基于深度学习的轻量化遥感图像目标检测方法

深度学习技术强大的特征表征能力为遥感图像目标检测提供了一个非常有效的工具。现有的主流深度学习模型参数数量巨大,存储和计算代价高,限制了其应用的推广。文中提出一种轻量化的深度学习模型用于遥感图像目标检测。实验结果表明,文中提出的方法在保持与Tiny_YOLOv3检测精度相当的情况下,模型大小仅为Tiny_YOLOv3的44. 7%。文中提出的模型在检测精度、模型大小和计算开销上可达到更好的平衡。     

Multi-task learning for video anomaly detection

We propose a multi-task learning framework for video anomaly detection based on a novel pipeline. Our model contains two crossing streams, one stream employs the backbone of Attention-R2U-net for future frame prediction, while the other is designed based on an encoder–decoder network to reconstruct the input optical flow maps. In addition, the latent layers of the two streams are merged together and assigned with a Deep SVDD-based loss at each location individually. Through the combination of these three tasks, the two-stream-crossing pipeline can be trained end-to-end to provide a comprehensive evaluation for the anomaly targets. Experimental results on several popular benchmark datasets show that our model outperforms the state-of-the-art competing models, which can be applied to different types of anomalous targets and meanwhile achieves remarkable precision.     

一种基于LSTM自动编码机的工业系统异常检测方法

在工业互联网的环境下,自动有效的异常检测方法对工业系统的安全、稳定生产具有重要的意义。传统的异常检测方法存在需要大量标注样本、不适应高维度时序数据等不足,提出一种基于LSTM自动编码机的工业系统异常检测方法。为克服现有方法依赖标注样本的不足,提出采用自动编码机,通过无监督的方式学习大量正常样本的特征和模式,在此基础上通过对样本进行重构和计算重构误差的方式进行异常检测。其次,为克服现有方法不适应高维度时序数据的不足,提出采用双向LSTM作为编码器,进而挖掘多维时序数据的潜在特征。基于一个真实造纸工业的数据集的实验表明,所提方法在各项指标上都对现有无监督异常检测方法有一定的提升,检测的总体精度达到了93.4%。     

Improving threat detection in networks using deep learning

Annals of Telecommunications - Detecting threats on the Internet is a key factor in maintaining data and information security. An intrusion detection system tries to prevent such attacks from...     

An anomaly intrusion detection method using fourier transform

A set of discrete points obtained from audit records on a behavior session is processed with Fourier transform. The criterion of selecting Fourier transform coefficients is introduced, and is used to find a unified value from the set of coefficients. This unified value is compared with a threshold to determine whether the session is abnormal. Finally simple test results are reported.     

An efficient artificial rabbits optimization-based user association problems in ultradense network using deep learning model

The ability of high splitting gain of dense small cells contributed to the rapid establishment of ultradense networks (UDNs). Its higher efficiency to deal with high traffic data demand made UDN a most-promising technology for the future 5G environment. However, the UDN creates concern about user association, which causes more complexities in providing a high data transmission rate and low latency rate. To tackle these complexities, in this paper, the ambient intelligence exploration multi-delay deep deterministic policy gradient-based artificial rabbits optimization (AEMDPG-ARO) algorithm is proposed for resolving data rate and the issues of latency in the small base station (SBS) and macro base station (MBS) of the wireless sensor network. The complexity in attaining lower latency and higher data rate is achieved through a novel technique AEMDPG-ARO. The ambient intelligence exploration multi-delay (AIEM) is combined with deep deterministic policy gradient (DDPG) for overcoming the local optimum and diversity issues of DDPG. The data sample for this study is obtained through the WINNER channel model. The proposed AEMDPG-ARO algorithm's efficiency is compared to varied state of art methods. The performance evaluation is carried out with regard to network lifetime, end-to-end delay, packet delivery ratio, sum rate overall energy consumption, latency, and minimum rate and maximum rate of the network. The proposed AEMDPG-ARO algorithm gives better performance with reduced time complexity and better metrics rate in the result analysis. The minimum latency achieved by the proposed AEMDPG-ARO algorithm is about 0.1 s.     

Enhancing IoT anomaly detection performance for federated learning

Federated Learning (FL) with mobile computing and the Internet of Things (IoT) is an effective cooperative learning approach. However, several technical challenges still need to be addressed. For instance, dividing the training process among several devices may impact the performance of Machine Learning (ML) algorithms, often significantly degrading prediction accuracy compared to centralized learning. One of the primary reasons for such performance degradation is that each device can access only a small fraction of data (that it generates), which limits the efficacy of the local ML model constructed on that device. The performance degradation could be exacerbated when the participating devices produce different classes of events, which is known as the class balance problem. Moreover, if the participating devices are of different types, each device may never observe the same types of events, which leads to the device heterogeneity problem. In this study, we investigate how data augmentation can be applied to address these challenges and improving detection performance in an anomaly detection task using IoT datasets. Our extensive experimental results with three publicly accessible IoT datasets show the performance improvement of up to 22.9% with the approach of data augmentation, compared to the baseline (without relying on data augmentation). In particular, stratified random sampling and uniform random sampling show the best improvement in detection performance with only a modest increase in computation time, whereas the data augmentation scheme using Generative Adversarial Networks is the most time-consuming with limited performance benefits.     

