Title :
A JPEG Steganalysis Algorithm with Embedded Rate Estimation Based on Multi-class SVM
Abstract :
The purpose of image steganalysis is to find out if the hidden messages exist in images. After differencing DCT coefficients of natural-images and stego-images, their histograms present different laws. This paper regards this as the feature vector to determine if a hiding message is in the image. Through the analysis of former feature vector, the author extracts a new feature to estimate the amount of hidden information in JPEG. The results of the experiments show this approach is effective.
Keywords :
Algorithm design and analysis; Data mining; Discrete cosine transforms; Feature extraction; Histograms; Information analysis; Information security; Signal processing algorithms; Steganography; Support vector machines; Histogram; Multi-Class SVM; Steganalysis;
Conference_Titel :
Image and Signal Processing, 2008. CISP '08. Congress on
Conference_Location :
Sanya, China
Print_ISBN :
978-0-7695-3119-9
DOI :
10.1109/CISP.2008.516
