检索结果-内蒙古大学图书馆

6th IEEE International Conference on Electronic Information and Communication Technology, ICEICT 2023

作者： Yuan, Wang Chen, Xiaolong Du, Xiaolin Yu, Gang He, Xiaoyang School of Computer and Control Engineering Yantai Unversity Yantai China Naval Aviation University Marine Target Detection Reaserch Group Yantai China School of Electrical Engineering Universiy of Jinan Jinan China

ISBN: (纸本)9798350399059

Accurately estimating the micro-motion parameters of unmanned aerial vehicles (UAVs) is of significant importance for UAV recognition and classification. Traditional time-frequency analysis methods have low resolution and suffer from severe cross-term interference, making it challenging to precisely estimate the micro-motion parameters of UAVs. In order to slove this problem, this paper proposes a high-resolution local maximum synchroextracting transform method(LSET). Firstly, echo data is collected using FMCW, and then the data is subjected to STFT. Next, an estimation of the instantaneous frequency is calculated, and finally, the value of the maximum energy in the time-frequency representation is extracted to retain useful information. The experimental results from the actual data testing show that compare with the traditional time-frequency analysis method, the proposed method reduces the average Renyi entropy by 4.7 dB. A smaller entropy value indicates higher time-frequency resolution and energy concentration. This method improves the average length estimation accuracy of rotating blades by 33.3%. © 2023 IEEE.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

ORB-Net: End-to-end Planning Using Feature-based Imitation Learning for Autonomous Drone Racing 56

ORB-Net: End-to-end Planning Using Feature-based Imitation L...

引用

56th International Symposium on Robotics, ISR Europe 2023

作者： Pham, Huy Xuan Heiß, Micha Tran, Dung Nguyen, Minh Anh Nguyen, Anh Quang Kayacan, Erdal Department of Electrical and Computer Engineering Aarhus University Aarhus C8000 Denmark Hanoi Viet Nam Automatic Control Group Department of Electrical Engineering and Information Technology Paderborn University Paderborn Germany

ISBN: (纸本)9783800761418

Deep learning-based methods have enhanced the performance of many robot applications thanks to their superior ability to robustly extract rich high-dimensional features. However, it comes with a high computational cost that often increases the latency of the overall system. On the other hand, traditional feature extraction methods, such as oriented FAST and rotated BRIEF (ORB), can be computed efficiently and remains the backbone of some critical robot algorithms, e.g. ORB-SLAM for robot state estimation. In this work, the usefulness of the aforementioned features for a robot navigation task is investigated. The features are experimentally incorporated with a deep-learning method, called ORB-Net1, which allows an agile aerial robot to learn a motion policy to complete a racing track in an autonomous drone racing context. The experimental studies demonstrate that it can be beneficial to reuse these computed features for end-to-end motion planning for the agile quadrotor, as the proposed method using combined input of ORB feature position and RGB images outperforms the baseline methods which use only RGB images. © VDE VERLAG GMBH Berlin Offenbach.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Optimizing Indoor Network Element Layout for Enhanced Signal Coverage and Security in Location-Based Services 7th

Optimizing Indoor Network Element Layout for Enhanced Signa...

引用

7th International Conference on Attacks and Defenses for Internet-of-Things, ADIoT 2024

作者： Yu, Xiaomin Yu, Xiaokun College of Computer and Control Engineering Qiqihar University Qiqihar China Heilongjiang Key Laboratory of Big Data Network Security Detection and Analysis Qiqihar University Qiqihar China Heilongjiang Jiaotong Polytechnic Qiqihar China

