Steganalysis Of Halftone Image Using Inverse Halftoning

摘要

An inverse halftoning based steganalysis method of halftone images is proposed. In this method, the wavelet statistic features extracted from the reconstructed gray level image through the inverse halftoning of a given halftone image are used in our support vector machines classifier. Different inverse halftoning techniques are investigated including filtering based methods and projection based methods in reconstructing the gray level image. Filtering based and projection based detectors are then proposed and the performance of the steganalysis is compared based on these two detectors.