Multi-dimensional correlation steganalysis

Abstract

Multi-dimensional spatial analysis of image pixels have not been much investigated for the steganalysis of the LSB Steganographic methods. Pixel distribution based steganalysis methods could be thwarted by intelligently compensating statistical characteristics of image pixels, as reported in several papers. Simple LSB replacement methods have been improved by introducing smarter LSB embedding approaches, e.g. LSB matching and LSB+ methods, but they are basically the same in the sense of the LSB alteration. A new analytical method to detect LSB stego images is proposed in this paper. Our approach is based on the relative locations of image pixels that are essentially changed in an LSB embedding system. Furthermore, we introduce some new statistical features including "local entropies sum" and "clouds min sum" to achieve a higher performance. Simulation results show that our proposed approach outperforms some well-known LSB steganalysis methods, in terms of detection accuracy and the embedding rate estimation.

Towards blind detection of low‐rate spatial embedding in image steganalysis

“…SPAM features [22] using transition probability matrices of Markov chains are well known for detection of spatial domain steganography. Multi‐dimensional correlation steganalysis [23] tries to aggregate the pixels correlation in spatial domain and finds the distortion of the image impose by LSB alternation. Gul and Kurugollu [24] present a method to steganalyse highly undetectable stego (HUGO) that extracts features from the downsampled 5‐variate PDF of the image, and then uses an optimised support vector machine (SVM) to do an efficient detection up to 85% on BOSSRank database.…”

CAPTAIN: Comprehensive Composition Assistance for Photo Taking

“…Take Q f av as user favorite shot. 10: end procedure user-specified preference vector, and S N (C j , q) is the similarity vector between query q and each photo C j in the preferred style set of the user C. The computational detail of the composition and matching steps has been explained in Algorithm 2 and inspired from (Diyanat et al, 2011;Farhat et al, 2011Farhat et al, , 2012Farhat and Ghaemmaghami, 2014).…”