Knowledge graph embedding models for link prediction often require a large number of model parameters, resulting in significant memory and computational costs that limit the scalability for large-scale knowledge graphs. To address the issue, this paper proposes a novel framework that balances expressiveness and computational efficiency through a dual-perspective modeling approach, named TransConvE, which combines a Transformer-based global dependency modeling with CNN-based local feature extraction for capturing both long-range dependencies and fine-grained local interactions. Experiments on standard link prediction benchmarks, FB15k-237 and WN18RR, demonstrate that the proposed TransConvE achieves competitive performance, reducing model parameters by an average of 66.7% compared to RotatE, TuckER, and CoKE, while maintaining comparable or even better effectiveness. This demonstrates the effectiveness of TransConvE in capturing complex patterns within knowledge graphs, balancing model complexity and performance, and ensuring scalability for large-scale knowledge graph embedding tasks.

To address the issue of low interpretability in deep learning-based methods for mining multi-layer features from synthetic aperture radar (SAR) target samples, the deep non-negative matrix factorization (DNMF) technique is proposed. However, DNMF needs to face the trade-off between high-precision recognition and efficient feature extraction as the number of SAR target samples increases. In this letter, we proposel2,1-constrained deep incremental NMF (l2,1-DINMF), an incremental formulation of DNMF withl2,1-constraint that effectively resolves the aforementioned dilemma. Within each layer,l2,1paradigm constraint is implemented to ensure the sparsity of the multi-layer feature extraction results. Additionally, the proposed incremental updating rule, derived from the approximate solution, exhibits faster convergence properties and significantly reduces time and memory loss in multi-layer feature extraction of DNMF. Several experimental results demonstrate the superiority of the proposed approach based on the MSTAR dataset. The proposed method consistently outperforms other NMF-based methods in terms of recognition performance, while exhibiting a time loss to reach the corresponding performance level that is merely 3.14% of the existing method. Moreover, the peak memory loss occupied by the proposed method in the feature extraction process is only 303.39KB.

In recent years, deep convolutional neural networks (DCNNs) have been widely used in the task of ship target detection in synthetic aperture radar (SAR) imagery. However, the vast storage and computational cost of DCNN limits its application to spaceborne or airborne onboard devices with limited resources. In this paper, a set of lightweight detection networks for SAR ship target detection are proposed. To obtain these lightweight networks, this paper designs a network structure optimization algorithm based on the multi-objective firefly algorithm (termed NOFA). In our design, the NOFA algorithm encodes the filters of a well-performing ship target detection network into a list of probabilities, which will determine whether the lightweight network will inherit the corresponding filter structure and parameters. After that, the multi-objective firefly optimization algorithm (MFA) continuously optimizes the probability list and finally outputs a set of lightweight network encodings that can meet the different needs of the trade-off between detection network precision and size. Finally, the network pruning technology transforms the encoding that meets the task requirements into a lightweight ship target detection network. The experiments on SSDD and SDCD datasets prove that the method proposed in this paper can provide more flexible and lighter detection networks than traditional detection networks.

The huge computing and storage requirements of deep con-volutional neural networks (DCNNs) limit their application on edge computing devices. In this article, we propose an attention mechanism based on the feature map quality evaluation algorithm (IQE). The knowledge distillation method based on the IQE attention mechanism uses the IQE method to identify important knowledge in the pre-trained SAR target recognition deep neural network. Then in the process of knowledge distillation, the lightweight network is forced to focus on the learning of important knowledge. Through this mechanism, the method proposed in this paper can efficiently transfer the knowledge of the pre-trained SAR target recognition network to the lightweight network, which makes it is possible to deploy the SAR target recognition algorithm on the edge computing platform. Comparison experiments with several commonly used knowledge distillation methods have proved the effectiveness of our proposed method. In addition, we also verified the performance of the lightweight network obtained by our method on the edge platform based on the K210 processor.

With the maturity of synthetic aperture radar (SAR) technology, the problem of imbalanced data has gradually emerged. This problem makes it difficult for the automatic target recognition (ATR) model to properly learn the classification boundaries of majority and minority category target samples. In this article, we propose an ATR model with new architecture, called the cost-sensitive awareness-based automatic target recognition (CA-ATR) model, which provides an effective way of solving the problem of imbalanced data. Aimed at the two issues caused by imbalanced data on ATR models, the proposed method solves the problems from both the data and algorithm levels. At the data level, CA-ATR avoids adverse correlations among the target samples through different oversampling methods. By making the ATR model cost-sensitive, the proposed method also avoids the empirical risk preference of the ATR model for majority category target samples at the algorithm-level. At the same time, CA-ATR can autonomously learn different cost-sensitive awareness from different imbalanced data sets. The awareness enables the ATR model to more accurately learn the classification boundaries between target samples that belong in different categories. Several experimental results show the superiority of the proposed approach based on the Moving and Stationary Target Acquisition and Recognition (MSTAR) data set. Compared with other imbalanced learning methods, the proposed method is able to solve different types of imbalanced data problems.

