Welcome to the InfoSci Platform
Reference Hub4
Steganalysis of AMR Based on Statistical Features of Pitch Delay

Steganalysis of AMR Based on Statistical Features of Pitch Delay

Yanpeng Wu, Huiji Zhang, Yi Sun, Minghui Chen
Copyright: © 2019 |Volume: 11 |Issue: 4 |Pages: 16
ISSN: 1941-6210|EISSN: 1941-6229|EISBN13: 9781522565178|DOI: 10.4018/IJDCF.2019100105
Cite Article Cite Article

MLA

Wu, Yanpeng, et al. "Steganalysis of AMR Based on Statistical Features of Pitch Delay." IJDCF vol.11, no.4 2019: pp.66-81. http://doi.org/10.4018/IJDCF.2019100105

APA

Wu, Y., Zhang, H., Sun, Y., & Chen, M. (2019). Steganalysis of AMR Based on Statistical Features of Pitch Delay. International Journal of Digital Crime and Forensics (IJDCF), 11(4), 66-81. http://doi.org/10.4018/IJDCF.2019100105

Chicago

Wu, Yanpeng, et al. "Steganalysis of AMR Based on Statistical Features of Pitch Delay," International Journal of Digital Crime and Forensics (IJDCF) 11, no.4: 66-81. http://doi.org/10.4018/IJDCF.2019100105

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

The calibrated matrix of the second-order difference of the pitch delay (C-MSDPD) feature has been proven to be effective in detecting steganography based on pitch delay. In this article, a new steganalysis scheme based on multiple statistical features of pitch delay is present. Analyzing the principle of the adaptive multi-rate (AMR) codec, the pitch delay values in the same frame is divided into groups, in each of which, a pitch delay has a closer correlation with the other ones. To depict the characteristic of the pitch delay, two new types of statistical features are adopted in this article. The new features and C-MSDPD feature are together employed to train a classifier based on support vector machine (SVM). The experimental result shows that, the proposed scheme outperforms the existing one at different embedding bit rates and with different speech lengths.