检索结果-内蒙古大学图书馆

Finding Stars From Fireworks: Improving Non-Cooperative Iris Tracking

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2022年第9期32卷 6137-6147页

作者： Lin, Chengdong Li, Xinlin Li, Zhenjiang Hou, Junhui City Univ Hong Kong Dept Comp Sci Hong Kong Peoples R China

We revisit the problem of iris tracking with RGB cameras, aiming to obtain iris contours from captured images of eyes. We find the reason that limits the performance of the state-of-the-art method in more general non-cooperative environments, which prohibits a wider adoption of this useful technique in practice. We believe that because the iris boundary could be inherently unclear and blocked, as its pixels occupy only an extremely limited percentage of those on the entire image of the eye, similar to the stars hidden in fireworks, we should not treat the boundary pixels as one class to conduct end-to-end recognition directly. Thus, we propose to learn features from iris and sclera regions first, and then leverage entropy to sketch the thin and sharp iris boundary pixels, where we can trace more precise parameterized iris contours. In this work, we also collect a new dataset by smartphone with 22 K images of eyes from video clips. We annotate a subset of 2 K images, so that label propagation can be applied to further enhance the system performance. Extensive experiments over both public and our own datasets show that our method outperforms the state-of-the-art method. The results also indicate that our method can improve the coarsely labeled data to enhance the iris contour's accuracy and support the downstream application better than the prior method.

关键词： Iris recognition Iris Cameras Training Noise measurement Neural networks Eyelids Iris tracking eye segmentation detection and localization

来源：评论

学校读者我要写书评

暂无评论

Neural Networks-Aided Insider Attack detection for the Average Consensus Algorithm

引用

IEEE ACCESS 2020年 8卷 51871-51883页

作者： Li, Gangqiang Wu, Sissi Xiaoxiao Zhang, Shengli Li, Qiang Shenzhen Univ Coll Elect & Informat Engn Shenzhen 518060 Peoples R China Peng Cheng Lab Shenzhen 518055 Peoples R China UESTC Sch Informat & Commun Engn Chengdu 611731 Peoples R China

To support the big-data processing needs of large-scale deployments of smart devices, there is significant interest in moving from cloud-computing to multi-agent (fog-computing) models, given these algorithms scalability and self-healing properties with respect to nodes and link failures. However, these algorithms are often based on the average consensus primitive, which is, unfortunately, vulnerable to data injection attacks. Recognizing this challenge, this work proposes three novel methods for detecting and localizing adversarial nodes using neural network (NN) models. The methods proposed are based on fully distributed algorithms, wherein each node locally updates its local states by exchanging information with its neighbors without supervision. Compared to the state-of-the-art, the proposed approach leverages the automatic learning characteristics of NN to reduce the dependence on pre-designed models and human expertise in complex internal attack scenarios. Simulation results show that the NN-based methods can significantly improve the attacker detection and localization performance.

关键词： Artificial neural networks Peer-to-peer computing Protocols Cloud computing Intrusion detection Simulation Gossip algorithm average consensus neural network (NN) insider attack detection and localization

来源：评论

学校读者我要写书评

暂无评论

A Review on the High-Efficiency detection and Precision Positioning Technology Application of Agricultural Robots

引用

PROCESSES 2024年第9期12卷 1833页

作者： Wang, Ruyi Chen, Linhong Huang, Zhike Zhang, Wei Wu, Shenglin Guangzhou Inst Sci & Technol Sch Intelligent Mfg & Elect Engn Guangzhou 510540 Peoples R China

The advancement of agricultural technology has increasingly positioned robotic detection and localization techniques at the forefront, ensuring critical support for agricultural development through their accuracy and reliability. This paper provides an in-depth analysis of various methods used in detection and localization, including UWB, deep learning, SLAM, and multi-sensor fusion. In the domain of detection, the application of deep algorithms in assessing crop maturity and pest analysis is discussed. For localization, the accuracy of different methods in target positioning is examined. Additionally, the integration of convolutional neural networks and multi-sensor fusion with deep algorithms in agriculture is reviewed. The current methodologies effectively mitigate environmental interference, significantly enhancing the accuracy and reliability of agricultural robots. This study offers directional insights into the development of robotic detection and localization in agriculture, clarifying the future trajectory of this field and promoting the advancement of related technologies.

关键词： detection and localization ultra-wideband (UWB) technology deep learning simultaneous localization and mapping (SLAM) multi-sensor fusion convolutional neural networks (CNN)

来源：评论

学校读者我要写书评

暂无评论

An ultrasonic approach to localization of fiducial markers for interactive, image-guided neurosurgery - Part I: Principles

引用

IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING 1998年第5期45卷 620-630页

作者： Lewis, JT Galloway, RL Schreiner, S Vanderbilt Univ Dept Biomed Engn Nashville TN 37235 USA Vanderbilt Univ Med Ctr Dept Neurol Surg Nashville TN 37232 USA Johns Hopkins Med Inst Dept Radiol & Radiol Sci Baltimore MD 21205 USA

