2019/10/17 by Pengpeng Yang, Yang, Pengpeng, Rongrong Ni +7
Computer Science · #Advanced Steganography and Watermarking Techniques #Digital Media Forensic Detection #FOS: Computer and information sciences #FOS: Electrical engineering #Generative Adversarial Networks and Image Synthesis #Image and Video Processing (eess.IV) #Multimedia (cs.MM) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1910.07992
openalex publication_date 2019/10/17 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Contrast enhancement (CE) forensics techniques have always been of great interest for image forensics community, as they can be an effective tool for recovering image history and identifying tampered images. Although several CE forensic algorithms have been proposed, their accuracy and robustness against some kinds of processing are still unsatisfactory. In order to attenuate such deficiency, in this paper we propose a new framework based on dual-domain fusion convolutional neural network to fuse the features of pixel and histogram domains for CE forensics. Specifically, we first present a pixel-domain convolutional neural network (P-CNN) to automatically capture the patterns of contrast-enhanced images in the pixel domain. Then, we present a histogram-domain convolutional neural network (H-CNN) to extract the features in the histogram domain. The feature representations of pixel and histogram domains are fused and fed into two fully connected layers for the classification of contrast-enhanced images. Experimental results show that the proposed method achieve better performance and is robust against pre-JPEG compression and anti-forensics attacks. In addition, a strategy for performance improvement of CNN-based forensics is explored, which could provide guidance for the design of CNN-based forensics tools.