Steganography detection based on Rich Model features is a hot research direction in steganalysis. However, Rich Model features usually result a large computation cost. To reduce the dimension of steganalysis features and improve the efficiency of steganalysis algorithm, differing from previous works normally proposed new feature extraction algorithm, this paper proposes a general steganalysis feature selection method based on decision rough set α-positive region reduction. First, it is pointed out that decision rough set α-positive region reduction is suitable for steganalysis feature selection. Second, a quantization method of attribute separability is proposed to measure the separability of steganalysis feature components. Third, steganalysis feature components selection algorithm based on decision rough set α-positive region reduction is given, thus stego images can be detected by the selected feature. The proposed method can significantly reduce the feature dimensions and maintain detection accuracy. Based on the BOSSbase-1.01 image database of 10000 images, a series of feature selection experiments are carried on two kinds of typical Rich Model features (35263-D J+SRM feature and 17000-D GFR feature). The results show that even though these two kinds of features are reduced to approximately 8000-D, the detection performance of steganalysis algorithms based on the selected features are also maintained with that of original features, which will remarkably improve the efficiency of feature extraction and stego image detection.

Feature selection based on the fuzzy neighborhood rough set model (FNRS) is highly popular in data mining. However, the dependent function of FNRS only considers the information present in the lower approximation of the decision while ignoring the information present in the upper approximation of the decision. This construction method may lead to the loss of some information. To solve this problem, this paper proposes a fuzzy neighborhood joint entropy model based on fuzzy neighborhood self-information measure (FNSIJE) and applies it to feature selection. First, to construct four uncertain fuzzy neighborhood self-information measures of decision variables, the concept of self-information is introduced into the upper and lower approximations of FNRS from the algebra view. The relationships between these measures and their properties are discussed in detail. It is found that the fourth measure, named tolerance fuzzy neighborhood self-information, has better classification performance. Second, an uncertainty measure based on the fuzzy neighborhood joint entropy has been proposed from the information view. Inspired by both algebra and information views, the FNSIJE is proposed. Third, the K–S test is used to delete features with weak distinguishing performance, which reduces the dimensionality of high-dimensional gene datasets, thereby reducing the complexity of high-dimensional gene datasets, and then, a forward feature selection algorithm is provided. Experimental results show that compared with related methods, the presented model can select less important features and have a higher classification accuracy.

Aiming at the problem that the traditional steganography based on carrier modification has the low steganographic capacity, a steganographic scheme based on Fully Convolutional Dense Connection Network (FC-DenseNet) is proposed. Since FC-DenseNet can effectively overcome the problems of gradient dissipation and gradient explosion, and a large number of features are multiplexed, the cascaded secret image and carrier image can reconstruct good image quality after entering the network. Effectively improve steganographic capacity. First, we reset the number of input channels of the first convolution block of FC-DenseNet and the number of output channels of the last convolution block and deleted the LogSoftmax() function. On the sender side, after the concatenated secret image and carrier image pass through the hidden network FC-DenseNet, the secret image is embedded in the carrier image to obtain a stego-image. At the receiving side, the extraction network reconstructs the secret image from the stego-image. Experimental results show that our proposed steganography scheme not only has a high Peak Signal-to-Noise Ratio(PSNR) and Structural Similarity(SSIM) but also can realize large-capacity image steganography, with an average image payload capacity of 23.96 bit per pixel.

Steganalysis feature selection shows excellent effectiveness on elevating the detection efficiency and decreasing time-space cost. However, the single evaluation criterion for features and the subjective selection basis always lead to valuable features neglect, which restricts the improvement of detection accuracy. To alleviate this predicament, this paper proposes a steganalysis feature selection method based on multidimensional evaluation and dynamic threshold allocation (MEDTA method). Firstly, to measure the feature components’ contribution degree to detection, the concept of partial entropy for steganalysis (ste- pe ) is defined and utilized to measure the mutual information between feature components. On this basis, the evaluation criterion for steganalysis feature components’ contribution degree is proposed, and the theoretical basis is given. Secondly, to measure the functional similarity of the feature components in distinguishing between cover images and stego images, by applying the property of cosine similarity between vectors, the evaluation criterion for steganalysis feature components’ contribution angle is proposed. Then, according to the Occam’s Razor, a multidimensional evaluation criterion based on contribution degree and contribution angle is proposed, which provides a basis for feature selection. In addition, to allocate the threshold for feature selection, this paper proposed a dynamic threshold allocation model, which combines the merits of several function models. Finally, feature selection with multidimensional evaluation and dynamic threshold allocation is proposed, which can achieve a comprehensive evaluation and objective selection for steganalysis features. Extensive experiments conducted on the BOSSbase1.01 image database demonstrate that the proposed MEDTA method could not only achieve highly competitive or even better performance in detection accuracy and feature dimension reduction, as compared with the state-of-the-art methods, but also get rid of depending on classifiers, so that the efficiency of feature extraction and steganalysis gets promoted.

To solve the problem of the low capacity of traditional information hiding and steganography, the deep neural network is used to construct a hidden network and a decoding network to hide one image into another. In this paper, the ResNet (Residual Network) has a good effect in feature extraction, and the residual block is used in the hidden network. First, the hidden network encoded the secret image into cover image. During the encoding process, the steganography image mimics the distribution of the cover image as much as possible. Then, the receiver sends the steganography image to the decoding network to obtain the secret image. Experimental results indicate that the scheme will solve the obvious problems of visual cues, also improve the ability of embedding.

Redundant steganalysis feature components in high-dimensional steganalysis feature of images increase the spatio-temporal complexity of steganalysis and even reduce the detection accuracy of the stego images. In order to reduce the image steganalysis feature dimension, improve the detection accuracy of the stego images and achieve fast feature selection, this paper proposes a general method for image steganalysis feature selection. Firstly, a feature metric algorithm based on the difference function is given, and this algorithm measures the difference of the steganalysis feature components between the cover image class and the stego image class, which provides the basis for selecting the steganalysis feature components contributing greatly to detect the stego images. Secondly, the Pearson correlation coefficient is improved and used to measure the correlation between the steganalysis feature components and the image classification results to provide the basis for removing the redundant steganalysis feature components. And then, by setting thresholds, the steganalysis feature components with a large difference function value are selected and with a small Pearson correlation coefficient are deleted. Finally, the steganalysis feature components retained are trained and detected as the final steganalysis feature. A series of experimental results indicate that this method can reduce the feature dimension effectively and the spatio-temporal complexity of steganalysis, while maintaining or even improving the detection accuracy of the stego images. Compared to existing steganalysis feature selection methods such as Fisher-GFR and Improved-Fisher, this method has a higher detection accuracy of the stego images after simplification.