基于一种新的级联分类器的目标检测系统

在机器视觉领域中,目标检测是一个至关重要的研究问题,提出一种新的目标检测算法。该算法在MB-LBP特征级联分类器的基础上再串联两个分类器,分别为基于SIFT特征分类器和基于SURF特征分类器。首先,当测试图像通过MB-LBP特征级联分类器时,所有的目标都没漏检,但是有部分非目标被误检为目标,接着依次通过基于SIFT特征分类器和基于SURF特征分类器,检测结果只剩下目标和极少数非目标。实验结果表明,由MB-LBP特征级联分类器、基于SIFT特征分类器和基于SURF特征分类器构成的新级联分类器可以有效提高查准率。     

Weakly supervised learning of a classifier for unusual event detection

In this paper, we present an automatic classification framework combining appearance based features and hidden Markov models (HMM) to detect unusual events in image sequences. One characteristic of the classification task is that anomalies are rare. This reflects the situation in the quality control of industrial processes, where error events are scarce by nature. As an additional restriction, class labels are only available for the complete image sequence, since frame-wise manual scanning of the recorded sequences for anomalies is too expensive and should, therefore, be avoided. The proposed framework reduces the feature space dimension of the image sequences by employing subspace methods and encodes characteristic temporal dynamics using continuous hidden Markov models (CHMMs). The applied learning procedure is as follows. 1) A generative model for the regular sequences is trained (one-class learning). 2) The regular sequence model (RSM) is used to locate potentially unusual segments within error sequences by means of a change detection algorithm (outlier detection). 3) Unusual segments are used to expand the RSM to an error sequence model (ESM). The complexity of the ESM is controlled by means of the Bayesian Information Criterion (BIC). The likelihood ratio of the data given the ESM and the RSM is used for the classification decision. This ratio is close to one for sequences without error events and increases for sequences containing error events. Experimental results are presented for image sequences recorded from industrial laser welding processes. We demonstrate that the learning procedure can significantly reduce the user interaction and that sequences with error events can be found with a small false positive rate. It has also been shown that a modeling of the temporal dynamics is necessary to reach these low error rates.     

BotCatcher:botnet detection system based on deep learning

Machine learning technology has wide application in botnet detection.However,with the changes of the forms and command and control mechanisms of botnets,selecting features manually becomes increasingly difficult.To solve this problem,a botnet detection system called BotCatcher based on deep learning was proposed.It automatically extracted features from time and space dimension,and established classifier through multiple neural network constructions.BotCatcher does not depend on any prior knowledge which about the protocol and the topology,and works without manually selecting features.The experimental results show that the proposed model has good performance in botnet detection and has ability to accurately identify botnet traffic .     

An efficient approach for facial action unit intensity detection using distance metric learning based on cosine similarity

Emotions of human beings are largely represented by facial expressions. Facial expressions, simple as well as complex, are well decoded by facial action units. Any facial expression can be detected and analyzed if facial action units are decoded well. In the presented work, an attempt has been made to detect facial action unit intensity by mapping the features based on their cosine similarity. Distance metric learning based on cosine similarity maps the data by learning a metric that measures orientation rather than magnitude. The motivation behind using cosine similarity is that change in facial expressions can be better represented by changes in orientation as compared to the magnitude. The features are applied to support vector machine for classification of various intensities of action units. Experimental results on the popularly accepted database such as DISFA database and UNBC McMaster shoulder pain database confirm the efficacy of the proposed approach.     

基于深度学习的航拍电网绝缘子缺陷检测

电网绝缘检测缺陷的准确检测是电网运行状态有效监测及故障诊断前提,基于无人机航拍电网绝缘子图像,为解决深度学习缺陷检测存在的误检测和局部信息丢失问题,提出基于改进深度学习全卷积网络的缺陷自动检测算法,算法通过改进FCN的VGG结构、扩展滤波器尺寸、取消全连接层Dropout及模型深度,实现FCN模型在绝缘子缺陷检测方面的有效改进,实验结果表示,改进模型在较少运行时间增加基础上,有效提高了对绝缘子缺陷检测的性能和对背影的鲁棒性,取得了比已有算法更有优势的检测结果。     

深度递归网络在物联网系统异常检测中应用研究

物联网系统采用感知层、传输层和应用层三层体系架构,物联网的三层结构具有时序性,上一层数据异常会链式反应到后续层,传统的物联网异常检测方法无法有效识别数据异常并快速定位异常发生在哪一层。文中提出以深度递归网络对物联网系统异常检测进行建模,感知层、传输层、应用层作为深度网络输出层,深度递归网络通过核函数变换能够提取高阶特征,并且深度递归网络本身的时序特性能够提升异常检测的准确性。实验结果表明,深度递归网络在物联网系统异常检测中能够获得较高的检测准确率。     

