scholarly journals Digital image steganalysis based on local textural features and double dimensionality reduction

2014 ◽  
Vol 9 (8) ◽  
pp. 729-736 ◽  
Author(s):  
Fengyong Li ◽  
Xinpeng Zhang ◽  
Hang Cheng ◽  
Jiang Yu
Keyword(s):  
Dimensionality Reduction ◽  
Digital Image ◽  
Textural Features ◽  
Image Steganalysis

scholarly journals A Novel Digital Image Steganalysis Approach for Investigation

2012 ◽  
Vol 47 (12) ◽  
pp. 18-21 ◽  
Author(s):  
Nanhay Singh ◽  
Bhoopesh Singh Bhati ◽  
R. S. Raw
Keyword(s):  
Digital Image ◽  
Image Steganalysis

scholarly journals Adaptive Steganalysis Based on Selection Region and Combined Convolutional Neural Networks

2017 ◽  
Vol 2017 ◽  
pp. 1-9 ◽  
Author(s):  
Donghui Hu ◽  
Qiang Shen ◽  
Shengnan Zhou ◽  
Xueliang Liu ◽  
Yuqi Fan ◽  
...  
Keyword(s):  
Performance Improvement ◽  
Digital Image ◽  
Information Hiding ◽  
Digital Image Analysis ◽  
Hybrid Approach ◽  
Selection Method ◽  
Detection Accuracy ◽  
Adaptive Steganography ◽  
Image Steganalysis ◽  
Region Selection

Digital image steganalysis is the art of detecting the presence of information hiding in carrier images. When detecting recently developed adaptive image steganography methods, state-of-art steganalysis methods cannot achieve satisfactory detection accuracy, because the adaptive steganography methods can adaptively embed information into regions with rich textures via the guidance of distortion function and thus make the effective steganalysis features hard to be extracted. Inspired by the promising success which convolutional neural network (CNN) has achieved in the fields of digital image analysis, increasing researchers are devoted to designing CNN based steganalysis methods. But as for detecting adaptive steganography methods, the results achieved by CNN based methods are still far from expected. In this paper, we propose a hybrid approach by designing a region selection method and a new CNN framework. In order to make the CNN focus on the regions with complex textures, we design a region selection method by finding a region with the maximal sum of the embedding probabilities. To evolve more diverse and effective steganalysis features, we design a new CNN framework consisting of three separate subnets with independent structure and configuration parameters and then merge and split the three subnets repeatedly. Experimental results indicate that our approach can lead to performance improvement in detecting adaptive steganography.

Pure spatial rich model features for digital image steganalysis

2015 ◽  
Vol 75 (5) ◽  
pp. 2897-2912 ◽  
Author(s):  
Pengfei Wang ◽  
Zhihui Wei ◽  
Liang Xiao
Keyword(s):  
Digital Image ◽  
Image Steganalysis ◽  
Model Features

scholarly journals Digital image steganalysis: A survey on paradigm shift from machine learning to deep learning based techniques

2020 ◽  
Author(s):  
Arivazhagan Selvaraj ◽  
Amrutha Ezhilarasan ◽  
Sylvia Lilly Jebarani Wellington ◽  
Ananthi Roy Sam
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Digital Image ◽  
Paradigm Shift ◽  
Image Steganalysis

scholarly journals A Research on the Combination Strategies of Multiple Features for Hyperspectral Remote Sensing Image Classification

2018 ◽  
Vol 2018 ◽  
pp. 1-14 ◽  
Author(s):  
Yuntao Ma ◽  
Ruren Li ◽  
Guang Yang ◽  
Lishuang Sun ◽  
Jingli Wang
Keyword(s):  
Remote Sensing ◽  
Dimensionality Reduction ◽  
Kernel Method ◽  
Hyperspectral Remote Sensing ◽  
Experimental Results ◽  
Systematic Research ◽  
Textural Features ◽  
Multiple Features ◽  
Combination Strategies ◽  
Multiple Feature

It has been common to employ multiple features in the identification of the images acquired by hyperspectral remote sensing sensors, since more features give more information and have complementary properties. Few studies have discussed the combination strategies of multiple feature groups. This study made a systematic research on this problem. We extracted different groups of features from the initial hyperspectral images and tried different combination scenarios. We integrated spectral features with different textural features and employed different dimensionality reduction algorithms. Experimental results on three widely used hyperspectral remote sensing images suggested that “dimensionality reduction before combination” performed better especially when textural features performed well. The study further compared different combination frameworks of multiple feature groups, including direct combination, manifold learning, and multiple kernel method. The experimental results demonstrated the effectiveness of direct combination with an autoweight calculation.

Sign in / Sign up

Export Citation Format

Share Document