Fiducial markers are reference points used in the registration of image space(s) with physical (patient) space. As applied to interactive, image-guided surgery, the registration of image space with physical space allows the current location of a surgical tool to be indicated on a computer display of patient-specific preoperative images. This intrasurgical guidance information is particularly valuable in surgery within the brain, where visual feedback is limited. The accuracy of the mapping between physical and image space depends upon the accuracy with which the fiducial markers were located in each coordinate system. To effect accurate space registration for interactive, image-guided neurosurgery, the use of permanent fiducial markers implanted into the surface of the skull is proposed in this paper. These small cylindrical markers are composed of materials that make them visible in the image sets. The challenge lies in locating the subcutaneous markers in physical space, This paper presents an ultrasonic technique for transcutaneously detecting the location of these markers. The technique incorporates an algorithm based on detection of characteristic properties of the reflected A-mode ultrasonic waveform. The results demonstrate that ultrasound is an appropriate technique for accurate transcutaneous marker localization. The companion paper to this article describes an automatic, enhanced implementation of the marker-localization theory described in this article.

关键词： A-mode ultrasound computer-assisted surgery detection and localization fiducial markers image-guided surgery registration stereotactic neurosurgery ultrasound

来源：评论

学校读者我要写书评

暂无评论

State Space System Identification Approach to Radar Data Processing

引用

IEEE TRANSACTIONS ON SIGNAL PROCESSING 2011年第8期59卷 3675-3684页

作者： Prasanth, R. K. BAE Syst Burlington MA 01803 USA

Space-time adaptive processing (STAP) algorithms typically consist of a data transformation step to reduce the number of degrees of freedom and a sampling step wherein radar returns from adjacent range bins are used to estimate interference statistics. The reduction in the number of degrees of freedom, inadequate sample support, presence of target in sampled data, and range dependence of interference are some of the main reasons for STAP performance loss. In this paper, we present an approach to target detection and localization that mitigates these performance losses using the well-known stochastic realization algorithm from system identification theory. We first identify a state space model from the radar return data in range-pulse domain for a given range bin, and then perform detection and localization using the identified state space matrices. As interference statistics are not directly computed and since there is no sampling from adjacent range bins, this approach is more robust to sample support issues, target in training and range dependence of clutter. A numerical comparison of the approach with beam-space post-Doppler STAP using simulated data is given.

关键词： Covariance detection and localization radar space-time adaptive processing (STAP) stochastic subspace system identification

来源：评论

学校读者我要写书评

暂无评论

Spatial-Temporal Data-Driven Model for Load Altering Attack detection in Smart Power Distribution Networks

引用

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 2024年第5期20卷 7414-7427页

作者： Ebtia, Afshin Rebbah, Dhiaa Elhak Debbabi, Mourad Kassouf, Marthe Ghafouri, Mohsen Mohammadi, Arash Soeanu, Andrei Concordia Univ Secur Res Ctr Gina Cody Sch Engn & Comp Sci Montreal PQ H3G 1M8 Canada Hydroquebec Res Inst IREQ Varennes PQ J3X 1S1 Canada

The widespread deployment of information and communication technologies in smart power distribution networks (SPDNs) exposes them to cyber threats. Among different types of cyber-attacks in such ICT-based SPDNs, load-altering attacks (LAAs) against high-wattage devices have received significant attention in recent years. In this context, this article proposes a data-driven detection model tailored for identifying and localizing LAAs in SPDNs. In this pursuit, first, the graph structure of an SPDN, which is obtained from the grid topology, and node features, i.e., measurements of the load's power, are fed to a graph attention network (GAT), and the spatial correlations among the nodes are captured. Alongside, the temporal correlations are captured using a long short-term memory model trained based on the graph representation obtained from the GAT. These spatial and temporal correlations are used by prediction and reconstruction models, i.e., a fully connected neural network and an auto-encoder. Finally, based on the error of the prediction and reconstruction blocks, an attack score for each load is calculated, and the compromised loads are detected and localized. To evaluate the performance of the proposed model, a co-simulation framework, which simulates the power system and emulates the communication network using real industrial protocols, i.e., IEC 60870-5-104, has been developed. The robustness of the model's performance against noisy data and non-attack outliers is confirmed with respect to different noise levels and data outliers. Also, the developed model's superior performance over existing models is demonstrated through various LAA scenarios applied to the IEEE 33- and the 123-Bus benchmarks.

关键词： Load modeling Correlation Predictive models Power systems Data models Power measurement Computational modeling Cosimulation framework detection and localization load-altering attack (LAA) prediction and reconstruction models smart power distribution network (SPDN) spatial and temporal correlation

来源：评论

学校读者我要写书评

暂无评论

Adaptive split spectrum processing for ultrasonic signal in the pulse echo test

引用

RUSSIAN JOURNAL OF NONDESTRUCTIVE TESTING 2015年第4期51卷 245-257页

作者： Bouden, T. Djerfi, F. Nibouche, M. Jijel Univ Dept Automat LEND Lab Jijel Algeria UWE Univ Bristol Dept Engn Design & Math Bristol Avon England

