Cross-Validation and Blind Feature Analysis of 25 Percent Embedding on JPEG Image Format using SVM
Deepa D. Shankar1, Prabhat Kumar Upadhyay2
1Deepa D.Shankar, Research Scholar, Banasthali Vidyapith, Rajasthan, India.
2Dr. Prabhat Kumar Upadhyay, Birla Institute of Technology Offshore Campus Ras al Khaimah, UAE.
Manuscript received on 15 September 2019 | Revised Manuscript received on 23 September 2019 | Manuscript Published on 11 October 2019 | PP: 1188-1191 | Volume-8 Issue-11S September 2019 | Retrieval Number: K124009811S19/2019©BEIESP | DOI: 10.35940/ijitee.K1240.09811S19
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: This paper provides a result assessment of traditional JPEG picture extraction function steganalysis compared to a cross-validation picture. Four distinct algorithms are used as steganographic systems in the spatial and transform domain. They are LSB Matching, LSB Replacement, Pixel Value Differencing and F5.A 25 percentage of embedding with text embedding information is considered in this paper. The characteristics regarded for evaluation are the First Order, Second Order, Extended DCT characteristics, and Markov characteristics. Support Vector Machine is the classifier used here. In statistical recovery, six distinct kernels and four distinct sampling techniques are used for evaluation.
Keywords: Crossvalidation, sampling, kernels, features, steganalysis, Support Vector Machines.
Scope of the Article: Image Analysis and Processing