ISBN: (纸本)9783031855924

Recently, with the rapid development of communication technology, location-based services (LBS) in indoor environments have penetrated into various aspects of people’s daily lives. Wireless local area networks (WLAN) have become the preferred choice for location-based services due to their broad deployment and low cost. However, the typical non-line-of-sight characteristics of indoor environments result in severe fluctuations in received signal strength (RSS), signal coverage vulnerabilities can exacerbate environmental interference, making positioning results more vulnerable to attacks or tampering. The layout of indoor network elements is a key factor that constrains signal coverage rate. This paper focuses on the optimization of indoor network element layout, proposes an indoor network element layout optimization model based on signal coverage and quality in indoor environments and introduces an Adaptive Simulation Annealing Non-inertial Opposite Particle Swarm Optimization algorithm (ASA-NRPSO). Firstly, a reverse strategy is introduced into the particle flight process of the Particle Swarm Optimization (PSO) algorithm, and high-quality particles are selected for the next generation iteration. In the selection of the reverse strategy, the simulated annealing idea is integrated to adaptively select the reverse strategy based on the needs of the particles at different stages, which avoids the particles falling into local optima. The inertia term in the particle swarm velocity update formula is replaced with group information for non-inertial updates. This can more fully utilize the global information of the population to guide the movement of the next generation of particles and improve the convergence speed of the particle swarm. The elite mutation strategy is used to increase the diversity of particles and improve the global search capability of the algorithm. This method ensures the rapid generation of network element layout to improve the signal coverage rate

关键词： Location based services

来源：评论

学校读者我要写书评

暂无评论

Hgo-Yolo: Advancing Anomaly Behavior Detection with Hierarchical Features and Lightweight Optimized Detection

SSRN

引用

SSRN 2024年

作者： Dong, Guanghui Zheng, Qizhi Luo, Zhongze Guo, Meiyan Wang, Xinzhu Wu, Renqimuge Meng, Qiu College of Computer and Control Engineering Northeast Forestry University Harbin150040 China

Accurate and real-time object detection is crucial for anomaly behavior detection, especially in scenarios constrained by hardware limitations, where achieving a balance between accuracy and speed is imperative to improve detection performance. In this study, we propose a model called HGO-YOLO, which integrates the HGNetv2 network structure into YOLOv8. This architectural combination expands the receptive field and captures a wider range of features while also simplifying the model complexity by replacing some convolutions with GhostConv. We introduce a new lightweight detection head, OptiConvDetect, which utilizes parameter sharing concepts to construct the detection head more efficiently. We evaluate the performance of the HGO-YOLO model on a collected dataset. Extensive experimental results show that the proposed algorithm achieves a mAP@0.5 of 87.4% and a recall rate of 81.1% with a remarkably compact model size of only 4.6 MB. When executed on the CPU, the algorithm achieves a frame rate of 56 FPS. Compared to the original YOLOv8n algorithm, our approach not only improves accuracy by 3.0% but also reduces the computational load by 51.69% (from 8.9 GFLOPs to 4.3 GFLOPs), while increasing the frame rate by a factor of 1.7. These results underline the excellent overall performance of the HGO-YOLO model in detecting anomalous behavior. © 2024, The Authors. All rights reserved.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Robotic Arm Dataset (RoAD): A Dataset to Support the Design and Test of Machine Learning-Driven Anomaly Detection in a Production Line 49

Robotic Arm Dataset (RoAD): A Dataset to Support the Design ...

引用

49th Annual Conference of the IEEE Industrial Electronics Society, IECON 2023

作者： Mascolini, Alessio Gaiardelli, Sebastiano Ponzio, Francesco Dall'ora, Nicola MacIi, Enrico Vinco, Sara Di Cataldo, Santa Fummi, Franco Polytechnic of Turin Dept. of Control and Computer Engineering Italy University of Verona Dept. of Engineering for Innovation Medicine Italy Polytechnic of Turin Dept. of Regional and Urban Studies and Planning Italy

ISBN: (纸本)9798350331820

The early detection of anomalous behaviors from a production line is a fundamental aspect of Industry 4.0, facilitated by the collection of massive amounts of data enabled by the Industrial Internet of Things. Nonetheless, the design and validation of anomaly detection algorithms, mostly based on sophisticated Machine Learning models, heavily rely on the availability of annotated datasets of realistic anomalies, which is very difficult to obtain in a real production line. To address this problem, we introduce the Robotic Arm Dataset (RoAD), specifically designed to support the development and validation of Multivariate Time Series Anomaly Detection (MTSAD) algorithms. We collect and annotate a large number of data and metadata to characterize the motion and energy consumption of a collaborative robotic arm in a full-fledged production line and annotate a comprehensive set of healthy as well as realistic anomalies scenarios. To prove the significance of RoAD and encourage future developments, we benchmark several state-of-the-art anomaly detection algorithms on our newly introduced dataset, and we freely release it to the scientific community. © 2023 IEEE.

