Alternative anti-forensics method for contrast enhancement

Abstract

Digital image forensic researchers have proposed different robust detection schemes to classify authentic images and contrast enhanced images. The main idea of the detectors is based on the peaks and gaps introduced in the histogram after performing contrast enhancement on an image. The classifier using these peak-gap artifacts as feature achieves high accuracy result. After that, Anti-forensic researchers proposed a method to remove the peak-gap artifacts by local random dithering method which significantly reduces the accuracy of current detection scheme. In this paper, an alternative anti-forensic method is proposed to avoid the peak-gap artifacts and still have good image quality in terms of PSNR. We perform the experiment by calculating PSNR between traditional contrast enhanced images and our result images. © 2012 Springer-Verlag.