基于卷积神经网络的雷达目标航迹识别研究

现代战争中雷达信号日趋复杂,如何快速准确地从种类繁多、数据量庞大的雷达检测数据中,获取目标航迹的类别信息,为战场指挥提供准确有效的信息是当前急需解决的难题。传统基于人的经验认知的雷达目标航迹识别方法已经无法有效应对瞬息万变的战场和海量数据。根据实际雷达数据特点,提出了使用对数的雷达航迹预处理方法,并构建了基于卷积神经网络的深度学习模型,实现了对雷达对抗中的目标航迹的识别与检测。基于模拟生成的雷达目标航迹数据对提出的数据预处理方法和构建的模型进行测试;实验表明,所提出的方法能很好地实现对目标航迹的检测与识别。     

Benchmarking Attention-Based Interpretability of Deep Learning in Multivariate Time Series Predictions

The adaptation of deep learning models within safety-critical systems cannot rely only on good prediction performance but needs to provide interpretable and robust explanations for their decisions. When modeling complex sequences, attention mechanisms are regarded as the established approach to support deep neural networks with intrinsic interpretability. This paper focuses on the emerging trend of specifically designing diagnostic datasets for understanding the inner workings of attention mechanism based deep learning models for multivariate forecasting tasks. We design a novel benchmark of synthetically designed datasets with the transparent underlying generating process of multiple time series interactions with increasing complexity. The benchmark enables empirical evaluation of the performance of attention based deep neural networks in three different aspects: (i) prediction performance score, (ii) interpretability correctness, (iii) sensitivity analysis. Our analysis shows that although most models have satisfying and stable prediction performance results, they often fail to give correct interpretability. The only model with both a satisfying performance score and correct interpretability is IMV-LSTM, capturing both autocorrelations and crosscorrelations between multiple time series. Interestingly, while evaluating IMV-LSTM on simulated data from statistical and mechanistic models, the correctness of interpretability increases with more complex datasets.     

A Hybrid Analysis-Based Approach to Android Malware Family Classification

With the popularity of Android, malware detection and family classification have also become a research focus. Many excellent methods have been proposed by previous authors, but static and dynamic analyses inevitably require complex processes. A hybrid analysis method for detecting Android malware and classifying malware families is presented in this paper, and is partially optimized for multiple-feature data. For static analysis, we use permissions and intent as static features and use three feature selection methods to form a subset of three candidate features. Compared with various models, including k-nearest neighbors and random forest, random forest is the best, with a detection rate of 95.04%, while the chi-square test is the best feature selection method. After using feature selection to explore the critical static features contained in this dataset, we analyzed a subset of important features to gain more insight into the malware. In a dynamic analysis based on network traffic, unlike those that focus on a one-way flow of traffic and work on HTTP protocols and transport layer protocols, we focused on sessions and retained protocol layers. The Res7LSTM model is then used to further classify the malicious and partially benign samples detected in the static detection. The experimental results show that our approach can not only work with fewer static features and guarantee sufficient accuracy, but also improve the detection rate of Android malware family classification from 71.48% in previous work to 99% when cutting the traffic in terms of the sessions and protocols of all layers.     

Unsupervised Anomaly Detection with Distillated Teacher-Student Network Ensemble

We address the problem of unsupervised anomaly detection for multivariate data. Traditional machine learning based anomaly detection algorithms rely on specific assumptions of normal patterns and fail to model complex feature interactions and relations. Recently, existing deep learning based methods are promising for extracting representations from complex features. These methods train an auxiliary task, e.g., reconstruction and prediction, on normal samples. They further assume that anomalies fail to perform well on the auxiliary task since they are never trained during the model optimization. However, the assumption does not always hold in practice. Deep models may also perform the auxiliary task well on anomalous samples, leading to the failure detection of anomalies. To effectively detect anomalies for multivariate data, this paper introduces a teacher-student distillation based framework Distillated Teacher-Student Network Ensemble (DTSNE). The paradigm of the teacher-student distillation is able to deal with high-dimensional complex features. In addition, an ensemble of student networks provides a better capability to avoid generalizing the auxiliary task performance on anomalous samples. To validate the effectiveness of our model, we conduct extensive experiments on real-world datasets. Experimental results show superior performance of DTSNE over competing methods. Analysis and discussion towards the behavior of our model are also provided in the experiment section.     

