检索结果-内蒙古大学图书馆

Concept Drift detection: Dealing With Missing Values via Fuzzy Distance Estimations

IEEE TRANSACTIONS ON FUZZY SYSTEMS 2021年第11期29卷 3219-3233页

作者： Liu, Anjin Lu, Jie Zhang, Guangquan Univ Technol Sydney Fac Engn & Informat Technol Decis Syst & E Serv Intelligence Lab Australian Artificial Intelligence Inst Sydney NSW 2007 Australia

In data streams, the data distribution of arriving observations at different time points may change-a phenomenon called concept drift. While detecting concept drift is a relatively mature area of study, solutions to the uncertainty introduced by observations with missing values have only been studied in isolation. No one has yet explored whether or how these solutions might impact drift detection performance. We, however, believe that data imputation methods may actually increase uncertainty in the data rather than reducing it. We also conjecture that imputation can introduce bias into the process of estimating distribution changes during drift detection, which can make it more difficult to train a learning model. Our idea is to focus on estimating the distance between observations rather than estimating the missing values, and to define membership functions that allocate observations to histogram bins according to the estimation errors. Our solution comprises a novel masked distance learning (MDL) algorithm to reduce the cumulative errors caused by iteratively estimating each missing value in an observation and a fuzzy-weighted frequency (FWF) method for identifying discrepancies in the data distribution. The concept drift detection algorithm proposed in this article is a singular and unified algorithm that can handle missing values, but not an imputation algorithm combined with a concept drift detection algorithm. Experiments on both synthetic and real-world datasets demonstrate the advantages of this method and show its robustness in detecting drift in data with missing values. The results show that compared to the best-performing algorithm that handles imputation and drift detection separately, MDL-FWF reduced the average drift detection difference from 10.75% to 5.83%. This is a nearly 46% improvement. These findings reveal that missing values exert a profound impact on concept drift detection, but using fuzzy set theory to model observations can produce more r

关键词： Task analysis Uncertainty Fuzzy set theory Estimation Histograms detection algorithms Computer aided instruction Concept drift fuzzy clustering fuzzy distance fuzzy weighting machine learning missing value

来源：评论

学校读者我要写书评

暂无评论

Target detection in Clutter/Interference Regions Based on Deep Feature Fusion for HFSWR

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 2021年 14卷 5581-5595页

作者： Wu, Maokai Zhang, Ling Niu, Jiong Wu, Q. M. Jonathan Ocean Univ China Coll Engn Qingdao 266100 Peoples R China Univ Windsor Dept Elect & Comp Engn Windsor ON N9B 3P4 Canada

High-frequency surface wave radar (HFSWR) is of great significance for maritime detection, but in the HFSWR echo signal, ship targets are often submerged in a variety of clutter and interference, making it difficult to detect vessels. In this paper, we propose an intelligent detection algorithm for targets concealed in strong clutter and complex interference environments. The algorithm has two stages: preprocessing and target detection. In the preprocessing stage, faster region-based convolutional neural networks Faster R-CNN are designed to identify and locate clutter and interference regions in the range Doppler spectrum;in the target detection stage, a two-level cascade algorithm is proposed. First, an extremum detection algorithm is proposed to identify suspicious target points in the clutter/interference regions, including real and false target points, to quickly obtain potential target positions. Second, in consideration of the characteristics of radar targets, two lightweight networks are designed to extract the CNN features and the stacked autoencoder features of the potential target locations. Then, fusion features are obtained and sent to an extreme learning machine that acts as a second-level classifier to distinguish between real and false target points. Experiments show that the proposed HFSWR target-detection algorithm has better performance for vessel detection in clutter/interference regions than the current mainstream detection algorithms.

关键词： Feature extraction Clutter Object detection Doppler effect Classification algorithms detection algorithms Detectors Convolutional neural network (CNN) extremum detection feature fusion high-frequency surface wave radar (HFSWR) stacked autoencoder (SAE) target detection

来源：评论

学校读者我要写书评

暂无评论

Near-ML detection Based on Multilevel SINR Thresholding for Mode Division Multiplexing Transmission With Mode-Dependent Loss

引用

IEEE ACCESS 2021年 9卷 20136-20142页

作者： Gong, Siyu Zhang, Jianyong Mi, Shuchao Fan, Fengju Yan, Baorui Beijing Jiaotong Univ Inst Lightwave Technol Beijing 100044 Peoples R China Beijing Jiaotong Univ Key Lab All Opt Network & Adv Telecommun Network Beijing 100044 Peoples R China

The probability distribution of the signal-to-interference plus noise ratio (SINR) is analyzed and approximated by the generalized gamma distribution for mode-dependent loss (MDL)-impaired fiber links. Based on the characteristics of SINR, an improved reduced-search algorithm (IRS) based on multilevel SINR thresholding (IRS-MST) is proposed. IRS-MST dynamically determines the size of the search space for each mode based on multiple SINR thresholds. Furthermore, a compounding function is used to balance the size of the search space and performance. Simulation results show that IRS-MST can achieve near ML performance for different MDLs. Compared to the conventional IRS algorithm, the complexity of IRS-MST is approximately 5.4 times smaller than that of the IRS.