基于深度学习的标签缺陷检测系统应用

依据深度学习算法可以自主进行特征学习和识别的特点,提出一种基于主成分分析法的深度学习模型,构建印刷标签检测系统,从中进行信息提取和学习处理。实验结果表明,该方法的分类识别率高达94.6%。与传统标签缺陷检测算法相比,这种研究方式更加注重原始图像的特征提取,无需进行复杂的模板制作,实现方法简单,适应性强。     

基于深度学习的软件定义网络应用策略冲突检测方法

在基于OpenFlow的软件定义网络（SDN）中,应用被部署时,相应的流表策略将被下发到OpenFlow交换机中,不同应用的流表项之间如果产生冲突,将会影响交换机的实际转发行为,进而扰乱特定应用的正确部署以及SDN的安全。随着SDN规模的扩大以及需要部署应用的数量的剧增,交换机中的流表数量呈现爆炸式增长。此时若采用传统的流表冲突检测算法,交换机将会耗费大量的系统计算时间。结合深度学习,首次提出了一种适合SDN中超大规模应用部署的智能流表冲突检测方法。实验结果表明,第一级深度学习模型的AUC达到97.04%,第二级模型的AUC达到99.97%,同时冲突检测时间与流表规模呈现线性增长关系。     

Keyed learning: An adversarial learning framework—formalization,challenges, and anomaly detection applications

We propose a general framework for keyed learning, where a secret key is used as an additional input of an adversarial learning system. We also define models and formal challenges for an adversary who knows the learning algorithm and its input data but has no access to the key value. This adversarial learning framework is subsequently applied to a more specific context of anomaly detection, where the secret key finds additional practical uses and guides the entire learning and alarm‐generating procedure.     

基于深度学习的电力设备红外与可见光图像配准

针对相对位置固定的红外和可见光双目相机配准任务,现有算法没有考虑到两者相对位置固定的先验知识,存在配准精度低、几何定位差异大等问题,适用性差。提出一种基于几何约束下区域搜索的红外可见光双目图像配准方法。首先借助红外和可见光双目相机的标定信息对红外和可见光图像进行立体校正使二者处于同一高度之上。接着借助于相位一致性计算红外与可见光的边缘特征图,然后在红外边缘图上提取特征点,最后提出两阶段的同名特征点搜索方法,以红外特征点为基准在可见光边缘图局部区域内搜索同名特征点。在第一阶段以归一化互相关（Normalized cross-correlation,NCC）为相似性度量计算两边缘图的整体水平偏移,预测同名特征点初始位置,在第二阶段提出多尺度加权NCC作为相似性度量,在初始同名特征点位置周围精确搜索同名特征点。在构造的真实环境数据集上进行实验,实验结果表明相对于其他对比算法,在特征点匹配数量和准确率以及主观视觉上的配准效果都优于其他对比算法。     

时间序列不确定数据流中异常数据检测方法

结合小波分析和不确定聚类方法的优点,提出一种基时间序列不确定数据流的异常数据检测方法,该方法主要考虑数据流中元组的不确定性,同时平衡检测的计算代价与检测精度。仿真实验证明,该检测方法能够良好地适应数据流的不确定性。在一定条件下可获得相当好的检测效果。     

An efficient tissue classifier for building patient-specific finiteelement models from X-ray CT images

The authors developed an efficient semiautomatic tissue classifier for X-ray computed tomography (CT) images which can be used to build patient- or animal-specific finite element (FE) models for bioelectric studies. The classifier uses a gray scale histogram for each tissue type and three-dimensional (3-D) neighborhood information. A total of 537 CT images from four animals (pigs) were classified with an average accuracy of 96.5% compared to manual classification by a radiologist. The use of 3-D, as opposed to 2-D, information reduced the error rate by 78%. Models generated using minimal or full manual editing yielded substantially identical voltage profiles. For the purpose of calculating voltage gradients or current densities in specific tissues, such as the myocardium, the appropriate slices need to be fully edited, however. The authors' classifier offers an approach to building FE models from image information with a level of manual effort that can be adjusted to the need of the application     

A simple and efficient hidden Markov model scheme for host-based anomaly intrusion detection

Extensive research activities have been observed on network-based intrusion detection systems (IDSs). However, there are always some attacks that penetrate trafficprofiling- based network IDSs. These attacks often cause very serious damages such as modifying host critical files. A host-based anomaly IDS is an effective complement to the network IDS in addressing this issue. This article proposes a simple data preprocessing approach to speed up a hidden Markov model (HMM) training for system-call-based anomaly intrusion detection. Experiments based on a public database demonstrate that this data preprocessing approach can reduce training time by up to 50 percent with unnoticeable intrusion detection performance degradation, compared to a conventional batch HMM training scheme. More than 58 percent data reduction has been observed compared to our prior incremental HMM training scheme. Although this maximum gain incurs more degradation of false alarm rate performance, the resulting performance is still reasonable.