSVM). The algorithm firstly catches characteristic of noise signal in wavelet domain of image, then utilizes GA search samples which are used to train, and finds the best characteristic of species, finally makes grey relational degree between sample characteristic and the best characteristic of species participate in training of SVM, thus constructs a GSVM to be a classifier of steganalysis. The result testing on the large numbers of images indicates that the proposed universal steganalysis algorithm has less false positive rate and better classifying performance compared to Holotyak’s algorithm which has the same characteristic with above algorithm, which indicates that GSVM can reduce effect of isolated samples.
Keywords: Steganalysis, Genetic Algorithm, Grey Relational Analysis.
APA Citation: Yuehong Wu, Shen Zhuang, Lihong Ma, Zixian Feng (2024). Universal Steganalysis for image Based on Genetic Algorithm and Grey-SVC. Transactions on Economics, Business and Management Research, 8(1), 211-216. https://doi.org/10.62051/ajeb9466

References

  1. I.AVCIBAS, N.MEMON, B.SANKUR, et al. Image steganalysis with binary similarity measures [A]. IEEE International Conference on Image Processing [C].New York: Rochester, 2022.56-59
  2. L. Siwei and H. Farid. Steganalysis Using Color Wavelet Statistics and One-Class Support Vector Machines [A]. SPIE Symposium on Electronic Imaging[C]. San Jose: 2014.168-170
  3. T. Holotyak, J. Fridrich, S. Voloshynovskiy. Blind Statistical Steganalysis of Additive Steganography Using Wavelet Higher Order Statistics [A].9th IFIP TC-6 TC-11 Conference on Communications and Multimedia Security[C]. 2015.273-274
  4. J. H. Holland. Adaptation in nature and artificial systems [M].University of Michigan Press, Ann Arbor, 1995.
  5. J. H. Wu and C. B. Chen. An alternative form for grey relational grades [J]. The Journal of Grey System, 2019, 11(1): 7-12.
  6. M. K. Mihcak, I. Kozintsev, K. Ramchandran, et al, Low-Complexity Image Denoising Based on Statistical Modeling of Wavelet Coefficients[J]. IEEE Signal Processing Letters, 2019, vol.6 (12):300-303.
  7. L.M.SCHMITT. Theory of genetic algorithms [J].Theoretical Computer Science, 2021, 259(1):1-61.
  8. V. Vapnik. Statistic learning theory [M]. New York: J.Wiley, 2008.
  9. NRCS Photo Gallery,http://photogallery.nrcs.usda.gov.