检索结果-内蒙古大学图书馆

Late sensor fusion approach with a designed multi-segmentation network

Neural Computing and Applications 2024年第27期36卷 17125-17143页

作者： Çaldıran, Bekir Eren Acarman, Tankut Acartec Intelligent Vehicle Technologies Sanayi Mah. Teknopark Bulvarı 1/4C Istanbul Turkey Computer Engineering Department Galatasaray University Çırağan Cad. Ortaköy Istanbul Turkey

sensors have different perceptive abilities against environment. sensor fusion plays a crucial role at achieving better perception by accumulating the information acquired at different times. But, the decision of the observation may conflict with each other due to usage of different algorithms, thresholds on processing algorithms, and different perceptive character of sensors. This study presents the late fusion method applied to outputs provided by deep learning models fed by camera and lidar sensor’s measurement data. For camera sensor, a deep learning model as a multi-task network is proposed to multi-classify cars, motorcycles, bicycles, buses, trucks, and pedestrians under the category of dynamic traffic objects. In addition, color classified traffic lights and traffic signs with a capability of segmenting drivable area and detecting lane lines are classified under the category of static traffic objects. The proposed multi-network is trained and tested with the BDD100K dataset and benchmarked with publicly available multi-networks. The presented method is the second fastest multi-network reaching at 52 FPS runtime, ranked second based on the drivable area segmentation and lane line detection performance. For segmentation of dynamic objects, the network performance is increased by 22.45%, and considering mIoU overall performance increase is 3.96%. For a lidar sensor, a different modality is presented to detect objects. Two sensors’ data are fused by proposed fusion algorithm, and results are tested and evaluated with the KITTI dataset. The proposed fusion methodology outperforms the stand-alone lidar methods about 3.58% and 3.63% on BEV and 3D detection MAP, respectively. Overall, benchmarking with two distinct fusion approaches illustrates the effectiveness of the proposed method. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Optical radar

来源：评论

学校读者我要写书评

暂无评论

Leveraging Residual Deep Neural networks and Multi-Device sensors for Heterogeneous Activity Recognition 47

Leveraging Residual Deep Neural Networks and Multi-Device Se...

引用

47th International conference on Telecommunications and Signal processing, TSP 2024

作者： Mekruksavanich, Sakorn Phaphan, Wikanda Jitpattanakul, Anuchit School of Information and Communication Technology University of Phayao Department of Computer Engineering Phayao Thailand Research Group in Statistical Learning and Inference King Mongkut's University of Technology North Bangkok Department of Applied Statistics Faculty of Applied Science Bangkok Thailand Intelligent and Nonlinear Dynamic Innovations Research Center King Mongkut's University of Technology North Bangkok Department of Mathematics Faculty of Applied Science Bangkok Thailand

ISBN: (纸本)9798350365597

This study introduces a novel approach to identifying human activities using wearable sensors, particularly smart-phones and smartwatches. By leveraging deep learning neural networks and data from the HHAR dataset, which includes accelerometer and gyroscope data from individuals engaged in various activities, our method, centered around the HAR-Res NeXt model, accurately detects six activities. Utilizing residual connections and multi-kernel blocks, our approach effectively captures temporal and spatial relationships in sensor data. Experimental results demonstrate superior performance to standard machine learning algorithms and other deep learning approaches for human activity recognition. HAR-ResNeXt achieves high accuracy rates, particularly in classifying smartphone sensor data, underscoring its adaptability across diverse scenarios. Comparative analysis reveals the effectiveness of smartphone sensors and emphasizes the importance of multi-modal sensor fusion for accurate activity detection. © 2024 IEEE.