Personalized Image Classification by Semantic Embedding and Active Learning

Currently, deep learning has shown state-of-the-art performance in image classification with pre-defined taxonomy. However, in a more real-world scenario, different users usually have different classification intents given an image collection. To satisfactorily personalize the requirement, we propose an interactive image classification system with an offline representation learning stage and an online classification stage. During the offline stage, we learn a deep model to extract the feature with higher flexibility and scalability for different users’ preferences. Instead of training the model only with the inter-class discrimination, we also encode the similarity between the semantic-embedding vectors of the category labels into the model. This makes the extracted feature adapt to multiple taxonomies with different granularities. During the online session, an annotation task iteratively alternates with a high-throughput verification task. When performing the verification task, the users are only required to indicate the incorrect prediction without giving the exact category label. For each iteration, our system chooses the images to be annotated or verified based on interactive efficiency optimization. To provide a high interactive rate, a unified active learning algorithm is used to search the optimal annotation and verification set by minimizing the expected time cost. After interactive annotation and verification, the new classified images are used to train a customized classifier online, which reflects the user-adaptive intent of categorization. The learned classifier is then used for subsequent annotation and verification tasks. Experimental results under several public image datasets show that our method outperforms existing methods.     

基于规范化样本拆分的轴承缺陷检测

表面缺陷对轴承的性能和寿命存在严重影响.近年来,深度学习在缺陷检测中发挥了重要的作用,然而对于轴承检测而言,缺陷样本的采集耗时耗力.选择轴承内径作为研究对象,根据轴承的对称性特性提出一种规范化样本拆分方法,可有效扩充轴承样本数据集.分别采用不同的样本处理方法,而后利用ResNet网络训练轴承缺陷检测模型,进行多组对比实...     

A traffic flow cellular automaton model to considering drivers’ learning and forgetting behaviour

It is known that the commonly used NaSch cellular automaton (CA) model and its modifications can help explain the internal causes of the macro phenomena of traffic flow.However,the randomization probability of vehicle velocity used in these models is assumed to be an exogenous constant or a conditional constant,which cannot reflect the learning and forgetting behaviour of drivers with historical experiences.This paper further modifies the NaSch model by enabling the randomization probability to be adjusted on the bases of drivers’ memory.The Markov properties of this modified model are discussed.Analytical and simulation results show that the traffic fundamental diagrams can be indeed improved when considering drivers’ intelligent behaviour.Some new features of traffic are revealed by differently combining the model parameters representing learning and forgetting behaviour.     

A Review on Machine Learning Approaches for Network Malicious Behavior Detection in Emerging Technologies

Network anomaly detection systems (NADSs) play a significant role in every network defense system as they detect and prevent malicious activities. Therefore, this paper offers an exhaustive overview of different aspects of anomaly-based network intrusion detection systems (NIDSs). Additionally, contemporary malicious activities in network systems and the important properties of intrusion detection systems are discussed as well. The present survey explains important phases of NADSs, such as pre-processing, feature extraction and malicious behavior detection and recognition. In addition, with regard to the detection and recognition phase, recent machine learning approaches including supervised, unsupervised, new deep and ensemble learning techniques have been comprehensively discussed; moreover, some details about currently available benchmark datasets for training and evaluating machine learning techniques are provided by the researchers. In the end, potential challenges together with some future directions for machine learning-based NADSs are specified.     

