Recent studies have shown that the steganalytic approaches based on deep learning frameworks cannot surpass their rich?model features based companions in peiformance. According to our analysis, one of the main causes of the unsatisfactory performance of deep learning frameworks is that training procedure tends to get stuck at local plateaus or even diverge when starting from a non-ideal initial state. In this paper we will try to investi?gate how to fit deep neural network to a rich-model features set. We regard it as a pre-training procedure and study its 4fect on deep learning for steganalysis. The state-of-the-art JPEG steganalytic features set DCTR is selected as the target and its features extraction procedure is divided into multiple sub-models. A deep learning framework with similar sub-networks is proposed. In the pre-training procedure we train theframeworkfrom bottom to up, fitting the output of each sub-network to the actual output of the corresponding sub-module of DCTR. The motivation behind the scenario is that we reinforce the proposed framework learn to fit the nonlinear mapping implicit in DCTR and expect when it is trainedfrom an initial state which represents an approximate so?lution of DCTR, we can get better peiformance compared to what DCTR has achieved.
Feature-based steganalysis has been an integral tool for detecting the presence of steganography in communication channels for a long time. In this paper, we explore the possibility to utilize powerful optimization algorithms available in convolutional neural network packages to optimize the design of rich features. To this end, we implemented a new layer that simulates the formation of histograms from truncated and quantized noise residuals computed by convolution. Our goal is to show the potential to compactify and further optimize existing features, such as the projection spatial rich model (PSRM).
With the rapid development of mobile devices and multimedia processing technologies, digital multimedia applications has become increasingly more popular in our daily life. Due to the nature of digital media, digital images can be easily modified without leaving obvious traces. Digital image forensics is an emerging research field which aims to address the major problems as forgery detection, source identification, image recovery and detecting the existence of hidden information, which is also referred as steganalysis. Steganography is the science of covert communication which aims to conceal the existence of the secret information hidden in the communication. Steganalysis is the study of detecting hidden information, which is embedded by using steganography techniques. In this paper we present a forensic mobile application developed on iOS platform. This application is designed to perform both steganalysis and steganorgraphy tasks in digital images; that is to conduct information analysis of given images and determine if there is any secret information hidden in the given images, and also fulfill the task of hiding information invisibly into digital images. There have been a number of well-established techniques in digital image steganalysis and steganography fields during recent years. However, there are very few mobile forensic tools that have been developed to comprehensively adopt these methods. Our forensic mobile application aims to systematically include significant methods both in steganography and steganalysis fields, so that people can use it as a forensic tool to determine if an image contains a hidden message, as well as use it as a security tool to perform covert communication by hiding information in an image.