关键词： Signal to noise ratio Interference Complexity theory MIMO communication Gamma distribution detection algorithms Estimation Mode division multiplexing mode-dependent loss MIMO detection reduced search SINR

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Based Object detection Improvement for Fine-Grained Birds

引用

IEEE ACCESS 2021年 9卷 67901-67915页

作者： Yang, Kuihe Song, Ziying Hebei Univ Sci & Technol Sch Informat Engn & Technol Shijiazhuang 050018 Hebei Peoples R China

When the object detection algorithm is applied to the bird protection project, there are many problems like large model parameters, high similarity between bird species and single sample scene. In order to further improve the detection accuracy and stability of the object detection model, a multi-object detection algorithm for fine-grained birds is proposed. Firstly, the algorithm introduces Depthwise separable convolution into the feature extraction layer of YOLOv3 algorithm. The convolution process is divided into two parts: deep convolution and point-by-point convolution. The separation between intra-channel convolution and inter-channel convolution is realized. On the basis of high detection accuracy, the number of algorithm model parameters and calculation amount are greatly reduced. Finally, Focal loss was added to the loss function to solve the serious imbalance of positive and negative samples. By reducing the weight of the large number of simple background classes, the algorithm was more focused on detecting foreground classes. The experimental results show that, in the bird data set, the average precision mean (mAP) index of this algorithm is 2.71% higher than YOLOv3 algorithm, the number of parameters is 79.88% lower than YOLOv3 basic model, and the number of frames per second (FPS) is 19.98% higher than YOLOv3 algorithm. This algorithm not only greatly reduces the number of model parameters and computation, but also improves the detection speed and mAP.

关键词： Birds Convolution Object detection Training detection algorithms Classification algorithms Feature extraction YOLOv3 algorithm Depthwise separable convolution Focal loss object detection

来源：评论

学校读者我要写书评

暂无评论

Intrusion detection for Cybersecurity of Smart Meters

引用

IEEE TRANSACTIONS ON SMART GRID 2021年第1期12卷 612-622页

作者： Sun, Chih-Che Cardenas, D. Jonathan Sebastian Hahn, Adam Liu, Chen-Ching Washington State Univ Sch Elect Engn & Comp Sci Pullman WA 99164 USA Virginia Polytech Inst & State Univ Power & Energy Ctr Blacksburg VA 24061 USA

The integration of Information and Communications Technology (ICT) enables real-time communication for smart meters to participate in power system operations. However, Advanced Metering Infrastructures (AMI) are vulnerable to cyber attacks. Both utilities and power consumers may become victims of cyber intrusions. In this paper, a two-stage cyber intrusion protection system is proposed. At the first stage of intrusion detection, a Support Vector Machine (SVM) is used as a detection algorithm to discover suspicious behaviors inside a smart meter. At the second stage, the Temporal Failure Propagation Graph (TFPG) technique is used to generate attack routes for identifying attack events. Finally, the proposed pattern recognition algorithm is used to calculate the similarity between a detected abnormal event and pre-defined cyber attacks. A higher similarity value implies a higher chance that a smart meter is under attack. An AMI security test platform has been developed to: (1) Collect training/testing data for SVM, (2) Simulate and analyze cyber attack events, and (3) Validate the proposed cyber attack protection system. The test platform consists of Network-Simulator 3 (NS-3) software to simulate an AMI network environment and single board computers (SBCs) to emulate the IEEE 802.15.4 communication between a grid router and a smart meter.

关键词： Smart meters Cyberattack Intrusion detection Support vector machines Meters detection algorithms Random access memory Advanced metering infrastructure (AMI) smart meters cyber-physical system security intrusion detection

来源：评论

学校读者我要写书评

暂无评论

A Test Bed for Detecting False Data Injection Attacks in Systems With Distributed Energy Resources

引用

IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS 2022年第1期10卷 1303-1315页

作者： Albunashee, Hamdi M. Farnell, Chris Suchanek, Andrew Haulmark, Kelby McCann, Roy A. Di, Jia Mantooth, Alan Univ Arkansas Dept Elect Engn Fayetteville AR 72701 USA Univ Arkansas Fayetteville AR 72701 USA Sandia Natl Labs Albuquerque NM 87123 USA Univ Arkansas Dept Comp Sci & Comp Engn Fayetteville AR 72701 USA

