Steganography has made great progress over the past few years due to the advancement of deep convolutional neural networks (DCNNs), which have been successfully used to multi-domains. Correspondingly the performance of steganalysis models inevitably encounters a bottleneck since the CNN based steganography models perform better. In this paper, we propose an Advanced General convolutional neural Network (AG-Net) for steganalysis to deal with this problem. We firstly design a confrontation module to extract and compare features of cover and stego images, which are captured from an unknown steganography network. Then, we construct the association between two adjacent confrontation modules according to the feature comparison of the previous module, to accumulate the differences of mid-and high-level features between the cover and stego images. Thirdly, we deliver the loss of the last confrontation module to a softmax layer after batch normalization and scalarization, to classify and detect stego images. Extensive experiments and evaluations demonstrate that the proposed AG-Net can achieve promising performance in response to different challenging steganographic algorithms.
Convolutional neural networks (CNNs) have been successfully used in remote sensing scene classification and identification due to their ability to capture deep spatial feature representations. However, the performance of deep models inevitably encounters a bottleneck when multimodality-dominated scene classification rather than single-modality-dominated scene classification is performed, due to the high similarity among different categories. In this study, we propose a novel multi-granularity fusion convolutional neural network (MGFN) to automatically capture the latent ontological features of remote sensing images. We firstly design a multigranularity module that can progressively crop input images to learn multigrained features, which can describe images to different degrees. Based on a comparison of different granularities, we then design a maxout-based module to learn the corresponding Gaussian covariance matrices of different granularities, which can extract second-order features to express the latent ontological essence of inputs and select the most distinguished inputs. We thirdly provide an adaptive fusion module to fuse all features via normalization to combine features of different degrees using the adaptive fused module. Finally, an SVM classifier is used to classify the fused matrix of every input image. Extensive experimentation and evaluations, particularly for multimodality-dominated scenes, demonstrate that the proposed network can achieve promising results for public remote sensing datasets.
In recent years, several steganalysis neural networks have been proposed and achieved satisfactory performances. However, these deep learning methods all encounter the problem of Training-Images-Shortage (TIS). In most cases, it is difficult for steganalyses to obtain enough signals about steganography from a game opponent. In order to solve the problem of TIS for steganalysis, we propose a novel steganalysis method based on generative learning and deep residual convolutional neural networks. Comparative experiments show that the proposed architecture can achieve promising performance in response to spatial domain steganalysis despite a lack of training images.
4 January 2022 in Discrete Dynamics in Nature and Society
Mission planning of air strike operations is hard because it has to give instructions to a large number of units during a relatively long period of time in an uncertain environment. If some instruction parameters can be calculated by an intelligent agent, better strategies can be found more quickly. In a specific combat scenario of air strike operations against islands, an intelligent model is proposed to improve the performance and flexibility of mission planning. The proposed intelligent mission planning model is based on rule‐based decision and uses a fully connected recurrent neural network to calculate some of the decision parameters. The proposed intelligent mission planning model shows better results as compared to rule‐based decision making with randomized parameters, and it performs as good as experts in the test set of the specific combat scenario.
Steganalysis based on deep learning has made noticeable progress over the past few years where the training is all based on paired images. However, scenes without paired training data exist. We present an architecture for learning to generate corresponding pseudo stego image from a cover-image in the absence of paired training images. We seek a mapping G that can generate pseudo stego images indistinguishable from the real but unpaired stego images using an adversarial loss. Because this mapping is highly under-constrained, we designed a CycleGAN and introduce spectrum of stego images to reinforce the adversarial loss. Qualitative comparisons demonstrate the superiority of our approach.