A Novel Lightweight Real-Time Traffic Sign Detection Integration Framework Based on YOLOv4

As a popular research direction in the field of intelligent transportation, various scholars have widely concerned themselves with traffic sign detection However, there are still some key issues that need to be further solved in order to thoroughly apply related technologies to real scenarios, such as the feature extraction scheme of traffic sign images, the optimal selection of detection methods, and the objective limitations of detection tasks. For the purpose of overcoming these difficulties, this paper proposes a lightweight real-time traffic sign detection integration framework based on YOLO by combining deep learning methods. The framework optimizes the latency concern by reducing the computational overhead of the network, and facilitates information transfer and sharing at diverse levels. While improving the detection efficiency, it ensures a certain degree of generalization and robustness, and enhances the detection performance of traffic signs in objective environments, such as scale and illumination changes. The proposed model is tested and evaluated on real road scene datasets and compared with the current mainstream advanced detection models to verify its effectiveness. In addition, this paper successfully finds a reasonable balance between detection performance and deployment difficulty by effectively reducing the computational cost, which provides a possibility for realistic deployment on edge devices with limited hardware conditions, such as mobile devices and embedded devices. More importantly, the related theories have certain application potential in technology industries such as artificial intelligence or autonomous driving.     

Progressive Traffic-Oriented Resource Management for Reducing Network Congestion in Edge Computing

Edge computing can deliver network services with low latency and real-time processing by providing cloud services at the network edge. Edge computing has a number of advantages such as low latency, locality, and network traffic distribution, but the associated resource management has become a significant challenge because of its inherent hierarchical, distributed, and heterogeneous nature. Various cloud-based network services such as crowd sensing, hierarchical deep learning systems, and cloud gaming each have their own traffic patterns and computing requirements. To provide a satisfactory user experience for these services, resource management that comprehensively considers service diversity, client usage patterns, and network performance indicators is required. In this study, an algorithm that simultaneously considers computing resources and network traffic load when deploying servers that provide edge services is proposed. The proposed algorithm generates candidate deployments based on factors that affect traffic load, such as the number of servers, server location, and client mapping according to service characteristics and usage. A final deployment plan is then established using a partial vector bin packing scheme that considers both the generated traffic and computing resources in the network. The proposed algorithm is evaluated using several simulations that consider actual network service and device characteristics.     

Two-Party Privacy-Preserving Set Intersection with FHE

A two-party private set intersection allows two parties, the client and the server, to compute an intersection over their private sets, without revealing any information beyond the intersecting elements. We present a novel private set intersection protocol based on Shuhong Gao’s fully homomorphic encryption scheme and prove the security of the protocol in the semi-honest model. We also present a variant of the protocol which is a completely novel construction for computing the intersection based on Bloom filter and fully homomorphic encryption, and the protocol’s complexity is independent of the set size of the client. The security of the protocols relies on the learning with errors and ring learning with error problems. Furthermore, in the cloud with malicious adversaries, the computation of the private set intersection can be outsourced to the cloud service provider without revealing any private information.     

基于优化残差网络的复杂纹理表面缺陷检测

产品表面缺陷检测是工业自动化生产的重要环节,准确率是评价自动检测系统可靠性的主要指标。基于复杂纹理表面缺陷检测的特殊性以及对检测方法的实时性、通用性等要求,提出了优化骨干网络并使用迁移学习特征映射构建复杂纹理表面缺陷的检测方法。该方法通过优化残差网络模型并建立仿真数据集的方式进行迁移学习,以解决实际情况中复杂纹理表面产品数据集样本数量少、数据集制作困难、相似问题难以互相兼容等问题。实验结果表明,提出的方法可以准确地检测随机复杂纹理的人造木质板材表面缺陷,平均准确率可达99.6%。现有实验条件下单张人造木质板材的检测时间为305 ms,可以满足在线检测的实时性要求。研究结果可为基于深度学习的复杂纹理表面缺陷检测提供新的思路与理论参考。     

Ultrasonic flaw detection using radial basis function networks (RBFNs)

Ultrasonic flaw detection has been studied many times in the literature. Schemes based on thresholding after a previous matched filter use to be the best solution, but results obtained with this method are only satisfactory when scattering and attenuation are not considered. In this paper, we propose an alternative solution to thresholding detection method. We deal with the usage of different flaw detection methods comparing them with the proposed one. The experiment tries to determinate whether a given ultrasonic signal contains a flaw echo or not. Starting with a set of 24,000 patterns with 750 samples each one, two subsets are defined for the experiments. The first one, the training set, is used to obtain the detection parameters of the different methods, and the second one is used to test the performance of them. The proposed method is based on radial basis functions networks, one of the most powerful neural network techniques. This signal processing technique tries to find the optimal decision criterion. Comparing this method with thresholding based ones, an improvement over 25-30% is obtained, depending on the probability of false alarm. So our new method is a good alternative to flaw detection problem.     

