检索结果-内蒙古大学图书馆

Improved decentralized multi-sensor navigation system for airborne applications

GPS SOLUTIONS 2018年第3期22卷 1-13页

作者： Jiang, Wei Li, Yong Rizos, Chris Beijing Jiaotong Univ Sch Elect & Informat Engn Beijing Engn Res Ctr EMC & GNSS Technol Rail Tran State Key Lab Rail Traff Control & Safety Beijing Peoples R China Navexplorer Technol Sydney NSW Australia Univ New South Wales Sch Civil & Environm Engn Sydney NSW Australia

The integration of Global Navigation Satellite System (GNSS) and Inertial Navigation System (INS) technologies is a very useful navigation option for high-accuracy positioning in many applications. However, its performance is still limited by GNSS satellite availability and satellite geometry. To address such limitations, a non-GNSS-based positioning technology known as "Locata" is used to augment a standard GNSS/INS system. The conventional methods for multi-sensor integration can be classified as being either in the form of centralized Kalman filtering (CKF), or decentralized Kalman filtering. However, these two filtering architectures are not always ideal for real-world applications. To satisfy both accuracy and reliability requirements, these three integration algorithms-CKF, federated Kalman filtering (FKF) and an improved decentralized filtering, known as global optimal filtering (GOF)-are investigated. In principle, the GOF is derived from more information resources than the CKF and FKF algorithms. These three algorithms are implemented in a GPS/Locata/INS integrated navigation system and evaluated using data obtained from a flight test. The experimental results show that the position, velocity and attitude solution derived from the GOF-based system indicate improvements of 30, 18.4 and 20.8% over the CKF- and FKF-based systems, respectively.

关键词： GNSS Locata INS Multi-sensor navigation Data fusion

来源：评论

学校读者我要写书评

暂无评论

RGB and LiDAR fusion based 3d semantic segmentation for autonomous driving

arXiv

引用

arXiv 2019年

作者： Madawi, Khaled El Rashed, Hazem Sallab, Ahmad El Nasr, Omar Kamel, Hanan Yogamani, Senthil Valeo R & D Cairo Egypt Cairo University Valeo Visions Systems Ireland

LiDAR has become a standard sensor for autonomous driving applications as they provide highly precise 3D point clouds. LiDAR is also robust for low-light scenarios at night-time or due to shadows where the performance of cameras is degraded. LiDAR perception is gradually becoming mature for algorithms including object detection and SLAM. However, semantic segmentation algorithm remains to be relatively less explored. Motivated by the fact that semantic segmentation is a mature algorithm on image data, we explore sensor fusion based 3D segmentation. Our main contribution is to convert the RGB image to a polar-grid mapping representation used for LiDAR and design early and mid-level fusion architectures. Additionally, we design a hybrid fusion architecture that combines both fusion algorithms. We evaluate our algorithm on KITTI dataset which provides segmentation annotation for cars, pedestrians and cyclists. We evaluate two state-of-the-art architectures namely SqueezeSeg and PointSeg and improve the mIoU score by 10% in both cases relative to the LiDAR only baseline. Copyright © 2019, The Authors. All rights reserved.

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Deep learning for human affect recognition: Insights and new developments

arXiv

引用

arXiv 2019年

作者： Rouast, Philipp v. Adam, Marc T.P. Chiong, Raymond School of Electrical Engineering and Computing University of Newcastle CallaghanNSW2308 Australia

Automatic human affect recognition is a key step towards more natural human-computer interaction. Recent trends include recognition in the wild using a fusion of audiovisual and physiological sensors, a challenging setting for conventional machine learning algorithms. Since 2010, novel deep learning algorithms have been applied increasingly in this field. In this paper, we review the literature on human affect recognition between 2010 and 2017, with a special focus on approaches using deep neural networks. By classifying a total of 950 studies according to their usage of shallow or deep architectures, we are able to show a trend towards deep learning. Reviewing a subset of 233 studies that employ deep neural networks, we comprehensively quantify their applications in this field. We find that deep learning is used for learning of (i) spatial feature representations, (ii) temporal feature representations, and (iii) joint feature representations for multimodal sensor data. Exemplary state-of-the-art architectures illustrate the progress. Our findings show the role deep architectures will play in human affect recognition, and can serve as a reference point for researchers working on related applications. Copyright © 2019, The Authors. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

An effective data fusion-based routing algorithm with time synchronization support for vehicular wireless sensor networks

引用

JOURNAL OF SUPERCOMPUTING 2018年第3期74卷 1267-1282页

作者： Tan, Chang Ji, Sai Gui, Ziyuan Shen, Jian Fu, De-Sheng Wang, Jin Nanjing Univ Informat Sci & Technol Jiangsu Engn Ctr Network Monitoring Nanjing 210044 Jiangsu Peoples R China Yangzhou Univ Sch Informat Engn Yangzhou 225009 Jiangsu Peoples R China