In this paper, an Adaptive Split Spectrum Processing technique (A-SSP) is proposed, to improve ultrasonic echoes detection. It is an arrangement of conventional Split Spectrum Processing (SSP) with an empirical method of analyzing nonlinear and non-stationary signals, called Empirical Mode Decomposition (EMD). This proposed technique allows breaking up the signal into several bands of frequencies in an adaptive way and intrinsic to the treated signal using EMD. It enables us to know the internal contents and the local changes of the ultrasonic signal and makes the detection of any desired targets more flexible for the coherent noise problem. In the combination phase of A-SSP, a linear operation for selected intrinsic mode functions and a non linear one for non selected intrinsic mode functions are used to reconstruct the signal with separated echoes. To evaluate the proposed techniques (A-SSP with different combination operations), firstly a mortar specimen with artificial defect is used to resolve the defects detection and localization problem. Secondly a paste cement specimen is also used to resolve the materials characterization problem. The signals were obtained using a technique applied in pulse-echo mode, known as the prism technique. Numerical and experimental tests were performed to verify the effectiveness and reliability of the proposed technique and to show its excellent performances.

关键词： Non-Destructive Testing (NDT) paste cement mortar propagation velocity thickness detection and localization Split Spectrum Processing (SSP) Empirical Mode Decomposition (EMD)

来源：评论

学校读者我要写书评

暂无评论

localization and classification of power quality disturbances using maximal overlap discrete wavelet transform and data mining based classifiers

引用

IFAC-PapersOnLine 2016年第1期49卷 437-442页

作者： Upadhyaya, Swarnabala Mohanty, Sanjeeb Department of Electrical Engineering National Institue of Technology Rourkela769008 India

This paper has proposed the time-series based maximal overlap discrete wavelet transform (MODWT) technique for detection and localization of different types of power quality (PQ) disturbances. Ten types of different PQ events of the voltage signal such as sag, swell, interruption, harmonic, spike, notch etc. are analyzed with the aforementioned wavelet transform (WT). Each of the signal is decomposed up to fourth level with the MODWT. The co-efficients of MODWT decomposition are further used for feature extraction which are the input to the classifiers like Support Vector Machine (SVM) and Decision Tree(DT). For the detection of the disturbances, the signals are decomposed up to four finer levels whereas for the classification, decomposition is carried out up to seventh finer levels. © 2016, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.

关键词： Data mining Decision trees Discrete wavelet transforms Power quality Signal reconstruction Support vector machines Wavelet transforms detection and localization Maximal overlap discrete wavelet transforms Power quality disturbances PQ event Types of power Voltage signals

来源：评论

学校读者我要写书评

暂无评论

Sensor fault diagnosis in inland navigation networks based on a grey-box model

引用

IFAC-PapersOnLine 2018年第24期51卷 742-747页

作者： Segovia, P. Blesa, J. Duviella, E. Rajaoarisoa, L. Nejjari, F. Puig, V. Terrassa Campus Gaia Building Rambla Sant Nebridi 22 Terrassa08222 Spain c/ Llorens i Artigas 4-6 Barcelona08028 Spain IMT Lille Douai Univ. Lille Unité de Recherche Informatique Automatique LilleF-59000 France

Inland navigation networks are equipped with limnimeters to measure and record water level data for the control of water levels and the management of water resources. When faults occur on sensors, corrupted data can be considered as correct, leading to undesirable management actions. Therefore, it is necessary to detect and localize these faults. In this paper, the detection and localization of sensor faults is performed through the analysis of the parameters of a grey-box model, which are obtained from available real data. The parameters are determined with a sliding window, with the exception of the delays, which are considered known a priori. A fault is detected and then localized when there is a change in the value of the parameters. This approach is well suited for constant faults and particularly well adapted for intermittent faults. Data of an inland navigation reach located in the north of France are used to highlight the performance of the proposed approach. © 2018

关键词： Fault detection Air navigation Electric fault currents Failure analysis Information management Inland waterways Large scale systems Navigation Water levels Water management Corrupted data detection and localization Grey box modeling Inland navigation Intermittent fault Sensor fault diagnosis Sliding Window Water level data

来源：评论

学校读者我要写书评

暂无评论

A Holistic Vision-based Hazard detection Framework for Asteroid Landings

引用

IFAC-PapersOnLine 2016年第17期49卷 218-223页

作者： Yan, Yunfeng Qi, Donglian Li, Chaoyong Yu, Miao Chen, Tuo College of Electrical Engineering Zhejiang University Hangzhou China

A holistic vision-based hazard detection framework for asteroid landing is introduced in this paper. The proposed holistic conjecture takes advantage of the Hough Transform method and decision forests, shares the simplicity and wide applicability of the Hough transform but bypasses the problem of obstacle multi-size detection and permits detection of multiple objects. The framework is conducive to hazard avoidance, satisfying the safe bound of the lander and the exploration task, and adaptive to environment, autonomous without additive complex filtering. The results show that the framework provides detection and localization performance close to that of human calibrations and can be further improved when larger and more diverse datasets are available. © 2016

关键词： Landing Asteroids Forestry Hazards Hough transforms Object detection Complex filtering Decision forest detection and localization Exploration tasks Hazard areas Holistic visions Safe landing Transform methods

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：