Adding subtle perturbations to an image can cause the classification model to misclassify, and such images are called adversarial examples. Adversarial examples threaten the safe use of deep neural networks, but when ...
Adding subtle perturbations to an image can cause the classification model to misclassify, and such images are called adversarial examples. Adversarial examples threaten the safe use of deep neural networks, but when combined with reversible data hiding(RDH) technology, they can protect images from being correctly identified by unauthorized models and recover the image lossless under authorized models. Based on this, the reversible adversarial example(RAE) is rising. However, existing RAE technology focuses on feasibility, attack success rate and image quality, but ignores transferability and time complexity. In this paper,we optimize the data hiding structure and combine data augmentation technology,which flips the input image in probability to avoid overfitting phenomenon on the dataset. On the premise of maintaining a high success rate of white-box attacks and the image's visual quality, the proposed method improves the transferability of reversible adversarial examples by approximately 16% and reduces the computational cost by approximately 43% compared to the state-of-the-art method. In addition, the appropriate flip probability can be selected for different application scenarios.
A compact Sub-6GHz multiple-input-multiple-output (MIMO) antenna is presented in this paper. The recommended MIMO antenna is electrically small (38mm × 38 mm× 1.6 mm). For good isolation and miniaturized siz...
详细信息
This paper introduces a compact wideband 6 port multiple-input multiple-output (MIMO) antenna for the 5G handset. Each antenna elements uses coupling feed, and I-shaped grounding structure is added to the unit to obta...
详细信息
The transformational and spatial proximities are important cues for identifying inliers from an appearance based match set because correct matches generally stay close in input images and share similar local transform...
详细信息
Aiming at the traditional blind source separation for the received single-channel mixed radar signal makes it difficult to accurately recover the source signal, a single-channel radar signal comprehensive separation m...
ISBN:
(数字)9781837240982
Aiming at the traditional blind source separation for the received single-channel mixed radar signal makes it difficult to accurately recover the source signal, a single-channel radar signal comprehensive separation method based on the combination of the FastICA algorithm and the variational modal decomposition (VMD) algorithm is proposed. The method utilizes VMD to first perform modal separation of single-channel mixed Linear Frequency Modulation (LFM) signal, then selects the modal component signal with the largest correlation coefficient with the mixed signal and expands it with single-channel mixed LFM signal into a virtual multi-channel mixed signal, and finally inputs it into FastICA algorithm to obtain the source signal. The simulation experiment results show that the comprehensive separation algorithm can accurately separate the LFM signal, overcoming the limitation that the FastICA algorithm cannot directly separate the single-channel mixed radar signal.
CNNs (Convolutional Neural Networks) have a good performance on most classification tasks, but they are vulnerable when meeting adversarial examples. Research and design of highly aggressive adversarial examples can h...
CNNs (Convolutional Neural Networks) have a good performance on most classification tasks, but they are vulnerable when meeting adversarial examples. Research and design of highly aggressive adversarial examples can help enhance the security and robustness of CNNs. The transferability of adversarial examples is still low in black-box settings. Therefore, an adversarial example method based on probability histogram equalization, namely HE-MI-FGSM (Histogram Equalization Momentum Iterative Fast Gradient Sign Method) is proposed. In each iteration of the adversarial example generation process, the original input image is randomly histogram equalized, and then the gradient is calculated to generate adversarial perturbations to mitigate overfitting in the adversarial example. The effectiveness of the method is verified on the ImageNet dataset. Compared with the advanced method I-FGSM (Iterative Fast Gradient Sign Method) and MI-FGSM (Momentum I-FGSM), the attack success rate in the adversarial training network increased by 27.9% and 7.7% on average, respectively.
A $\boldsymbol{40}\times\boldsymbol{40}\times \boldsymbol{1.6}\mathbf{mm}^{3}$ ultra-wideband MIMO antenna is proposed in this paper. The antenna realizes four-notched band characteristics by embedding different typ...
详细信息
A $\boldsymbol{40}\times\boldsymbol{40}\times \boldsymbol{1.6}\mathbf{mm}^{3}$ ultra-wideband MIMO antenna is proposed in this paper. The antenna realizes four-notched band characteristics by embedding different types of slits in the radiating element. The antenna has good UWB performance, with four-band rejection to avoid interference caused by four narrow frequency bands WiMAX(3.3-3.8GHz), WLAN(S.1-S.8GHz), INSAT(6.7-7.3GHz) and ITV 8GHz (8.13-8.38GHz). It achieves better than - 15dB of isolation, and impedance bandwidth from 3.06-15.5GHz.
Despite the Graph Neural Networks' (GNNs) pro-ficiency in analyzing graph data, achieving high-accuracy and interpretable predictions remains challenging. Existing GNN interpreters typically provide post-hoc expla...
详细信息
ISBN:
(数字)9798350317152
ISBN:
(纸本)9798350317169
Despite the Graph Neural Networks' (GNNs) pro-ficiency in analyzing graph data, achieving high-accuracy and interpretable predictions remains challenging. Existing GNN interpreters typically provide post-hoc explanations disjointed from GNNs' predictions, resulting in misrepresentations. Self-explainable GNNs offer built-in explanations during the training process. However, they cannot exploit the explanatory outcomes to augment prediction performance, and they fail to provide high-quality explanations of node features and require additional processes to generate explainable subgraphs, which is costly. To address the aforementioned limitations, we propose a self-explained and self-supervised graph neural network (SES) to bridge the gap between explainability and prediction. SES comprises two processes: explainable training and enhanced predictive learning. During explainable training, SES employs a global mask generator co-trained with a graph encoder and directly produces crucial structure and feature masks, reducing time consumption and providing node feature and subgraph explanations. In the enhanced predictive learning phase, mask-based positive-negative pairs are constructed utilizing the ex-planations to compute a triplet loss and enhance the node representations by contrastive learning. Extensive experiments demonstrate the superiority of SES on multiple datasets and tasks. SES outperforms baselines on real-world node classification datasets by notable margins of up to 2.59% and achieves state-of-the-art (SOTA) performance in explanation tasks on synthetic datasets with improvements of up to 3.0%. Moreover, SES delivers more coherent explanations on real-world datasets, has a fourfold increase in Fidelity+ score for explanation quality, and demonstrates faster training and expla-nation generating times. To our knowledge, SES is a pioneering GNN to achieve SOTA performance on both explanation and prediction tasks.
Graph neural networks (GNNs) have shown great success in graph processing. However, current message-passing-based GNNs have limitations in feature aggregations and update mechanisms that rely on a fixed mode, resultin...
详细信息
Interpolation technology has evolved into a powerful tool for reversible data hiding in the image processing ***,existing interpolated algorithms only have a trivial impact on image *** this paper,an innovative interp...
详细信息
ISBN:
(数字)9789887581536
ISBN:
(纸本)9781665482561
Interpolation technology has evolved into a powerful tool for reversible data hiding in the image processing ***,existing interpolated algorithms only have a trivial impact on image *** this paper,an innovative interpolation and matrix-based algorithm is proposed.A novel concept of the difference between interpolated pixels is represented to dramatically improve the visual quality of the image,which lays a solid foundation for the subsequent data hiding *** is growing evidence that the Tetris matrix plays a vital role in improving embedding *** is worth mentioning that our scheme can intensely resist different attacks of various *** experimental findings demonstrate that the effect of our proposed scheme is unprecedentedly perfect even though a higher capacity is embedded than with traditional steganography approaches.
暂无评论