关键词： Smartphones

来源：评论

学校读者我要写书评

暂无评论

Synchronous Wireless sensor and Sink Placement Method Using Dual-Population Co-evolutionary Constrained Multiobjective Optimization Algorithm

引用

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 2023年第6期19卷 7561-7571页

作者： Yu, Qianqian Yang, Chen Dai, Guangming Peng, Lei Chen, Xiaoyu China Univ Geosci Sch Comp Sci Wuhan 430074 Peoples R China Beijing Inst Technol Sch Automat Beijing 100081 Peoples R China

Optimal wireless sensor placement (OWSP) plays a pivotal role in structural health monitoring. This study proposes a method to determine the simultaneous placement of sensors and sinks that minimizes energy consumption and maximizes information effectiveness. Network connectivity and reliability are critical constraints that determine the lifetime of wireless sensor networks. In this study, OWSP was formulated as a constrained multi-objective optimization problem with mixed-integer programming. Accordingly, a dual-population constrained multiobjective optimization (DCCMO) algorithm, which includes new crossover and mutation operators, was developed. In DCCMO, weak cooperation between two offspring populations is exploited to improve the efficiency of the solution search. The performance of DCCMO was compared to that of five other state-of-the-art algorithms using numerical examples with varying network parameters. DCCMO not only successfully matches the constrained Pareto front but also balances energy consumption and information effectiveness while exhibiting greater diversity and faster convergence than all other tested algorithms.

关键词： Constrained multiobjective optimization mixed-integer programming (MIP) optimal sensor and sink placement structural health monitoring wireless sensor networks (WSNs)

来源：评论

学校读者我要写书评

暂无评论

2CAP: A Novel Curve Crash Avoidance Protocol to Handle Curve Crashes in Vehicular Ad-Hoc Network

引用

IEEE ACCESS 2024年 12卷 60601-60619页

作者： Bashir, Raja Rizwan Saeed, Yousaf Ali, Abid Algarni, Abeer D. Muthanna, Ammar Hijjawi, Mohammad Alsboui, Tariq Univ Haripur Dept Informat Technol Haripur 22620 Pakistan KTS Haripur Govt Akhtar Nawaz Khan Shaheed Degree Coll Dept Comp Sci Haripur 22620 Pakistan Univ Engn & Technol Dept Comp Sci Taxila 48080 Pakistan Princess Nourah Bint Abdul Rahman Univ Coll Comp & Informat Sci Dept Informat Technol POB 84428 Riyadh 11671 Saudi Arabia Bonch Bruevich St Petersburg State Univ Telecommun Dept Telecommun Networks & Data Transmiss St Petersburg 193232 Russia Appl Sci Private Univ Dept Comp Sci Amman 11921 Jordan Univ Huddersfield Dept Comp Sci Huddersfield HD1 3DH England

Curves are the leading cause of road departure crashes that turn into deaths and severe injuries. Most crashes occur on curved roads because they have different geometrical circumstances from straight roads. A sharp turn on the road creates more difficulty for the driver than a normal curve road. To Avoid crashes on the curved road, there is a prerequisite to identify the curved road for drivers that this turn is problematic. Responding intelligently to save precious human life and the country's gross domestic product is necessary. Our paper proposed a Curve Crash Avoidance Protocol (2CAP) in Vehicular Ad-hoc Network (VANET). The intelligent Curve Crash Avoidance Algorithm is proposed to avoid disastrous vehicle crashes. The sensing Node Operation algorithm works for multiple sensors embedded in the Onboard Unit to gather information on vehicles and road environments. On the other side, road-side unit requests for this collected data via a secure communication channel for processing. A Linear Regression machine learning technique implements the intelligent Unit Operations algorithm. The intelligent Unit decides to notify the onboard Unit based on gathered data and trained dataset. We implement the scheme using the Linear Regression Machine Learning Model. Multiple sensors, Global Positioning Systems, and Global information systems use the dataset methodology to classify and predict results. The proposed model is expected to be effective for proper coordination with wireless sensor network equipment.

关键词： WHO VANET Curve Crash Avoidance LR MSE OBU RSU

来源：评论

学校读者我要写书评

暂无评论

Improved DOA Estimation of Sound Source of Small Amplitudes Using a Single Acoustic Vector sensor 19th

Improved DOA Estimation of Sound Source of Small Amplitudes ...