Fatigue Detection with Spatial-Temporal Fusion Method on Covariance Manifolds of Electroencephalography

With the increasing pressure of current life, fatigue caused by high-pressure work has deeply affected people and even threatened their lives. In particular, fatigue driving has become a leading cause of traffic accidents and deaths. This paper investigates electroencephalography (EEG)-based fatigue detection for driving by mining the latent information through the spatial-temporal changes in the relations between EEG channels. First, EEG data are partitioned into several segments to calculate the covariance matrices of each segment, and then we feed these matrices into a recurrent neural network to obtain high-level temporal information. Second, the covariance matrices of whole signals are leveraged to extract two kinds of spatial features, which will be fused with temporal characteristics to obtain comprehensive spatial-temporal information. Experiments on an open benchmark showed that our method achieved an excellent classification accuracy of 93.834% and performed better than several novel methods. These experimental results indicate that our method enables better reliability and feasibility in the detection of fatigued driving.     

A New Deep Dual Temporal Domain Adaptation Method for Online Detection of Bearings Early Fault

With the quick development of sensor technology in recent years, online detection of early fault without system halt has received much attention in the field of bearing prognostics and health management. While lacking representative samples of the online data, one can try to adapt the previously-learned detection rule to the online detection task instead of training a new rule merely using online data. As one may come across a change of the data distribution between offline and online working conditions, it is challenging to utilize the data from different working conditions to improve detection accuracy and robustness. To solve this problem, a new online detection method of bearing early fault is proposed in this paper based on deep transfer learning. The proposed method contains an offline stage and an online stage. In the offline stage, a new state assessment method is proposed to determine the period of the normal state and the degradation state for whole-life degradation sequences. Moreover, a new deep dual temporal domain adaptation (DTDA) model is proposed. By adopting a dual adaptation strategy on the time convolutional network and domain adversarial neural network, the DTDA model can effectively extract domain-invariant temporal feature representation. In the online stage, each sequentially-arrived data batch is directly fed into the trained DTDA model to recognize whether an early fault occurs. Furthermore, a health indicator of target bearing is also built based on the DTDA features to intuitively evaluate the detection results. Experiments are conducted on the IEEE Prognostics and Health Management (PHM) Challenge 2012 bearing dataset. The results show that, compared with nine state-of-the-art fault detection and diagnosis methods, the proposed method can get an earlier detection location and lower false alarm rate.     

基于深度学习的故障检测方法

针对传统故障诊断方法中特征提取技术难度大、故障样本获取困难等问题,在深度学习计算框架下提出了一种半监督训练的故障检测方法,利用深度信念网络中的受限波茨曼机堆栈结构实现了数据高层特征的自动提取,结合支持向量数据描述方法实现了异常数据检测,只需利用正常工况的数据样本进行网络训练和模型拟合,无需故障样本数据,也无需人工干预进行信号特征提取,即能实现对故障数据进行的实时检测和判别。经采用标准轴承实验数据的三组故障数据进行验证,故障识别率达到100%,具有很强的工程应用价值。     