关键词： Process monitoring

来源：评论

学校读者我要写书评

暂无评论

Evaluating the Efficacy of Cut-and-Paste Data Augmentation in Semantic Segmentation for Satellite Imagery

Evaluating the Efficacy of Cut-and-Paste Data Augmentation i...

引用

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Ionut M. Motoi Leonardo Saraceni Daniele Nardi Thomas A. Ciarfuglia Department of Computer Control and Management Engineering Sapienza University of Rome Rome Italy

ISBN: (数字)9798350360325

ISBN: (纸本)9798350360332

Satellite imagery is crucial for tasks like environmental monitoring and urban planning. Typically, it relies on semantic segmentation or Land Use Land Cover (LULC) classification to categorize each pixel. Despite the advancements brought about by Deep Neural Networks (DNNs), their performance in segmentation tasks is hindered by challenges such as limited availability of labeled data, class imbalance and the inherent variability and complexity of satellite images. In order to mitigate those issues, our study explores the effectiveness of a Cut-and-Paste augmentation technique for semantic segmentation in satellite images. We adapt this augmentation, which usually requires labeled instances, to the case of semantic segmentation. By leveraging the connected components in the semantic segmentation labels, we extract instances that are then randomly pasted during training. Using the DynamicEarthNet dataset and a U-Net model for evaluation, we found that this augmentation significantly enhances the mIoU score on the test set from 37.9 to 44.1. This finding highlights the potential of the Cut-and-Paste augmentation to improve the generalization capabilities of semantic segmentation models in satellite imagery.

关键词： Training Adaptation models Semantic segmentation Semantics Urban planning Land surface Satellite images

来源：评论

学校读者我要写书评

暂无评论

A Micro Architectural Events Aware Real-Time Embedded System Fault Injector

A Micro Architectural Events Aware Real-Time Embedded System...

引用

Latin American Test Workshop, LATW

作者： Enrico Magliano Alessio Carpegna Alessadro Savino Stefano Di Carlo Department of Control and Computer Engineering(DAUIN) Politecnico di Torino Turin Italy

ISBN: (数字)9798350365559

ISBN: (纸本)9798350365566

In contemporary times, the increasing complexity of the system poses significant challenges to the reliability, trustworthiness, and security of the Safety-Critical Real-Time Embedded Systems (SACRES). Key issues include the susceptibility to phenomena such as instantaneous voltage spikes, electromagnetic interference, neutron strikes, and out-of-range temperatures. These factors can induce switch state changes in transistors, resulting in bit-flipping, soft errors, and transient corruption of stored data in memory. The occurrence of soft errors, in turn, may lead to system faults that can propel the system into a hazardous state. Particularly in critical sectors like automotive, avionics, or aerospace, such malfunctions can have real-world implications, potentially causing harm to *** paper introduces a fault injector designed with the novelty to facilitate the monitoring, aggregation, and examination of micro-architectural events. This is achieved by harnessing the microprocessor’s Performance Monitoring Unit (PMU) and the debugging interface, explicitly focusing on ensuring the repeatability of fault injections. The fault injection methodology targets bit-flipping within the memory system, affecting CPU registers and RAM. The outcomes of these fault injections enable a thorough analysis of the impact of soft errors in the final output and timing predictability demanded by SACRES.

关键词： Embedded systems Random access memory Voltage Real-time systems Phasor measurement units Registers Timing

来源：评论

学校读者我要写书评

暂无评论

Maintenance of High Fuzzy Utility Itemsets Using the Pre-Large-Itemset Concept and Tree Structure 23

Maintenance of High Fuzzy Utility Itemsets Using the Pre-Lar...