引用

19th National conference on Man-Machine Speech Communication

作者： Zhu, Bing Zhang, Wen Northwestern Polytech Univ Sch Marine Sci & Technol Ctr Intelligent Acoust & Immers Commun Xian Peoples R China

ISBN: (纸本)9789819610440;9789819610457

The localization of the sound source is crucial in audio signal processing, especially when using acoustic vector sensors. Traditional methods typically utilize active intensity vector information to estimate the direction of arrival (DOA). However, these methods often fail to estimate the DOA of small-amplitude sound sources in the presence of high interference. This study introduces an innovative approach that uses the reactive intensity vector as a clue to enhance the estimation of DOA for sound sources. The simulation results confirm that our proposed method consistently outperforms traditional techniques in estimating DOA for small-amplitude sound sources.

关键词： Active intensity vector Reactive intensity vector DOA estimation

来源：评论

学校读者我要写书评

暂无评论

Multiple intelligent Reflecting Surfaces Collaborative Wireless Localization System

引用

IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS 2025年第1期24卷 134-148页

作者： Zhang, Ziheng Chen, Wen Wu, Qingqing Li, Zhendong Zhu, Xusheng Chen, Jingfeng Cheng, Nan Shanghai Jiao Tong Univ Dept Elect Engn Shanghai 200240 Peoples R China Xi An Jiao Tong Univ Sch Informat & Commun Engn Xian 710049 Peoples R China Xidian Univ Sch Telecommun Engn Xian 710071 Peoples R China

This paper studies a multiple intelligent reflecting surfaces (IRSs) collaborative localization system where multiple semi-passive IRSs are deployed in the network to locate one or more targets based on time-of-arrival. It is assumed that each semi-passive IRS is equipped with reflective elements and sensors, which are used to establish the line-of-sight links from the base station (BS) to multiple targets and process echo signals, respectively. Based on the above model, we derive the Fisher information matrix of the echo signal with respect to the time delay. By employing the chain rule and exploiting the geometric relationship between time delay and position, the Cram & eacute;r-Rao bound (CRB) for estimating the target's Cartesian coordinate position is derived. Then, we propose a two-stage algorithmic framework to minimize CRB in single- and multi-target localization systems by joint optimizing active beamforming at BS, passive beamforming at multiple IRSs and IRS selection. For the single-target case, we derive the optimal closed-form solution for multiple IRSs coefficients design and propose a low-complexity algorithm based on alternating direction method of multipliers to obtain the optimal solution for active beaming design. For the multi-target case, alternating optimization is used to transform the original problem into two subproblems where semi-definite relaxation and successive convex approximation are applied to tackle the quadraticity and indefiniteness in the CRB expression, respectively. Finally, numerical simulation results validate the effectiveness of the proposed algorithm for multiple IRSs collaborative localization system compared to other benchmark schemes as well as the significant performance gains.

关键词： sensors Location awareness Collaboration Array signal processing Estimation Optimization Interference Wireless sensor networks Vectors Transforms intelligent reflecting surface collaborative localization time-of-arrival Cramer-Rao bound

来源：评论

学校读者我要写书评

暂无评论

intelligent Fruit Quality Monitoring System using sensor networks and Deep Learning

Intelligent Fruit Quality Monitoring System using Sensor Net...

引用

2024 International conference on Smart Electronics and Communication Systems, ISENSE 2024

作者： Kumar, Arvind Jaya Krishna, N. Chandra, Mahesh Kumar, Rampravesh Sahoo, Jaya Prakash Poojitha, N. Karan, Biswajit Kishore, Kamlesh Gitam University Department of EECE Bengaluru India School of ECE Reva University Bengaluru India Sandip University Department of EEE Nashik India University of Stellenbosch Department of EEE Stellenbosch South Africa RTC Institute of Technology Department of ECE Bengaluru India

ISBN: (纸本)9798331528126

This paper presents a novel approach for fruit quality detection using a multi-modal system that combines sensor data and image processing techniques with deep learning. In the first stage of our experiment, we employed various sensors to measure temperature, gas levels, and moisture content using DHT 11 and MQ-135 sensors. These sensors provide valuable insights into the environmental conditions surrounding the fruits, which are crucial indicators of their freshness. In the second stage, we utilized image processing algorithms to pre-process fruit images and train a SqueezeNet convolutional neural network (CNN) to classify fruits into two classes: fresh and rotten. By integrating sensor data with image-based classification, our proposed method offers a comprehensive solution for real-time fruit quality assessment. Experimental results demonstrate the effectiveness of our approach in accurately detecting fruit quality, paving the way for advanced applications in food safety and quality control. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Research on intelligent warning mechanism of accurate personnel positioning system based on fuzzy neural network

Research on intelligent warning mechanism of accurate person...