一种基于无监督主动学习的苹果品质光谱无损检测模型构建方法

利用光谱技术实现农产品、食品品质无损检测的实质是建立样本光谱信息与样本品质参数之间的机器学习模型。为了获得具有良好泛化性能的机器学习模型,通常需要大量的标记样本,然而,获取样本的光谱信息相对容易,但标注样本品质参数的过程往往涉及到大量的时间和经济成本,并且具有破坏性。主动学习是一种减少训练集有标记样本数量的方法,通过选择最有价值的样本进行标记,而不是随机选择。因此,主动学习能够控制向训练集添加哪些样本,模型不再是被动地接受用于建模的样本。在分类任务中已经提出较多关于主动学习的算法,但回归任务中的研究却相对较少,且现有的用于回归任务的主动学习算法大多是有监督的,即需要少量有标记样本训练初始模型。本文提出了一种基于无监督主动学习方法的训练样本选择策略。该方法首先通过层次凝聚聚类对无标记（标准值）光谱数据集进行多样性划分,获得不同的聚类簇;然后通过局部线性重建算法在每个聚类簇中选择最具代表性的样本构成训练样本集,最后基于训练集构建模型。利用两个年份三个品种苹果的近红外光谱数据,构建了其可溶性固形物含量和硬度的偏最小二乘预测模型,用于验证所提出方法的有效性。实验结果表明：所提出的方法要优于已有的样本选择策略,可以有效地提高模型精度,减少在模型训练中的破坏性理化实验。同时,与随机采样（RS）、Kennard-Stone算法（KS）、光谱-理化值共生距离算法（SPXY）这三种光谱领域常用的样本选择算法相比,该研究所提出的方法表现出了最佳的性能, 基于所提出的无监督主动学习算法选取200个样本作为训练集所建立的可溶性固形物含量预测模型的预测均方根误差相对于其他三种算法降低了2.0%~13.2%,硬度预测模型的预测均方根误差相对降低了1.2%~15.7%。     

A new macro model with consideration of the traffic interruption probability

In this paper, we present a new macro model which involves the effects that the probability of traffic interruption has on the car-following behavior through formulating the inner relationship between micro and macro variables. Linear stability analysis shows that consideration of the traffic interruption probability can improve the stability of traffic flow if and only if the drivers’ reactive time required for adjusting their acceleration based on the traffic interruption probability p is not greater than that one based on the non-interruption probability 1−p. Numerical results verify that the new model can be used to analyze the effects of traffic interruption probability and traffic interruption on shock, rarefaction wave, small perturbation and uniform flow. The model has been applied in reproducing some complex traffic phenomena resulted by some traffic interruptions (e.g., signal light, pedestrian and tolling station).     

基于机器学习的偏振遥感云检测优化算法

偏振遥感经验阈值云检测算法受主观因素影响较强,极易在亮地表上空出现云检测不准确的问题.针对该问题,本文提出了一种主动和被动遥感卫星相结合的机器学习云检测算法.该算法基于POLDER3载荷多通道多角度偏振特性以及CALIOP载荷高精度云垂直特性展开研究,利用POLDER3载荷和CALIOP载荷观测重合区域数据,搭建了粒子...     

Adaptive Block-Based Compressed Video Sensing Based on Saliency Detection and Side Information

The setting of the measurement number for each block is very important for a block-based compressed sensing system. However, in practical applications, we only have the initial measurement results of the original signal on the sampling side instead of the original signal itself, therefore, we cannot directly allocate the appropriate measurement number for each block without the sparsity of the original signal. To solve this problem, we propose an adaptive block-based compressed video sensing scheme based on saliency detection and side information. According to the Johnson–Lindenstrauss lemma, we can use the initial measurement results to perform saliency detection and then obtain the saliency value for each block. Meanwhile, a side information frame which is an estimate of the current frame is generated on the reconstruction side by the proposed probability fusion model, and the significant coefficient proportion of each block is estimated through the side information frame. Both the saliency value and significant coefficient proportion can reflect the sparsity of the block. Finally, these two estimates of block sparsity are fused, so that we can simultaneously use intra-frame and inter-frame correlation for block sparsity estimation. Then the measurement number of each block can be allocated according to the fusion sparsity. Besides, we propose a global recovery model based on weighting, which can reduce the block effect of reconstructed frames. The experimental results show that, compared with existing schemes, the proposed scheme can achieve a significant improvement in peak signal-to-noise ratio (PSNR) at the same sampling rate.