Since there are differences in the natural frequency of various synthetic aperture radar (SAR) target samples in reality, the problem of imbalanced data on the automatic target recognition (ATR) model has gradually appeared in recent years. The problem makes the classification boundary learned by the ATR model often fuzzy or even wrong. In this article, an SAR target sample generation method was proposed, called demand-driven generative adversarial nets (DDGANs), which provided an effective way to implement imbalanced data learning. When the imbalanced data exacerbated the deterioration of the minority category target samples distribution, the proposed method generated samples to alleviate this negative impact. The proposed method innovatively used two convolutional neural networks to form the discriminator of DDGAN. Among them, a convolutional neural network was used to determine whether the generated sample is real or fake. Moreover, another convolutional neural network can simultaneously dig out the generation demands of different categories of target samples when recognizing the generated samples. The generation demands enabled DDGAN to allocate different generation capabilities to different target samples on demand, thereby alleviating the negative impact of data imbalance. At the same time, DDGAN can autonomously learn the generation demands from imbalanced training sets. Several experimental results based on the Moving and Stationary Target Acquisition and Recognition (MSTAR) dataset showed the advantages of DDGAN. Compared with existing imbalanced learning algorithms, the proposed method had obvious superiority in recognition performance and data generation efficiency.

The rapid development of generative adversarial nets (GANs) has led to an increasing number of applications for the synthetic aperture radar (SAR) automatic target recognition (A-TR) with a small sample set in the past few years. However, the generated samples by the GAN s sometimes even lead to a decrease in the performance of the ATR model. In this paper, we propose a filtering approach to address this harm of generated samples. The proposed filtering approach is based on a stable generation model. The stable generation model can continuously and stably generate different batches of target samples. Then, multiple SVMs trained by different SAR target sample sets provide pseudo-labels to the other SVMs to improve the accuracy of the filtering results. Therefore, the proposed approach improves the recognition ability of the A-TR model dynamically while continuously filtering generated target samples. Several experimental results show the superiority of the proposed filtering approach based on the Moving and Stationary Target Acquisition and Recognition (MSTAR) data set. When the number of training samples is 14.5% of the original training set, the recognition rate of the ATR model still reaches 91.27% with the help of the proposed approach.

Deep convolutional neural networks (DCNNs) have been widely used for SAR image target recognition. However, the huge demands of DCNNs for computing, storage, and energy resources limit their use on edge computing devices. In this article, we propose a method based on image quality evaluation (IQE) criterion to prune deep neural networks. We use IQE criterion to identify unimportant filters, and then remove them, to obtain a lightweight network while maintaining the performance of the neural network as much as possible. Besides, we verified the effectiveness of our method on the MSTAR dataset with cheap edge computing devices.

Under the framework of a supervised learning-based automatic target recognition (ATR) approach, recognition performance is primarily dependent on the amount of training samples. However, shortage in training samples is a consistent issue for ATR. In this article, we propose a new image to image generation method, called label-directed generative adversarial networks (LDGANs), which will provide labeled samples to be used for recognition model training. We define an entirely new loss function for the LDGAN, which utilizes the Wasserstein distance to replace the original distance measurement of the conventional generative adversarial networks (GANs), thus efficiently avoiding the collapse mode problem. The label information is also added to the loss function of the LDGAN to avoid generating a large number of unlabeled target images. More importantly, the proposed method also makes corresponding changes to the network architecture regarding the new GANs. At the same time, the detailed algorithm about the LDGAN is also introduced in this article to deal with the issue that characteristically GANs are not easy to train. Based on comparisons with other directed generation methods, the experimental results show comparative results of several types of generated images in statistical features, gradient features, classic features of synthetic aperture radar (SAR) targets and the independence from the real image. While demonstrating that the images generated by the LDGAN produced better results using the assumptions of independent and identical distribution, the experiment also explores the performance of the generated image in the ATR. A comparison of these experimental results demonstrates a better way to use the generated image for ATR. The experimental results also prove that the proposed method does have the ability to supplement information for ATR when the training sample information is insufficient.

Noncontact gesture recognition is gradually being applied to emerging applications, such as smart cars and smart phones. Negative latency gesture recognition (recognition before a gesture is finished) is desirable due to the instantaneous feedback. However, it is difficult for existing methods to achieve a high precision and negative latency gesture recognition. A fragment can provide too few features to directly identify all gestures well. By observing a large number of existing gesture sets and people's daily operating habits, we found that some high frequency used gestures are similar. To the best of our knowledge, it is the first time to redivide the gestures into two subsets according to their movement physical states. We divided the gestures with different shapes or motion states into a parent-class subset, and further divided each pair of parent-class gestures to obtain a child-class subset. In order to achieve a better tradeoff between the high-precision and negative latency, an approach of motion pattern and behavior intention (MPBI) is proposed. Taking full advantage of the characteristics of each subset, MPBI includes two models. First, pattern model coarsely classify the parent-class gestures by a convolutional network, and then intention model further classifies child-class gestures according to their opposite motion direction. MPBI is evaluated on a 340-GHz terahertz radar. With the advantage of its accurate ranging, intention model can recognize child-class gestures directly without training. MPBI is evaluated on 12 gestures and achieves a recognition accuracy of 94.13%, which only needs a 0.033-s gesture fragment as an input sample.