Deep Residual Network for Steganalysis of Digital Images

Boroumand, Mehdi; Chen, Mo; Fridrich, Jessica

doi:10.1109/tifs.2018.2871749

Cited by 698 publications

(449 citation statements)

References 51 publications

Supporting

Mentioning

435

Contrasting

Unclassified

Order By: Relevance

“…To evaluate the difficulty of IStego100K and provide benchmark results for researchers who subsequently use this dataset, we tested four latest and widely used image steganalysis methods on this proposed dataset, which are DCTR [11], GFR [12], XuNet [23] and SRNet [13]. DCTR [11] extracts the first-order statistics of quantized noise residuals obtained from the inputted image using 64 kernels of the discrete cosine transform (DCT) as features for steganalysis.…”

Section: Benchmark Methods and Evaluation Metricsmentioning

confidence: 99%

“…GFR [12] extracts features based on 2-dimensional (2D) Gabor filters, which have certain optimal joint localization properties in the spatial domain and in the spatial frequency domain and can describe the image texture features from different scales and orientations, therefore it can detect the changes of statistical feature distribution before and after steganography. XuNet [23] and SRNet [13] and based on convolutional neural networks (CNN), for which, XuNet [23] contains a 20-layer CNN and SRNet [13] designed a deep residual architecture to minimize the use of heuristics and extract features, finally these features are sent to classifiers for steganlysis. We use several evaluation indicators commonly used in classification tasks to evaluate the performance of our model, which are precision (P), recall (R), F1-score (F1) and accuracy (Acc).…”

Section: Benchmark Methods and Evaluation Metricsmentioning

confidence: 99%

See 1 more Smart Citation

IStego100K: Large-Scale Image Steganalysis Dataset

Yang

et al. 2020

Digital Forensics and Watermarking

View full text Add to dashboard Cite

In order to promote the rapid development of image steganalysis technology, in this paper, we construct and release a multivariable large-scale image steganalysis dataset called IStego100K. It contains 208,104 images with the same size of 1024*1024. Among them, 200,000 images (100,000 cover-stego image pairs) are divided as the training set and the remaining 8,104 as testing set. In addition, we hope that IS-tego100K can help researchers further explore the development of universal image steganalysis algorithms, so we try to reduce limits on the images in IStego100K. For each image in IStego100K, the quality factors is randomly set in the range of 75-95, the steganographic algorithm is randomly selected from three well-known steganographic algorithms, which are J-uniward, nsF5 and UERD, and the embedding rate is also randomly set to be a value of 0.1-0.4. In addition, considering the possible mismatch between training samples and test samples in real environment, we add a test set (DS-Test) whose source of samples are different from the training set. We hope that this test set can help to evaluate the robustness of steganalysis algorithms. We tested the performance of some latest steganalysis algorithms on IStego100K, with specific results and analysis details in the experimental part. We hope that the IStego100K dataset will further promote the development of universal image steganalysis technology 4 .

show abstract

Section: Benchmark Methods and Evaluation Metricsmentioning

confidence: 99%

Section: Benchmark Methods and Evaluation Metricsmentioning

confidence: 99%

IStego100K: Large-Scale Image Steganalysis Dataset

Yang

et al. 2020

Digital Forensics and Watermarking

View full text Add to dashboard Cite

show abstract

“…is set is called A train . If a feature extraction-based machine learning algorithm is applied (not all the machine learning algorithms need feature extraction [3]), we need to extract the features of the images. e usual methodology in machine learningbased steganalysis is to use the set A train to train a classi er.…”

Section: Preliminariesmentioning

confidence: 99%

“…In Experiment 7 (Table 6), the results using the SRNet [3] classication method are presented, and di erent training databases are used. Again, the prediction of the classi cation errors is accurate.…”

Section: Prediction Of the Classi Cation Errormentioning

confidence: 99%

See 1 more Smart Citation

Detection of Classifier Inconsistencies in Image Steganalysis

Lerch-Hostalot

Megías

2019

Proceedings of the ACM Workshop on Information Hiding and Multimedia Security

View full text Add to dashboard Cite

In this paper, a methodology to detect inconsistencies in classi cation-based image steganalysis is presented. e proposed approach uses two classi ers: the usual one, trained with a set formed by cover and stego images, and a second classi er trained with the set obtained a er embedding additional random messages into the original training set. When the decisions of these two classi ers are not consistent, we know that the prediction is not reliable. e number of inconsistencies in the predictions of a testing set may indicate that the classi er is not performing correctly in the testing scenario. is occurs, for example, in case of cover source mismatch, or when we are trying to detect a steganographic method that the classi er is no capable of modelling accurately. We also show how the number of inconsistencies can be used to predict the reliability of the classi er (classi cation errors).

show abstract