引用

10th Multidisciplinary International Social Networks Conference, MISNC 2023

作者： Hong, Tzung-Pei Hung, Wei-Teng Huang, Wei-Ming Department of Computer Science and Information Engineering National University of Kaohsiung Taiwan Department of Computer Science and Engineering National Sun Yat-Sen University Library and Information Center National University of Kaohsiung Taiwan Department of Electrical and Control China Steel Inc. China

ISBN: (纸本)9798400708176

Utility mining has recently attracted much attention in real-world applications because it fits actual situations. Fuzzy utility mining approaches can discover important high-utility patterns in linguistic terms. Data, however, usually come intermittently, and users may want to know currently updated mining results instead of only the ones at the previous stage. Thus, maintaining correct knowledge is very important to this requirement. In the past, we used the fast-updated (FUP) approach and tree structures to manage the maintenance problem of fuzzy utility mining for incremental databases. In this work, we adopt the pre-large-itemset concept to speed up the tree-based maintenance of high fuzzy utility itemsets. The proposed method uses the header table and the pre-large threshold to reduce database scans for efficiency improvement. From the experimental results, the proposed approach performs better than the FUP-tree-based and batching method. © 2023 ACM.

关键词： Maintenance

来源：评论

学校读者我要写书评

暂无评论

Real-time and Energy-aware Scheduling for Edge-to-Cloud Continuum based on Reinforcement Learning

Real-time and Energy-aware Scheduling for Edge-to-Cloud Cont...

引用

International Workshop/Symposium on Distributed Simulation and Real-Time Applications (DS-RT)

作者： Andrea Panceri Gabriele Proietti Mattia Roberto Beraldi Department of Computer Control and Management Engineering "A. Ruberti" Sapienza University of Rome

ISBN: (数字)9798331527211

ISBN: (纸本)9798331527228

By spreading out computing workloads over multiple levels, the Edge-to-Cloud continuum paradigm improves the performance of applications that are sensitive to latency. However, real-time scheduling is difficult on the Edge computing layer since it consists of a variety of nodes with varying uptime. In this paper, we address this issue by proposing an online and adaptive scheduling algorithm based on a continuous learning Reinforcement Learning. Our algorithm determines each work request individually, optimizing scheduling policies to meet real-time application requirements while taking environmental energy and battery limits into account. We validate the efficacy of our approach in dynamically assigning tasks, particularly in scenarios where Edge nodes exhibit variable speeds and unpredictable failures, while efficiently managing energy resources and battery constraints through extensive simulations and comparisons with static scheduling strategies.

关键词： Processor scheduling Heuristic algorithms Computational modeling Reinforcement learning Quality of service Dynamic scheduling Real-time systems Batteries Testing Load modeling

来源：评论

学校读者我要写书评

暂无评论

Xception CNN-Ensemble Learning Based Facial Emotion Recognition 6

Xception CNN-Ensemble Learning Based Facial Emotion Recognit...

引用

6th International Conference on Computing, Communication, control and Automation, ICCUBEA 2022

作者： Shirsath, Supriya Vikhe, Vaishali Vikhe, Pratap Pravara Rural Engineering College Dept. of Computer Engineering Ahmednagar India Pravara Rural Engineering College Dept. of Instrumentation & Control Engineering Ahmednagar India

ISBN: (纸本)9781665484510

Emotions in human is an effective medium to study the mindset of a person. Since, expression on the face of human is significant approach to understand the condition and communicate with him to release the pressure or stress of person. The emotions and behavior have a strong relationship and is noteworthy in non-verbal form of communication. The advancement in medical science and use of image processing tools like Artificial Intelligence (AI) and Machine Learning (ML), can be helpful to recognize emotion, detect stress and depression level of a person. Thus, in this research work emotion recognition system is developed using Convolution Neural Networks (CNNs) with increased depth and width. CNNs have been shown to enhance prediction accuracy. In terms of proper emotion categorization and accurate prediction, the suggested ensemble technique is effective. In this paper Xception CNN architecture is proposed for accurate facial emotions prediction along with an ensemble model, using Max Voting ensemble technique, mainly contributes in accurate classification. In proposed technique trained CNN models were loaded and for each trained model prediction probabilities were generated. Furthermore, it is then used for Max Voting to generate final emotion prediction. The CNN models are trained and evaluated on FER-2013 dataset. © 2022 IEEE.

关键词： Emotion Recognition

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：