In the field of vehicular wireless sensor networks-based structural health monitoring, the structural damage identification is achieved by two structural features, namely natural frequencies and mode shapes. The kind of data fusion-based routing algorithm in specific applications needs to meet time synchronization requirements and meet certain constraints, such as the single-hop communication between cluster head node and each node in cluster, the overlap between different clusters and so on. To meet the special constraints for data fusion-based routing algorithm in structural health monitoring, this paper proposed a new method based on an improved flooding time synchronization protocol, which is called time synchronization and enhanced greedy algorithm based on D(v) (TSDEGA) routing algorithm. The TSDEGA method can achieve the minimum connected cover by node's own degree D(v), and it can also meet the structural health monitoring routing constraints. The simulation experiments show that TSDEGA has better energy resistance and longer network lifetime, and it is superior to the traditional greedy algorithms. The proposed algorithm can effectively eliminate interference of outliers and improve the accuracy in order to meet time synchronization requirements in structural health monitoring applications.

关键词： Data fusion Routing algorithm Energy efficiency Time synchronization vehicular wireless sensor networks

来源：评论

学校读者我要写书评

暂无评论

An automatic parallelization ubiquitous machine learning accelerator on FPGA

引用

AIP Conference Proceedings 2022年第1期2519卷 030048页

作者： v. vijaya Baskar v. G. Sivakumar S. P. vimal M. vadivel. Sathyabama Institute of Science and Technology Chennai Tamil Nadu India VidyaJyothi Institute of Technology Hyderabad Telangana India Sri Ramakrishna Engineering College Coimbatore Tamil Nadu India Vidya Jyothi Institute of Technology Hyderabad Telangana India

Convolutionary neural networks have been commonly used for object recognition because, by emulating the nervous system’s action in living organisms, they can achieve high precision. Recently, development and deployment have further boosted the exponential growth of advanced businesses focused on intelligent systems. In specific, different fuel cells have been suggested for deep CNN based on the FPGA platform since it has characteristics of good efficiency, reconfigurability, and rapid growth, respectively. The processor architecture room has not been well explored, although existing FPGA prototypes have shown better performance over standard computers. One crucial concern is that computing efficiency does not fit the memory capacity offered by an FPGA platform. Due to its excellent precision, a Convolutionary computer program sees uses in a range of image processing applications ranging from target identification and tracking to scene comprehension. various algorithms exist for calculating CNNs. Mostly in CPU, three service systems are found: artificial neural, convolutionary artificial neural, genetic analysis, and collective processing. The multiplication uses an out-of-order optimization technique to facilitate simultaneous information flow processing to optimize coarse-grained instruction computing. Next, we suggest the fused structure in CNNs will naturally fuse several layers, repurposing the intermediary details. We explore diverse architectures to optimize the performance of a CNN centered on this working to create. To decide the fusion and algorithm technique for each sheet, we develop an optimum algorithm. We are also designing an integrated toolchain using Xilinx to ease the routing from the template to the FPGA bitstream. Studies using commonly used vGG and Alex Net illustrate that, relative to the previous fusion-based FPGA acceleration for CNNs, our architecture achieves up to 1.5X fine aggregate.

关键词：

来源：评论

学校读者我要写书评

暂无评论

RGB and LiDAR fusion based 3D Semantic Segmentation for Autonomous Driving

RGB and LiDAR fusion based 3D Semantic Segmentation for Auto...

引用

International Conference on Intelligent Transportation

作者： Khaled El Madawi Hazem Rashed Ahmad El Sallab Omar Nasr Hanan Kamel Senthil Yogamani Valeo R&D Cairo Egypt Cairo University Valeo Visions Systems Ireland

ISBN: (纸本)9781538670255

关键词： Laser radar Three-dimensional displays Semantics Image segmentation Feature extraction Computer architecture Cameras

来源：评论

学校读者我要写书评

暂无评论

实景三维环境中多模态数据融合的关键技术与挑战

引用

工程建设(维泽科技) 2024年第12期7卷 112-114页

作者：孙伟超大连市勘察测绘研究院集团有限公司辽宁大连116021

实景三维环境中的多模态数据融合技术在城市规划、智能交通、环境监测和虚拟现实等领域具有广泛应用。伴随着传感器技术的突破和计算能力的提升,多样化的数据模态(例如激光雷达数据、RGB图像、热成像等)的采集变得更加便捷。在进行数据... 详细信息

实景三维环境中的多模态数据融合技术在城市规划、智能交通、环境监测和虚拟现实等领域具有广泛应用。伴随着传感器技术的突破和计算能力的提升,多样化的数据模态(例如激光雷达数据、RGB图像、热成像等)的采集变得更加便捷。在进行数据融合时,遭遇众多技术难题,如数据结构的多样性、时间同步的难题、噪声与不确定性的处理,以及融合算法在效率与精确度上的挑战。文中旨在审视现实三维场景下,多模态数据融合技术的核心与难题,评述目前学术界的成果,并针对问题提出解决策略。研究显示,借助高效的数据预处理、前沿的融合算法和机器学习技术,能够增进数据融合的精确度和实用性,这为相关应用提供了更加充实的决策支持。

关键词：实景三维环境多模态数据融合传感器技术数据处理智能交通

来源：评论

学校读者我要写书评

暂无评论

Data-driven Representation Learning in Multimodal Feature fusion

Data-driven Representation Learning in Multimodal Feature Fu...