Previous theoretical studies have identified possible conditions for an undetectable false data injection attack (FDIA) vector that bypasses bad data detectors in power system state estimators. The formulation of the FDIA vector has attracted many researchers to propose detection algorithms. However, neither the attack formulation nor the proposed detection algorithms have been experimentally proven or validated. In this article, a framework for developing cybersecurity test beds that evaluate FDIA vectors for utility-scale distributed energy resources (DERs) and SE in real-time is proposed and implemented. The test bed is a combination of: 1) simulation models of utility-scale DER and distribution systems using OPAL-RT;2) intelligent electronic devices (SEL-351S, SEL-311C, and SEL-3355);3) network devices;and 4) graphical user interface (LabVIEW). The combination of actual devices, modeled systems, and cyber-physical interfaces adds realistic constraints, such as sampling error, network jitter, and data time alignment that exist during actual power system operations. Using the test bed framework, results show that the onset of the FDIA against a state estimator is experimentally detectable. Next, the test bed is used to expose an unknown vulnerability of existing phase-locked loop (PLL) techniques by presenting a new class of attack that causes PLL reporting of false phase angle and frequency values to the DER controllers. This type of attack would, then, induce the DERs to perform an adverse control action. In this article, a new detection algorithm based on a Kalman filter is evaluated to detect this type of PLL attack. A detailed evaluation of various attack scenarios is presented. The results indicate the benefits of employing this type of test bed framework to improve the development of cybersecurity defenses for realistic inverter-based DER environments.

关键词： Power systems detection algorithms Security State estimation Detectors Real-time systems Phase locked loops Cybersecurity distributed energy resources (DERs) state estimation (SE)

来源：评论

学校读者我要写书评

暂无评论

Stationary Marine Target detection With Time-Series SAR Imagery

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 2021年 14卷 6406-6413页

作者： An, Wentao Lin, Mingsen Yang, Haijun Minist Nat Resources Natl Satellite Ocean Applicat Serv Key Lab Space Ocean Remote Sensing & Applicat Beijing 100081 Peoples R China Hainan Trop Ocean Univ Remote Sensing Sanya 572022 Peoples R China

Stationary marine targets, such as oil rigs and offshore wind turbines, usually show up like bright spots without or with trivial position shifts in multitemporal synthetic aperture radar (SAR) imagery. They will trigger considerable false alarms in target detection applications tasked with ship detection, if they are not identified. An algorithm for stationary marine target detection, developed based on the assumption that the apparent positions of a stationary target in multitemporal imagery are also stationary or nearly so, is proposed in this article. This algorithm requires a strict time-series of SAR images in temporal sequence as input. For each input SAR image, all targets on sea surface are initially detected with an iterative cell-averaging constant false-alarm rate detection algorithm, and their longitude and latitude positions are then used to identify whether they are stationary targets. Under this algorithm, a stationariness index with five levels ("unknown target," "suspected new stationary target," "stationary target," "suspected removed stationary target," and "removed target") is defined for each target and must be iteratively updated with the latest level of identification. The proposed algorithm is promising for monitoring the status of stationary marine targets over a large sea area, because the processing of all input SAR images follows the same procedure and meanwhile the stationariness index is generated and kept updated. Two examples with GF-3 and RADARSAT-2 images are presented to illustrate the effectiveness of the proposed algorithm in detecting both offshore wind turbines and oil rigs.

关键词： Radar polarimetry Object detection Oils Synthetic aperture radar Indexes detection algorithms Databases Offshore oil rig detection offshore wind turbine detection stationary marine target detection synthetic aperture radar (SAR) target detection

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Eye Blink Artifact detection From EEG With Gaussian Mixture Model

引用

IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS 2021年第8期25卷 2895-2905页

作者： Cao, Jiuwen Chen, Long Hu, Dinghan Dong, Fang Jiang, Tiejia Gao, Weidong Gao, Feng Hangzhou Dianzi Univ Machine Learning I Hlth Int Cooperat Base Zhejian Hangzhou 310018 Zhejiang Peoples R China Hangzhou Dianzi Univ Artificial Intelligence Inst Hangzhou 310018 Zhejiang Peoples R China Zhejiang Lab Res Ctr Intelligent Sensing Hangzhou 311100 Zhejiang Peoples R China Zhejiang Univ City Coll Sch Informat & Elect Engn Hangzhou 310015 Zhejiang Peoples R China Zhejiang Univ Childrens Hosp Natl Clin Res Ctr Child Hlth Dept NeurolSch Med Hangzhou 310003 Zhejiang Peoples R China Chinese Univ Hong Kong Sch Sci & Engn Shenzhen Hong Kong Peoples R China