引用

2023 International conference on Computer Graphics, Artificial Intelligence, and Data processing, ICCAID 2023

作者： Tang, Lijun Chongqing Vocational Institute of Engineering Chongqing402260 China

ISBN: (纸本)9781510677685

In order to improve the intelligent level of accurate personnel positioning system to deal with underground emergencies, this paper shows an intelligent warning mechanism based on fuzzy neural network. The fuzzy neural network is a multi-input single-output model structure model, which has four layers, input layer, fuzzification layer, fuzzy inference layer and output layer. The intelligent warning mechanism standardizes the data of sensors in the location of the personnel label, and then uses the trained fuzzy neural network to predict. Before the disaster accident, the staff is automatically reminded to evacuate the dangerous area as soon as possible through the sound and light alarm function of the personnel label. The test shows the intelligent warning mechanism can remind the underground staff to evacuate the dangerous area as soon as possible before the disaster accident. Compared with the 70% prediction accuracy of the single sensor warning mechanism, the intelligent warning mechanism using fuzzy neural network can achieve 95% prediction accuracy. © 2024 SPIE.

关键词： Fuzzy neural networks

来源：评论

学校读者我要写书评

暂无评论

Poster Abstract: Representation Learning from Multimodal sensor Data with Maximally Correlated Autoencoders 21

Poster Abstract: Representation Learning from Multimodal Sen...

引用

21st ACM/IEEE International conference on information processing in sensor networks (IPSN)

作者： Ma, Fei Gu, Weixi Ni, Shiguang Zhang, Lin Tsinghua Univ Tsinghua Berkeley Shenzhen Inst Shenzhen Peoples R China China Acad Ind Internet Dept Intelligent Ind Res Beijing Peoples R China Tsinghua Univ Tsinghua Shenzhen Int Grad Sch Shenzhen Peoples R China

ISBN: (纸本)9781665496247

With the development of sensing technology, multiple sensors are widely used in Internet of Things (IoT) devices. A key challenge is to learn feature representations from multimodal sensor data to combine the information of different sensors. Although progress has been made by previous works, the correlation between different sensors is still not well exploited, which may limit the performance of representation learning. To address this problem, we propose a deep learning approach to learn representations from multimodal sensor data with maximally correlated autoencoders (MCA). It can efficiently capture high dependence between different modalities at different feature levels. The learned representations are further used for the recognition task. Experimental results on the real-world RGB-D dataset demonstrate the high effectiveness of MCA.

关键词： Representation learning Performance evaluation Deep learning Correlation Multimodal sensors information processing Semisupervised learning

来源：评论

学校读者我要写书评

暂无评论

Integrated intelligent Educational Environment - Opportunities for STEM Education 22

Integrated Intelligent Educational Environment - Opportuniti...

引用

22nd IFAC conference on Technology, Culture and International Stability (TECIS)

作者： Terzieva, Valentina Ilchev, Svetozar Djambazova, Edita Bulgarian Acad Sci Inst Informat & Commun Technol Bl 2 Acad G Bonchev St Sofia 1113 Bulgaria

The paper reviews the technology integration in STEM education within the scope of an Integrated intelligent Educational Environment (IIEE). A conceptual model of such an environment is suggested. The model consists of an intelligent Educational System that communicates through a gateway with sensors and actuators within the integrated environment. A possible usage of different technology devices within IIEE regarding STEM education is proposed. A common information model for both sensors and actuators is presented. A custom packet-based communication protocol is designed to support multiple wired and wireless communication media. IIEEs that include various technological devices provide an explorative and engaging context for STEM education. Toolkits for projects on engineering and technology can be prepared and play an active role in STEM classrooms. Copyright (C) 2024 The Authors.

关键词： intelligent educational environment STEM education sensor networks ICT Digitalization

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：