引用

作者： Song, Huan Arizona State University

学位级别：Ph.D.

Modern machine learning systems leverage data and features from multiple modalities to gain more predictive power. In most scenarios, the modalities are vastly different and the acquired data are heterogeneous in nature. Consequently, building highly effective fusion algorithms is at the core to achieve improved model robustness and inferencing performance. This dissertation focuses on the representation learning approaches as the fusion strategy. Specifically, the objective is to learn the shared latent representation which jointly exploit the structural information encoded in all modalities, such that a straightforward learning model can be adopted to obtain the prediction. We first consider sensor fusion, a typical multimodal fusion problem critical to building a pervasive computing platform. A systematic fusion technique is described to support both multiple sensors and descriptors for activity recognition. Targeted to learn the optimal combination of kernels, Multiple Kernel Learning (MKL) algorithms have been successfully applied to numerous fusion problems in computer vision etc. Utilizing the MKL formulation, next we describe an auto-context algorithm for learning image context via the fusion with low-level descriptors. Furthermore, a principled fusion algorithm using deep learning to optimize kernel machines is developed. By bridging deep architectures with kernel optimization, this approach leverages the benefits of both paradigms and is applied to a wide variety of fusion problems. In many real-world applications, the modalities exhibit highly specific data structures, such as time sequences and graphs, and consequently, special design of the learning architecture is needed. In order to improve the temporal modeling for multivariate sequences, we developed two architectures centered around attention models. A novel clinical time series analysis model is proposed for several critical problems in healthcare. Another model coupled with triplet ranking loss a

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-sensor data fusion using NIHS transform and decomposition algorithms

引用

MULTIMEDIA TOOLS AND applications 2018年第23期77卷 30381-30402页

作者： Ankarao, v. Sowmya, v. Soman, K. P. Amrita Vishwa Vidyapeetham Amrita Sch Engn Ctr Computat Engn & Networking CEN Coimbatore Tamil Nadu India

Multi-spectral image fusion is to enhance the details present in multi-spectral bands with the spatial information available in the panchromatic image. Fused images have the effect of spectral distortions and lack of structural similarity. To overcome these limitations, three methods are proposed using intensity, hue, saturation (IHS) and nonlinear IHS (NIHS) transform along with the Dynamic Mode Decomposition (DMD) and 2D-Empirical Mode Decomposition (2D-EMD or IEMD). An intensity plane is calculated from the NIHS transform. The modes are constructed using DMD by considering the variations between the intensity plane computed using NIHS transforms of a low resolution multi-spectral image and a panchromatic image. Similarly, 2D-EMD is also used for image fusion. Modes are subjected to weighted fusion rule to get an intensity plane with spatial and edge information. Finally, the calculated intensity plane is concatenated along with the hue and saturation plane of low-resolution multi-spectral image and transformed into RGB color space. Thus, the fused images have high spatial and edge information on spectral bands. The experiments and its quality assessment assure that proposed methods perform better than the existing methods.

关键词： Remote sensing Multi-spectral fusion Dynamic mode decomposition Image empirical mode decomposition Non-linearIHS transform

来源：评论

学校读者我要写书评

暂无评论

Kinect 3D Reconstruction fusion Mapping with C# Algorithm by RGB-D sensor

Kinect 3D Reconstruction Fusion Mapping with C# Algorithm by...

引用

International Conference on Advances in Science, Engineering and Robotics Technology

作者： Ashikur Rahman Shoumik Rahman Mehedy Rashedur M. Rahman Dept. of Computer Science and Engineering Brac University Dept. of Electrical and Computer Engineering North South University

ISBN: (纸本)9781728134468

The fundamental objective of this paper is to describe the applications of Microsoft Kinect sensor with robust algorithm and solution to present 3-D surface reconstruction mapping with different measurements. In the line of sensor operations, there could be two possible categories into 3-D reconstructions such as exorbitant laser scanner and other is by taking photos from various angles. In this paper, we have used indoor environmental scenes by depth sensor camera Microsoft Kinect. Data Pre-operations, sensor posture evaluations, computational fusion of the depth sensor data and 3-D surface extractions are the four major part of this experiment. The algorithm of this paper has been built uniquely with C# programming.

关键词： Microsoft kinect sensor 3D reconstruction Depth-sensor camera Image processing algorithms three-dimensional displays algorithms Image processing game player sensors laser scanners three-dimensional surface paper Microsoft

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：