Eye blink is one of the most common artifacts in electroencephalogram (EEG) and significantly affects the performance of the EEG related applications, such as epilepsy recognition, spike detection, encephalitis diagnosis, etc. To achieve an accurate and efficient eye blink detection, a novel unsupervised learning algorithm based on a hybrid thresholding followed with a Gaussian mixture model (GMM) is presented in this paper. The EEG signal is priliminarily screened by a cascaded thresholding method built on the distributions of signal amplitude, amplitude displacement, as well as the cross channel correlation. Then, the channel correlation of the two frontal electrodes (FP1, FP2), the fractal dimension, and the mean of amplitude difference between FP1 and FP2, are extracted to characterize the filtered EEGs. The GMM trained on these features is applied for the eye blink detection. The performance of the proposed algorithm is studied on two EEG datasets collected by the Temple University Hospital (TUH) and the Children's Hospital, Zhejiang University School of Medicine (CHZU), where the datasets are recorded from epilepsy and encephalitis patients, and contain a lot of eye blink artifacts. Experimental results show that the proposed algorithm can achieve the highest detection precision and F1 score over the state-of-the-art methods.

关键词： Electroencephalography Correlation Epilepsy Electrodes detection algorithms Brain modeling Support vector machines Eye blink artifact detection unsupervised learning GMM amplitude displacement channel correlation

来源：评论

学校读者我要写书评

暂无评论

Pavement Crack detection Algorithm Based on Densely Connected and Deeply Supervised Network

引用

IEEE ACCESS 2021年 9卷 11835-11842页

作者： Li, Haifeng Zong, Jianping Nie, Jingjing Wu, Zhilong Han, Hongyang Civil Aviat Univ China Coll Comp Sci & Technol Tianjin 300300 Peoples R China Chengdu Tianfu Int Airport Chengdu 641419 Peoples R China

In order to improve the accuracy and robustness of existing automated crack detection methods, a fully convolutional neural network for pixel-level detection based on densely connected and deeply supervised network is proposed. First, the densely connected layers are applied for enhancing the propagation and reuse of crack features. Then, the deeply supervised modules are designed to make network extract more significant features through multi-scale levels. Finally, the feature maps from different scales are fused to achieve complementarity at different levels. In addition, a class-balanced cross-entropy loss function is designed to balance backgrounds and cracks by increasing the weight of crack pixel loss. The proposed method is tested on three public datasets, and the experiments show that our method is superior to state-of-the-art methods in accuracy, speed and robustness.

关键词： Feature extraction Convolution Semantics Image edge detection detection algorithms Training Surface cracks Crack detection deep learning densely connected network deeply supervised network

来源：评论

学校读者我要写书评

暂无评论

A Traffic-Sign detection Algorithm Based on Improved Sparse R-cnn

引用

IEEE ACCESS 2021年 9卷 122774-122788页

作者： Cao, Jinghao Zhang, Junju Jin, Xin Nanjing Univ Sci & Technol Sch Elect & Opt Engn Nanjing 210094 Jiangsu Peoples R China

Automatic traffic-sign detection is a hot topic in computer vision and one of the critical technologies of intelligent transportation. The Transformer structure has recently become a research hotspot due to its excellent performance. We hope to apply this structure to the design of traffic sign detection algorithms. Therefore, we make some improvements to Sparse R-cnn, a neural network model inspired by Transformer. Sparse R-cnn is a novel model, and its core idea is to replace hundreds of thousands of candidate anchors in the RPN network with a small set of proposal boxes. The experiments in our paper proved that the performance of the Sparse R-cnn model is better than other existing general object detection models. Based on the original Sparse R-cnn inspiration, an improved Sparse R-cnn model is proposed. First, a novel backbone for the task of traffic-sign detection is proposed. Multi-scale fusion structure is the essential method of improving the algorithm for small target detection, so improving the multi-scale capability of the backbone is a required method for designing traffic sign detection. So, we made further improvements to the existing backbone ResNest. We enhanced the multi-scale representation ability of the backbone by constructing hierarchical residual-like connections within each single radix block in the original ResNest. We call the improved backbone Res2Nest. The novel backbone proposed by us shows better performance without introducing excessive computational costs to the model. In addition, the attention mechanism is also an effective method to improve the detection of traffic signs, so we set up a branch network for recalibrating the channel feature response adaptively through the Global Average Pooling (GAP) operation and a fully connected layer. It can also be seen as the implementation of the cross-channel self-attention mechanism. After experiments by TT100K dataset, our method would attain a better accuracy and robustness.

关键词： Feature extraction Proposals Object detection Task analysis Heuristic algorithms Head detection algorithms Deep learning object detection traffic-sign detection self-attention mechanism improved Sparse R-cnn

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：